id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.19608
|
Zhiyuan Wang
|
Zhiyuan Wang and Bokui Chen
|
ChordPrompt: Orchestrating Cross-Modal Prompt Synergy for Multi-Domain Incremental Learning in CLIP
|
Accepted by the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD 2025)
| null | null | null |
cs.AI cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Continual learning (CL) empowers pre-trained vision-language models to adapt effectively to novel or previously underrepresented data distributions without comprehensive retraining, enhancing their adaptability and efficiency. While vision-language models like CLIP show great promise, they struggle to maintain performance across domains in incremental learning scenarios. Existing prompt learning methods face two main limitations: 1) they primarily focus on class-incremental learning scenarios, lacking specific strategies for multi-domain task incremental learning; 2) most current approaches employ single-modal prompts, neglecting the potential benefits of cross-modal information exchange. To address these challenges, we propose the \ChordPrompt framework, which facilitates a harmonious interplay between visual and textual prompts. \ChordPrompt introduces cross-modal prompts to leverage interactions between visual and textual information. Our approach also employs domain-adaptive text prompts to select appropriate prompts for continual adaptation across multiple domains. Comprehensive experiments on multi-domain incremental learning benchmarks demonstrate that \ChordPrompt outperforms state-of-the-art methods in zero-shot generalization and downstream task performance.
|
[
{
"version": "v1",
"created": "Tue, 24 Jun 2025 13:22:06 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 12:23:15 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wang",
"Zhiyuan",
""
],
[
"Chen",
"Bokui",
""
]
] |
2506.19992
|
G\'abor Petneh\'azi
|
Gabor Petnehazi and Bernadett Aradi
|
HERCULES: Hierarchical Embedding-based Recursive Clustering Using LLMs for Efficient Summarization
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The explosive growth of complex datasets across various modalities necessitates advanced analytical tools that not only group data effectively but also provide human-understandable insights into the discovered structures. We introduce HERCULES (Hierarchical Embedding-based Recursive Clustering Using LLMs for Efficient Summarization), a novel algorithm and Python package designed for hierarchical k-means clustering of diverse data types, including text, images, and numeric data (processed one modality per run). HERCULES constructs a cluster hierarchy by recursively applying k-means clustering, starting from individual data points at level 0. A key innovation is its deep integration of Large Language Models (LLMs) to generate semantically rich titles and descriptions for clusters at each level of the hierarchy, significantly enhancing interpretability. The algorithm supports two main representation modes: `direct' mode, which clusters based on original data embeddings or scaled numeric features, and `description' mode, which clusters based on embeddings derived from LLM-generated summaries. Users can provide a `topic\_seed' to guide LLM-generated summaries towards specific themes. An interactive visualization tool facilitates thorough analysis and understanding of the clustering results. We demonstrate HERCULES's capabilities and discuss its potential for extracting meaningful, hierarchical knowledge from complex datasets.
|
[
{
"version": "v1",
"created": "Tue, 24 Jun 2025 20:22:00 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 05:10:20 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Petnehazi",
"Gabor",
""
],
[
"Aradi",
"Bernadett",
""
]
] |
2506.20044
|
Roldao da Rocha
|
V. F. Guimar\~aes, R. T. Cavalcanti, R. da Rocha
|
Hair imprints of the gravitational decoupling and hairy black hole spectroscopy
|
40 pages, 12 figures
|
Class. Quantum Grav. 42 (2025) 175011
|
10.1088/1361-6382/adfda5
| null |
gr-qc hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Hairy black holes by gravitational decoupling (GD) are probed to derive the gravitational waveform produced by perturbation theory applied to these compact objects. Using the Regge-Wheeler and Zerilli equations governing the metric perturbations and applying a higher-order WKB method, the quasinormal modes (QNMs) are computed and discussed. Compared to the QNMs produced in the ringdown phase of Reissner-Nordstr\"om black hole solutions, it yields a clear physical signature of primary hair imprinting the hairy GD black hole gravitational waveforms.
|
[
{
"version": "v1",
"created": "Tue, 24 Jun 2025 22:50:43 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 19:13:42 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Guimarães",
"V. F.",
""
],
[
"Cavalcanti",
"R. T.",
""
],
[
"da Rocha",
"R.",
""
]
] |
2506.20229
|
Tetsuro Tsuji
|
Tetsuro Tsuji, Koichiro Takita, Satoshi Taguchi
|
Slip-flow theory for thermo-osmosis based on a kinetic model with near-wall potential
|
47 pages, 11 figures
| null | null | null |
physics.flu-dyn cond-mat.mes-hall cond-mat.stat-mech physics.app-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, thermal-slip coefficients in slip boundary conditions of the Stokes equation are derived using the generalized slip-flow theory, with special interest in the role of near-wall potential in micro- and nanoscale flows. As the model of fluids and fluid-solid interaction, we employ the model Boltzmann equation for dilute gases and the diffuse-reflection boundaries with near-wall potential, respectively. It is found that, when the mean free path of gas molecules and the effective range of potential are of the same order of magnitude, the thermal-slip boundary condition can be derived in the near-continuum limit. In the derived slip-flow theory, the thermal-slip coefficient and the boundary-layer corrections (i.e., Knudsen-layer corrections) are determined by solving the kinetic boundary-layer problems (i.e., Knudsen-layer problems) that include external-force terms and inhomogeneous terms both driven by the potential. As an application of the slip-flow theory, thermo-osmosis between two parallel plates with uniform temperature gradients is analyzed. The results of the slip-flow theory are validated by comparing them with those of the direct numerical analysis of the same problem. Furthermore, it is found that thermo-osmosis and thermal slip on the plates are significantly affected by the features of the near-wall potential; even the gas-flow direction can be reversed when the near-wall potential is repulsive. Such a flow reversal is qualitatively similar to thermo-osmosis in liquids reported in existing molecular dynamics simulation.
|
[
{
"version": "v1",
"created": "Wed, 25 Jun 2025 08:21:33 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 07:47:19 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Tsuji",
"Tetsuro",
""
],
[
"Takita",
"Koichiro",
""
],
[
"Taguchi",
"Satoshi",
""
]
] |
2506.20536
|
Eric Borowski
|
E. S. Borowski, R. I. Hynes, Q. Hunt, A. J. Tetarenko, R. M. Plotkin, T. Shahbaz, P. Gandhi, T. J. Maccarone, J. C. A. Miller-Jones, C. O. Heinke, A. W. Shaw, T. D. Russell, G. R. Sivakoff, P. A. Charles, E. V. Palaiologou, P. Reig
|
The Mid-infrared Emitting Jet in the Black Hole V404 Cygni in Quiescence
|
13 pages, 2 figures. Accepted for publication in ApJL
| null | null | null |
astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
Observations of some quiescent black hole X-ray binaries have revealed an excess of mid-infrared (MIR) emission above that expected from their donor stars. In one system, V404 Cygni, this excess has been variously suggested to arise from the accretion disk, circumbinary material, or a compact relativistic jet. Here we present simultaneous James Webb Space Telescope (JWST), Atacama Large Millimeter/submillimeter Array (ALMA), and complementary multi-wavelength observations undertaken to resolve this uncertainty. We observed large-amplitude 21 $\mu$m variability on short timescales with JWST, particularly a dramatic flare which swiftly rose to $\approx2.4$ mJy, over 10 times the lowest observed MIR flux density. Similar variability was simultaneously observed from radio to X-ray wavelengths with other facilities throughout the campaign. This variability and the flat radio/mm/MIR spectral index ($\alpha = 0.04 \pm 0.01$) suggest that the MIR excess at and longward of 21 $\mu$m in V404 Cyg does not arise from the accretion disk or circumbinary material. Instead, the emission at 21 $\mu$m is dominated by synchrotron radiation from a jet which persists into quiescence. This result reinforces the ubiquity of the disk-jet connection in accreting black holes across a range of masses and accretion rates.
|
[
{
"version": "v1",
"created": "Wed, 25 Jun 2025 15:24:53 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 20:06:03 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Borowski",
"E. S.",
""
],
[
"Hynes",
"R. I.",
""
],
[
"Hunt",
"Q.",
""
],
[
"Tetarenko",
"A. J.",
""
],
[
"Plotkin",
"R. M.",
""
],
[
"Shahbaz",
"T.",
""
],
[
"Gandhi",
"P.",
""
],
[
"Maccarone",
"T. J.",
""
],
[
"Miller-Jones",
"J. C. A.",
""
],
[
"Heinke",
"C. O.",
""
],
[
"Shaw",
"A. W.",
""
],
[
"Russell",
"T. D.",
""
],
[
"Sivakoff",
"G. R.",
""
],
[
"Charles",
"P. A.",
""
],
[
"Palaiologou",
"E. V.",
""
],
[
"Reig",
"P.",
""
]
] |
2506.20723
|
Marie Cornelius
|
Marie Cornelius (NBIA and DARK, Niels Bohr Institute), Irene Tamborra (NBIA and DARK, Niels Bohr Institute), Malte Heinlein (MPI Astrophysics, Garching and TUM, Garching), Hans-Thomas Janka (MPI Astrophysics, Garching)
|
Diagnosing electron-neutrino lepton number crossings in core-collapse supernovae: A comparison of methods
|
13 pages, including 3 figures and 1 appendix. Minor changes. Matches version accepted for publication in Phys. Rev. D
|
Phys. Rev. D 112 (2025) 6, 063004
|
10.1103/gqd7-4ynz
| null |
astro-ph.HE hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Fast neutrino flavor conversion may impact the explosion mechanism and nucleosynthesis in core-collapse supernovae. A necessary condition for fast flavor conversion is the presence of crossings in the angular distribution of the electron-neutrino lepton number (ELN) crossing. Because of the computational costs, flavor-dependent angular distributions are not computed by the vast majority of state-of-the-art hydrodynamical simulations; instead, angular distributions are reconstructed employing approximate methods in post-processing. In this work, we evaluate the performance of four methods adopted to diagnose the existence of ELN crossings. For selected post-bounce times, we extract the fluid and thermodynamic properties from spherically symmetric supernova simulations for an $18.6 M_\odot$ progenitor, testing cases with and without muons as well as with and without mixing-length treatment of proto-neutron star convection. We compare the occurrence of crossings in the angular distributions obtained by solving the Boltzmann equations with those in distributions reconstructed from angular moments of our Boltzmann solutions by using the maximum entropy and Minerbo schemes, and also with crossings identified via a polynomial weighting function applied to the angular moments. Our results show that the polynomial method and the Minerbo closure scheme have severe limitations. The maximum entropy approach captures most of the forward crossings, although it fails to reproduce or misidentifies crossings in a subset of our models. These findings highlight the need for robust modeling of the neutrino angular properties in order to assess the impact of flavor conversion on the supernova mechanism.
|
[
{
"version": "v1",
"created": "Wed, 25 Jun 2025 18:00:03 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 15:26:16 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Cornelius",
"Marie",
"",
"NBIA and DARK, Niels Bohr Institute"
],
[
"Tamborra",
"Irene",
"",
"NBIA and DARK, Niels Bohr Institute"
],
[
"Heinlein",
"Malte",
"",
"MPI Astrophysics, Garching and TUM, Garching"
],
[
"Janka",
"Hans-Thomas",
"",
"MPI Astrophysics, Garching"
]
] |
2506.20870
|
Marcelo Sarandy
|
Alan Duriez, Andreia Saguia, Marcelo S. Sarandy
|
Variational simulation of quantum phase transitions induced by boundary fields
|
v2: 11 pages, 10 figures. Published version
|
Phys. Rev. B 112, 094401 (2025)
| null | null |
quant-ph cond-mat.stat-mech
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The characterization of quantum phase transitions is a fundamental task for the understanding of quantum phases of matter, with a number of potential applications in quantum technologies. In this work, we use digital quantum simulation as a resource to theoretically and experimentally study quantum phase transitions. More specifically, we implement the variational quantum eigensolver (VQE) algorithm to the one-dimensional spin-$1/2$ transverse-field Ising chain in the presence of boundary magnetic fields. Such fields can induce a rich phase diagram, including a first-order line and also a continuous wetting transition, which is a quantum version of the classical wetting surface phenomenon. We present results for noiseless simulations of the associated quantum circuits as well as hardware results taken from a superconducting quantum processor. For different regions of the phase diagram, the quantum algorithm allows us to predict the critical value of the magnetic fields responsible for either the first or second-order transitions occurring in the system.
|
[
{
"version": "v1",
"created": "Wed, 25 Jun 2025 22:42:07 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 13:59:24 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Duriez",
"Alan",
""
],
[
"Saguia",
"Andreia",
""
],
[
"Sarandy",
"Marcelo S.",
""
]
] |
2506.21433
|
Rui-Qing Xiao
|
John Ellis, Hong-Jian He, Rui-Qing Xiao, Shi-Ping Zeng
|
Probing Neutral Triple Gauge Couplings via $ZZ$ Production at $e^+e^-$ Colliders with Machine Learning
|
36 pages, 25 Figs + Tables. Improved version
| null | null |
KCL-PH-TH/2025-28, CERN-TH-2025-122
|
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Neutral triple gauge couplings (nTGCs) first arise from the dimension-8 operators of the Standard Model Effective Field Theory (SMEFT), rather than the dimension-4 SM Lagrangian and dimension-6 SMEFT operators, opening up a unique window for probing new physics at the dimension-8 level. In this work, we formulate the nTGC form factors of $ZZV^*$ ($V\!\!=\!Z,\gamma$) that are compatible with the spontaneous breaking of the SU(2)$\otimes$U(1) electroweak gauge symmetry and consistently match the dimension-8 nTGC operators in the broken phase. We study the sensitivities for probing both the $ZZV^*$ form factors and the corresponding new physics scales through $ZZ$ production (with visible/invisible fermionic $Z$ decays) at high energy $e^+e^-$ colliders including CEPC, FCC-ee, ILC and CLIC. In particular, we identify the dimension-8 operator that contributes to the pure triple $Z$ boson coupling $ZZZ^*$ alone, but not the mixed $ZZ\gamma^*$ coupling. We further study the correlations between probes of the $ZZZ^*$ and $ZZ\gamma^*$ couplings. Using machine learning, we show that angular distributions of the final-state fermions can play key roles in suppressing the SM backgrounds. The sensitivities can be further improved by using polarized $e^\mp$ beams. We demonstrate that machine learning is advantageous for handling the 4-body final states from $ZZ$ decays and improves significantly the sensitivity reaches of probes of nTGCs in $e^+e^-$ collisions. We find that nTGC new physics scales can be probed up to the multi-TeV scale at the proposed $e^+e^-$ colliders.
|
[
{
"version": "v1",
"created": "Thu, 26 Jun 2025 16:18:51 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 14:34:28 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ellis",
"John",
""
],
[
"He",
"Hong-Jian",
""
],
[
"Xiao",
"Rui-Qing",
""
],
[
"Zeng",
"Shi-Ping",
""
]
] |
2506.21619
|
Siyi Zhou
|
Siyi Zhou, Yiquan Zhou, Yi He, Xun Zhou, Jinchao Wang, Wei Deng, Jingchen Shu
|
IndexTTS2: A Breakthrough in Emotionally Expressive and Duration-Controlled Auto-Regressive Zero-Shot Text-to-Speech
| null | null | null | null |
cs.CL cs.AI cs.SD eess.AS
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Existing autoregressive large-scale text-to-speech (TTS) models have advantages in speech naturalness, but their token-by-token generation mechanism makes it difficult to precisely control the duration of synthesized speech. This becomes a significant limitation in applications requiring strict audio-visual synchronization, such as video dubbing. This paper introduces IndexTTS2, which proposes a novel, general, and autoregressive model-friendly method for speech duration control. The method supports two generation modes: one explicitly specifies the number of generated tokens to precisely control speech duration; the other freely generates speech in an autoregressive manner without specifying the number of tokens, while faithfully reproducing the prosodic features of the input prompt. Furthermore, IndexTTS2 achieves disentanglement between emotional expression and speaker identity, enabling independent control over timbre and emotion. In the zero-shot setting, the model can accurately reconstruct the target timbre (from the timbre prompt) while perfectly reproducing the specified emotional tone (from the style prompt). To enhance speech clarity in highly emotional expressions, we incorporate GPT latent representations and design a novel three-stage training paradigm to improve the stability of the generated speech. Additionally, to lower the barrier for emotional control, we designed a soft instruction mechanism based on text descriptions by fine-tuning Qwen3, effectively guiding the generation of speech with the desired emotional orientation. Finally, experimental results on multiple datasets show that IndexTTS2 outperforms state-of-the-art zero-shot TTS models in terms of word error rate, speaker similarity, and emotional fidelity. Audio samples are available at: https://index-tts.github.io/index-tts2.github.io/
|
[
{
"version": "v1",
"created": "Mon, 23 Jun 2025 08:33:40 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 10:46:35 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zhou",
"Siyi",
""
],
[
"Zhou",
"Yiquan",
""
],
[
"He",
"Yi",
""
],
[
"Zhou",
"Xun",
""
],
[
"Wang",
"Jinchao",
""
],
[
"Deng",
"Wei",
""
],
[
"Shu",
"Jingchen",
""
]
] |
2506.21807
|
Cyril Voyant
|
Cyril Voyant and Alan Julien and Milan Despotovic and Gilles Notton and Luis Antonio Garcia-Gutierrez and Claudio Francesco Nicolosi and Philippe Blanc and Jamie Bright
|
Stochastic Coefficient of Variation: Assessing the Variability and Forecastability of Solar Irradiance
| null |
Volume 256, Part B, 1 January 2026, 123913
|
10.1016/j.renene.2025.123913
| null |
physics.ao-ph physics.data-an
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This work presents a robust framework for quantifying solar irradiance variability and forecastability through the Stochastic Coefficient of Variation (sCV) and the Forecastability (F). Traditional metrics, such as the standard deviation, fail to isolate stochastic fluctuations from deterministic trends in solar irradiance. By considering clear-sky irradiance as a dynamic upper bound of measurement, sCV provides a normalized, dimensionless measure of variability that theoretically ranges from 0 to 1. F extends sCV by integrating temporal dependencies via maximum autocorrelation, thus linking sCV with F. The proposed methodology is validated using synthetic cyclostationary time series and experimental data from 68 meteorological stations (in Spain). Our comparative analyses demonstrate that sCV and F proficiently encapsulate multi-scale fluctuations, while addressing significant limitations inherent in traditional metrics. This comprehensive framework enables a refined quantification of solar forecast uncertainty, supporting improved decision-making in flexibility procurement and operational strategies. By assessing variability and forecastability across multiple time scales, it enhances real-time monitoring capabilities and informs adaptive energy management approaches, such as dynamic outage management and risk-adjusted capacity allocation
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 19:21:40 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Voyant",
"Cyril",
""
],
[
"Julien",
"Alan",
""
],
[
"Despotovic",
"Milan",
""
],
[
"Notton",
"Gilles",
""
],
[
"Garcia-Gutierrez",
"Luis Antonio",
""
],
[
"Nicolosi",
"Claudio Francesco",
""
],
[
"Blanc",
"Philippe",
""
],
[
"Bright",
"Jamie",
""
]
] |
2506.21824
|
Hikaru Watanabe
|
Hikaru Watanabe, Rikuto Oiwa, Gakuto Kusuno, Takuya Satoh, Ryotaro Arita
|
Symmetry analysis of cross-circular and parallel-circular Raman optical activity
|
13 pages, 2 figures, 3 tables
| null |
10.1103/45jq-j23c
| null |
cond-mat.mtrl-sci
|
http://creativecommons.org/publicdomain/zero/1.0/
|
The Raman scattering regarding the circularly-polarized incident and scattered lights is closely related to the circular activity of a given system. We investigate the symmetry of its activity, called the cross-circular and parallel-circular Raman optical activity. The analysis is systematically performed with the magnetic point groups and indicates that the response allows for a useful diagnosis of the symmetry of materials like chirality and (magneto-)axiality. It is also shown that the Stokes and anti-Stokes processes are related to each other by the conserved antiunitary symmetry for the time-reversal operation and that combined with the mirror reflection.
|
[
{
"version": "v1",
"created": "Thu, 26 Jun 2025 23:56:43 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 03:23:40 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Watanabe",
"Hikaru",
""
],
[
"Oiwa",
"Rikuto",
""
],
[
"Kusuno",
"Gakuto",
""
],
[
"Satoh",
"Takuya",
""
],
[
"Arita",
"Ryotaro",
""
]
] |
2506.21915
|
Dake Sun
|
D. Sun and S. Zhou
|
An Effective Two-Phase Genetic Algorithm for Solving the Resource Constrained Project Scheduling Problem (RCPSP)
|
12 pages
| null | null | null |
cs.NE math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
This note presents a simple and effective variation of genetic algorithm (GA) for solving RCPSP, denoted as 2-Phase Genetic Algorithm (2PGA). The 2PGA implements GA parent selection in two phases: Phase-1 includes the best current solutions in the parent pool, and Phase-2 excludes the best current solutions from the parent pool. The 2PGA carries out the GA evolution by alternating the two phases iteratively. In exploring a solution space, the Phase-1 emphasizes intensification in current neighborhood, while the Phase-2 emphasizes diversification to escape local traps. The 2PGA was tested on the standard benchmark problems in PSPLIB, the results have shown that the algorithm is effective and has improved some of the best heuristic solutions.
|
[
{
"version": "v1",
"created": "Fri, 27 Jun 2025 04:59:48 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 16:48:00 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Sun",
"D.",
""
],
[
"Zhou",
"S.",
""
]
] |
2506.23294
|
Lars Hupel
|
Mostafa Abdelrahman, Filip Rezabek, Lars Hupel, Kilian Glas, Georg Carle
|
Threshold Signatures for Central Bank Digital Currencies
| null | null | null | null |
cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Digital signatures are crucial for securing Central Bank Digital Currencies (CBDCs) transactions. Like most forms of digital currencies, CBDC solutions rely on signatures for transaction authenticity and integrity, leading to major issues in the case of private key compromise. Our work explores threshold signature schemes (TSSs) in the context of CBDCs. TSSs allow distributed key management and signing, reducing the risk of a compromised key. We analyze CBDC-specific requirements, considering the applicability of TSSs, and use Filia CBDC solution as a base for a detailed evaluation. As most of the current solutions rely on ECDSA for compatibility, we focus on ECDSA-based TSSs and their supporting libraries. Our performance evaluation measured the computational and communication complexity across key processes, as well as the throughput and latency of end-to-end transactions. The results confirm that TSS can enhance the security of CBDC implementations while maintaining acceptable performance for real-world deployments.
|
[
{
"version": "v1",
"created": "Sun, 29 Jun 2025 15:31:11 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 08:22:01 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Abdelrahman",
"Mostafa",
""
],
[
"Rezabek",
"Filip",
""
],
[
"Hupel",
"Lars",
""
],
[
"Glas",
"Kilian",
""
],
[
"Carle",
"Georg",
""
]
] |
2506.23367
|
Paige Tutt\"os\'i
|
Paige Tutt\"os\'i, H. Henny Yeung, Yue Wang, Jean-Julien Aucouturier, Angelica Lim
|
You Sound a Little Tense: L2 Tailored Clear TTS Using Durational Vowel Properties
|
Accepted to ISCA Speech Synthesis Workshop, 2025, Project webpage here: https://rosielab.github.io/clear_speech/ Code here: https://github.com/chocobearz/Matcha-TTS-L2-clarity
| null | null | null |
cs.SD cs.CL eess.AS
|
http://creativecommons.org/licenses/by/4.0/
|
We present the first text-to-speech (TTS) system tailored to second language (L2) speakers. We use duration differences between American English tense (longer) and lax (shorter) vowels to create a "clarity mode" for Matcha-TTS. Our perception studies showed that French-L1, English-L2 listeners had fewer (at least 9.15%) transcription errors when using our clarity mode, and found it more encouraging and respectful than overall slowed down speech. Remarkably, listeners were not aware of these effects: despite the decreased word error rate in clarity mode, listeners still believed that slowing all target words was the most intelligible, suggesting that actual intelligibility does not correlate with perceived intelligibility. Additionally, we found that Whisper-ASR did not use the same cues as L2 speakers to differentiate difficult vowels and is not sufficient to assess the intelligibility of TTS systems for these individuals.
|
[
{
"version": "v1",
"created": "Sun, 29 Jun 2025 18:55:05 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 08:54:41 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Tuttösí",
"Paige",
""
],
[
"Yeung",
"H. Henny",
""
],
[
"Wang",
"Yue",
""
],
[
"Aucouturier",
"Jean-Julien",
""
],
[
"Lim",
"Angelica",
""
]
] |
2506.23903
|
Hamza Rasaee
|
Hamza Rasaee, Taha Koleilat, Hassan Rivaz
|
GroundingDINO-US-SAM: Text-Prompted Multi-Organ Segmentation in Ultrasound with LoRA-Tuned Vision-Language Models
|
11 pages, 3 figures, 7 tables
| null |
10.1109/TUFFC.2025.3605285
| null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Accurate and generalizable object segmentation in ultrasound imaging remains a significant challenge due to anatomical variability, diverse imaging protocols, and limited annotated data. In this study, we propose a prompt-driven vision-language model (VLM) that integrates Grounding DINO with SAM2 to enable object segmentation across multiple ultrasound organs. A total of 18 public ultrasound datasets, encompassing the breast, thyroid, liver, prostate, kidney, and paraspinal muscle, were utilized. These datasets were divided into 15 for fine-tuning and validation of Grounding DINO using Low Rank Adaptation (LoRA) to the ultrasound domain, and 3 were held out entirely for testing to evaluate performance in unseen distributions. Comprehensive experiments demonstrate that our approach outperforms state-of-the-art segmentation methods, including UniverSeg, MedSAM, MedCLIP-SAM, BiomedParse, and SAMUS on most seen datasets while maintaining strong performance on unseen datasets without additional fine-tuning. These results underscore the promise of VLMs in scalable and robust ultrasound image analysis, reducing dependence on large, organ-specific annotated datasets. We will publish our code on code.sonography.ai after acceptance.
|
[
{
"version": "v1",
"created": "Mon, 30 Jun 2025 14:33:44 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 00:41:31 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Rasaee",
"Hamza",
""
],
[
"Koleilat",
"Taha",
""
],
[
"Rivaz",
"Hassan",
""
]
] |
2507.00790
|
Huaqiu Li
|
Huaqiu Li and Yong Wang and Tongwen Huang and Hailang Huang and Haoqian Wang and Xiangxiang Chu
|
LD-RPS: Zero-Shot Unified Image Restoration via Latent Diffusion Recurrent Posterior Sampling
| null | null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Unified image restoration is a significantly challenging task in low-level vision. Existing methods either make tailored designs for specific tasks, limiting their generalizability across various types of degradation, or rely on training with paired datasets, thereby suffering from closed-set constraints. To address these issues, we propose a novel, dataset-free, and unified approach through recurrent posterior sampling utilizing a pretrained latent diffusion model. Our method incorporates the multimodal understanding model to provide sematic priors for the generative model under a task-blind condition. Furthermore, it utilizes a lightweight module to align the degraded input with the generated preference of the diffusion model, and employs recurrent refinement for posterior sampling. Extensive experiments demonstrate that our method outperforms state-of-the-art methods, validating its effectiveness and robustness. Our code and data are available at https://github.com/AMAP-ML/LD-RPS.
|
[
{
"version": "v1",
"created": "Tue, 01 Jul 2025 14:25:09 GMT"
},
{
"version": "v2",
"created": "Fri, 04 Jul 2025 20:39:37 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 08:45:06 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Li",
"Huaqiu",
""
],
[
"Wang",
"Yong",
""
],
[
"Huang",
"Tongwen",
""
],
[
"Huang",
"Hailang",
""
],
[
"Wang",
"Haoqian",
""
],
[
"Chu",
"Xiangxiang",
""
]
] |
2507.00917
|
Qingrui Zhao
|
Xiaoxiao Long, Qingrui Zhao, Kaiwen Zhang, Zihao Zhang, Dingrui Wang, Yumeng Liu, Zhengjie Shu, Yi Lu, Shouzheng Wang, Xinzhe Wei, Wei Li, Wei Yin, Yao Yao, Jia Pan, Qiu Shen, Ruigang Yang, Xun Cao, Qionghai Dai
|
A Survey: Learning Embodied Intelligence from Physical Simulators and World Models
|
Update with recent progresses. 49pages, 25figures, 6tables, github repository avalible in https://github.com/NJU3DV-LoongGroup/Embodied-World-Models-Survey
| null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
The pursuit of artificial general intelligence (AGI) has placed embodied intelligence at the forefront of robotics research. Embodied intelligence focuses on agents capable of perceiving, reasoning, and acting within the physical world. Achieving robust embodied intelligence requires not only advanced perception and control, but also the ability to ground abstract cognition in real-world interactions. Two foundational technologies, physical simulators and world models, have emerged as critical enablers in this quest. Physical simulators provide controlled, high-fidelity environments for training and evaluating robotic agents, allowing safe and efficient development of complex behaviors. In contrast, world models empower robots with internal representations of their surroundings, enabling predictive planning and adaptive decision-making beyond direct sensory input. This survey systematically reviews recent advances in learning embodied AI through the integration of physical simulators and world models. We analyze their complementary roles in enhancing autonomy, adaptability, and generalization in intelligent robots, and discuss the interplay between external simulation and internal modeling in bridging the gap between simulated training and real-world deployment. By synthesizing current progress and identifying open challenges, this survey aims to provide a comprehensive perspective on the path toward more capable and generalizable embodied AI systems. We also maintain an active repository that contains up-to-date literature and open-source projects at https://github.com/NJU3DV-LoongGroup/Embodied-World-Models-Survey.
|
[
{
"version": "v1",
"created": "Tue, 01 Jul 2025 16:23:00 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Jul 2025 16:06:05 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 01:44:58 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Long",
"Xiaoxiao",
""
],
[
"Zhao",
"Qingrui",
""
],
[
"Zhang",
"Kaiwen",
""
],
[
"Zhang",
"Zihao",
""
],
[
"Wang",
"Dingrui",
""
],
[
"Liu",
"Yumeng",
""
],
[
"Shu",
"Zhengjie",
""
],
[
"Lu",
"Yi",
""
],
[
"Wang",
"Shouzheng",
""
],
[
"Wei",
"Xinzhe",
""
],
[
"Li",
"Wei",
""
],
[
"Yin",
"Wei",
""
],
[
"Yao",
"Yao",
""
],
[
"Pan",
"Jia",
""
],
[
"Shen",
"Qiu",
""
],
[
"Yang",
"Ruigang",
""
],
[
"Cao",
"Xun",
""
],
[
"Dai",
"Qionghai",
""
]
] |
2507.01203
|
Mark Raizen
|
Mark G. Raizen
|
Testing Distinguishability and Determinism in Newly Created Atoms
| null | null | null | null |
quant-ph physics.atom-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We propose and analyze experiments to test distinguishability and determinism in atoms. As a first step, we consider experiments to test if newly created atoms are distinguishable from each other due to small differences in the nuclear magnetic moment. The idea is to measure the hyperfine splitting for newly created stable isotopes and compare them to the same isotopes found in nature. The three cases analyzed are strontium-87 with an odd number of neutrons, lutetium-175 with an odd number of protons, and lutetium-176 with an odd number of protons and neutrons. We also propose that a time-dependent nuclear magnetic moment may enable deterministic tracking of the life cycle of an isotope, and prediction of age since creation and time until decay of radioisotopes. The case of thulium-170 is considered, it is particularly interesting as a radioisotope with an odd number of protons and neutrons. More generally, we argue that any atomic transition could change in time, either due to the atomic nucleus or to the electron charge distribution. The latter could be tested in a dual atomic clock in a ladder configuration to search for the aging of an electronic transition.
|
[
{
"version": "v1",
"created": "Tue, 01 Jul 2025 21:44:33 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 20:12:46 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Raizen",
"Mark G.",
""
]
] |
2507.02654
|
Rui Xie
|
Rui Xie, Asad Ul Haq, Yunhua Fang, Linsen Ma, Sanchari Sen, Swagath Venkataramani, Liu Liu, Tong Zhang
|
Breaking the HBM Bit Cost Barrier: Domain-Specific ECC for AI Inference Infrastructure
| null | null | null | null |
cs.AR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
High-Bandwidth Memory (HBM) delivers exceptional bandwidth and energy efficiency for AI workloads, but its high cost per bit, driven in part by stringent on-die reliability requirements, poses a growing barrier to scalable deployment. This work explores a system-level approach to cost reduction by eliminating on-die ECC and shifting all fault management to the memory controller. We introduce a domain-specific ECC framework combining large-codeword Reed--Solomon~(RS) correction with lightweight fine-grained CRC detection, differential parity updates to mitigate write amplification, and tunable protection based on data importance. Our evaluation using LLM inference workloads shows that, even under raw HBM bit error rates up to $10^{-3}$, the system retains over 78\% of throughput and 97\% of model accuracy compared with systems equipped with ideal error-free HBM. By treating reliability as a tunable system parameter rather than a fixed hardware constraint, our design opens a new path toward low-cost, high-performance HBM deployment in AI infrastructure.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 14:18:41 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 14:58:07 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Xie",
"Rui",
""
],
[
"Haq",
"Asad Ul",
""
],
[
"Fang",
"Yunhua",
""
],
[
"Ma",
"Linsen",
""
],
[
"Sen",
"Sanchari",
""
],
[
"Venkataramani",
"Swagath",
""
],
[
"Liu",
"Liu",
""
],
[
"Zhang",
"Tong",
""
]
] |
2507.03034
|
Yiming Li
|
Yiming Li, Shuo Shao, Yu He, Junfeng Guo, Tianwei Zhang, Zhan Qin, Pin-Yu Chen, Michael Backes, Philip Torr, Dacheng Tao, Kui Ren
|
Rethinking Data Protection in the (Generative) Artificial Intelligence Era
|
Perspective paper for a broader scientific audience. The first two authors contributed equally to this paper. 13 pages
| null | null | null |
cs.LG cs.AI cs.CR cs.CV cs.CY
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The (generative) artificial intelligence (AI) era has profoundly reshaped the meaning and value of data. No longer confined to static content, data now permeates every stage of the AI lifecycle from the training samples that shape model parameters to the prompts and outputs that drive real-world model deployment. This shift renders traditional notions of data protection insufficient, while the boundaries of what needs safeguarding remain poorly defined. Failing to safeguard data in AI systems can inflict societal and individual, underscoring the urgent need to clearly delineate the scope of and rigorously enforce data protection. In this perspective, we propose a four-level taxonomy, including non-usability, privacy preservation, traceability, and deletability, that captures the diverse protection needs arising in modern (generative) AI models and systems. Our framework offers a structured understanding of the trade-offs between data utility and control, spanning the entire AI pipeline, including training datasets, model weights, system prompts, and AI-generated content. We analyze representative technical approaches at each level and reveal regulatory blind spots that leave critical assets exposed. By offering a structured lens to align future AI technologies and governance with trustworthy data practices, we underscore the urgency of rethinking data protection for modern AI techniques and provide timely guidance for developers, researchers, and regulators alike.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 02:45:51 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Jul 2025 06:52:55 GMT"
},
{
"version": "v3",
"created": "Sat, 19 Jul 2025 05:58:02 GMT"
},
{
"version": "v4",
"created": "Wed, 03 Sep 2025 06:34:12 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Li",
"Yiming",
""
],
[
"Shao",
"Shuo",
""
],
[
"He",
"Yu",
""
],
[
"Guo",
"Junfeng",
""
],
[
"Zhang",
"Tianwei",
""
],
[
"Qin",
"Zhan",
""
],
[
"Chen",
"Pin-Yu",
""
],
[
"Backes",
"Michael",
""
],
[
"Torr",
"Philip",
""
],
[
"Tao",
"Dacheng",
""
],
[
"Ren",
"Kui",
""
]
] |
2507.03180
|
Ian Roederer
|
Ian U. Roederer, Rana Ezzeddine
|
Habitable Worlds Observatory: The Nature of the Astrophysical r-process
|
5 pages (Science Case Development Document submitted to the ASP proceedings for the HWO2025 conference)
| null | null | null |
astro-ph.IM astro-ph.GA astro-ph.SR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present the science case for characterizing the origin of the heaviest elements on the periodic table, with a focus on those produced by the rapid neutron-capture process (r-process), using the Habitable Worlds Observatory (HWO). High-resolution ultraviolet (UV) spectroscopy can increase the number of r-process elements detectable in cool stars by more than 50% relative to optical and infrared spectra. These elements are key to characterizing the physical conditions that govern the r-process and identify the nature, sites, and environments where r-process events occurred. HWO has the potential to greatly expand the sample of stars where rarely studied heavy elements can be detected beyond the Solar neighborhood to the Galactic halo, globular clusters, and dwarf galaxies.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 21:26:33 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 02:42:40 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Roederer",
"Ian U.",
""
],
[
"Ezzeddine",
"Rana",
""
]
] |
2507.03181
|
Ian Roederer
|
Ian U. Roederer, Rana Ezzeddine, Jennifer S. Sobeck
|
Habitable Worlds Observatory: The Nature of the First Stars
|
5 pages (Science Case Development Document submitted to the ASP proceedings for the HWO2025 conference)
| null | null | null |
astro-ph.IM astro-ph.GA astro-ph.SR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present the science case for characterizing the nature of the first stars using the Habitable Worlds Observatory (HWO). High-resolution ultraviolet (UV) spectroscopy with the HWO has the potential to confirm any surviving low-mass zero-metallicity first stars by placing unprecedented low limits on their metal abundances. It also has the potential to substantially increase the number of elements detectable in the spectra of known long-lived low-mass stars, which exhibit extremely low metal abundances that reveal the metals produced by the first stars. Elements important for this science case with UV transitions include C, Mg, Al, Si, P, S, Sc, V, Cr, Mn, Fe, Co, Ni, Cu, and Zn. HWO would expand the discovery space when compared with the Hubble Space Telescope by enabling high-resolution UV spectroscopy for much fainter stars throughout the Milky Way and neighboring stellar systems.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 21:27:08 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 02:39:51 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Roederer",
"Ian U.",
""
],
[
"Ezzeddine",
"Rana",
""
],
[
"Sobeck",
"Jennifer S.",
""
]
] |
2507.03907
|
Su Gao
|
Su Gao and Feng Li
|
On the Isomorphism Relation for Omnigenous Locally Finite Groups
| null | null | null | null |
math.LO math.GR
|
http://creativecommons.org/licenses/by/4.0/
|
The concept of an omnigenous locally finite group was introduced in [2] as a generalization of Hall's universal countable locally finite group. In this paper we show that the class of all countable omnigenous locally finite groups is Borel complete, hence it has the maximum Borel cardinality of isomorphism types among all countable structures.
[2] M. Etedadialiabadi, S. Gao, F. Le Ma\^{i}tre, J. Melleray, Dense locally finite subgroups of automorphism groups of ultraextensive spaces, Adv. Math. 391 (2021), 107966.
|
[
{
"version": "v1",
"created": "Sat, 05 Jul 2025 05:46:28 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 02:05:00 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Gao",
"Su",
""
],
[
"Li",
"Feng",
""
]
] |
2507.04233
|
Xiaochen Wei
|
Xiaochen Wei, Weiwei Guo, Zenghui Zhang, Wenxian Yu
|
Grid-Reg: Detector-Free Gridized Feature Learning and Matching for Large-Scale SAR-Optical Image Registration
|
This work has been submitted to the IEEE for possible publication
| null | null | null |
eess.IV cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
It is highly challenging to register large-scale, heterogeneous SAR and optical images, particularly across platforms, due to significant geometric, radiometric, and temporal differences, which most existing methods struggle to address. To overcome these challenges, we propose Grid-Reg, a grid-based multimodal registration framework comprising a domain-robust descriptor extraction network, Hybrid Siamese Correlation Metric Learning Network (HSCMLNet), and a grid-based solver (Grid-Solver) for transformation parameter estimation. In heterogeneous imagery with large modality gaps and geometric differences, obtaining accurate correspondences is inherently difficult. To robustly measure similarity between gridded patches, HSCMLNet integrates a hybrid Siamese module with a correlation metric learning module (CMLModule) based on equiangular unit basis vectors (EUBVs), together with a manifold consistency loss to promote modality-invariant, discriminative feature learning. The Grid-Solver estimates transformation parameters by minimizing a global grid matching loss through a progressive dual-loop search strategy to reliably find patch correspondences across entire images. Furthermore, we curate a challenging benchmark dataset for SAR-to-optical registration using real-world UAV MiniSAR data and Google Earth optical imagery. Extensive experiments demonstrate that our proposed approach achieves superior performance over state-of-the-art methods.
|
[
{
"version": "v1",
"created": "Sun, 06 Jul 2025 03:43:18 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 04:09:53 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wei",
"Xiaochen",
""
],
[
"Guo",
"Weiwei",
""
],
[
"Zhang",
"Zenghui",
""
],
[
"Yu",
"Wenxian",
""
]
] |
2507.04416
|
Xiuying Wei
|
Xiuying Wei, Anunay Yadav, Razvan Pascanu, Caglar Gulcehre
|
RAT: Bridging RNN Efficiency and Attention Accuracy via Chunk-based Sequence Modeling
| null | null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Transformers have become the cornerstone of modern large-scale language models, but their reliance on softmax attention poses a computational bottleneck at both training and inference. Recurrent models offer high efficiency, but compressing the full sequence into a fixed-size and holistic representation suffers from memory degradation in long contexts and limits fine-grained retrieval. To address this, we propose RAT, an intermediate design that bridges the efficiency of RNNs and capacity of attention. RAT partitions the input into chunks, applies recurrence within each chunk for local dependencies, and softmax-based attention across chunks for long-range interactions. This design mitigates memory degradation and enables direct access to distant tokens, while retaining computational efficiency. Empirically, with a chunk size of 16, the RAT block achieves a 7x improvement in training speed with 100K token sequences and 9x in generation at the 4K position, while maintaining similar performance compared to standard attention. We demonstrate this by training 1.3B parameter models from scratch and performing large-scale evaluations, including short- and long-context benchmarks, as well as supervised fine-tuning~(SFT). We further propose a hybrid architecture that interleaves RAT with local attention. By combining efficient long-range modeling with strong local interactions, this hybrid design not only improves inference speed and reduces cache memory usage, but also consistently enhances performance and shows the overall best results. Code is available at https://github.com/CLAIRE-Labo/RAT.
|
[
{
"version": "v1",
"created": "Sun, 06 Jul 2025 15:08:49 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 14:28:23 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wei",
"Xiuying",
""
],
[
"Yadav",
"Anunay",
""
],
[
"Pascanu",
"Razvan",
""
],
[
"Gulcehre",
"Caglar",
""
]
] |
2507.04618
|
Yan Gong
|
CSST Collaboration: Yan Gong, Haitao Miao, Hu Zhan, Zhao-Yu Li, Jinyi Shangguan, Haining Li, Chao Liu, Xuefei Chen, Haibo Yuan, Jilin Zhou, Hui-Gen Liu, Cong Yu, Jianghui Ji, Zhaoxiang Qi, Jiacheng Liu, Zigao Dai, Xiaofeng Wang, Zhenya Zheng, Lei Hao, Jiangpei Dou, Yiping Ao, Zhenhui Lin, Kun Zhang, Wei Wang, Guotong Sun, Ran Li, Guoliang Li, Youhua Xu, Xinfeng Li, Shengyang Li, Peng Wu, Jiuxing Zhang, Bo Wang, Jinming Bai, Yi-Fu Cai, Zheng Cai, Jie Cao, Kwan Chuen Chan, Jin Chang, Xiaodian Chen, Xuelei Chen, Yuqin Chen, Yun Chen, Wei Cui, Subo Dong, Pu Du, Wenying Duan, Junhui Fan, LuLu Fan, Zhou Fan, Zuhui Fan, Taotao Fang, Jianning Fu, Liping Fu, Zhensen Fu, Jian Gao, Shenghong Gu, Yidong Gu, Qi Guo, Zhanwen Han, Bin Hu, Zhiqi Huang, Luis C. Ho, Linhua Jiang, Ning Jiang, Yipeng Jing, Xi Kang, Xu Kong, Cheng Li, Chengyuan Li, Di Li, Jing Li, Nan Li, Yang A. Li, Shilong Liao, Weipeng Lin, Fengshan Liu, Jifeng Liu, Xiangkun Liu, Zhuokai Liu, Ruiqing Mao, Shude Mao, Xianmin Meng, Xiaoying Pang, Xiyan Peng, Yingjie Peng, Huanyuan Shan, Juntai Shen, Shiyin Shen, Zhiqiang Shen, Sheng-Cai Shi, Yong Shi, Siyuan Tan, Hao Tian, Jianmin Wang, Jun-Xian Wang, Xin Wang, Yuting Wang, Hong Wu, Jingwen Wu, Xuebing Wu, Chun Xu, Xiang-Xiang Xue, Yongquan Xue, Ji Yang, Xiaohu Yang, Qijun Yao, Fangting Yuan, Zhen Yuan, Jun Zhang, Pengjie Zhang, Tianmeng Zhang, Wei Zhang, Xin Zhang, Gang Zhao, Gongbo Zhao, Hongen Zhong, Jing Zhong, Liyong Zhou, Wei Zhu, and Ying Zu
|
Introduction to the Chinese Space Station Survey Telescope (CSST)
|
45 pages, 12 figures, 1 table. Discussions, references, and contributors are added
| null | null | null |
astro-ph.IM astro-ph.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Chinese Space Station Survey Telescope (CSST) is a next-generation Stage-IV sky survey telescope, distinguished by its large field of view (FoV), high image quality, and multi-band observation capabilities. It can simultaneously conduct precise measurements of the Universe by performing multi-color photometric imaging and slitless spectroscopic surveys. The CSST is equipped with five scientific instruments, i.e. Multi-band Imaging and Slitless Spectroscopy Survey Camera (SC), Multi-Channel Imager (MCI), Integral Field Spectrograph (IFS), Cool Planet Imaging Coronagraph (CPI-C), and THz Spectrometer (TS). Using these instruments, CSST is expected to make significant contributions and discoveries across various astronomical fields, including cosmology, galaxy and active galactic nuclei (AGN), the Milky Way and nearby galaxies, stars, exoplanets, Solar System objects, astrometry, and transients and variable sources. This review aims to provide a comprehensive overview of the CSST instruments, observational capabilities, data products, and scientific potential.
|
[
{
"version": "v1",
"created": "Mon, 07 Jul 2025 02:26:08 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 07:11:20 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"CSST Collaboration",
"",
""
],
[
"Gong",
"Yan",
""
],
[
"Miao",
"Haitao",
""
],
[
"Zhan",
"Hu",
""
],
[
"Li",
"Zhao-Yu",
""
],
[
"Shangguan",
"Jinyi",
""
],
[
"Li",
"Haining",
""
],
[
"Liu",
"Chao",
""
],
[
"Chen",
"Xuefei",
""
],
[
"Yuan",
"Haibo",
""
],
[
"Zhou",
"Jilin",
""
],
[
"Liu",
"Hui-Gen",
""
],
[
"Yu",
"Cong",
""
],
[
"Ji",
"Jianghui",
""
],
[
"Qi",
"Zhaoxiang",
""
],
[
"Liu",
"Jiacheng",
""
],
[
"Dai",
"Zigao",
""
],
[
"Wang",
"Xiaofeng",
""
],
[
"Zheng",
"Zhenya",
""
],
[
"Hao",
"Lei",
""
],
[
"Dou",
"Jiangpei",
""
],
[
"Ao",
"Yiping",
""
],
[
"Lin",
"Zhenhui",
""
],
[
"Zhang",
"Kun",
""
],
[
"Wang",
"Wei",
""
],
[
"Sun",
"Guotong",
""
],
[
"Li",
"Ran",
""
],
[
"Li",
"Guoliang",
""
],
[
"Xu",
"Youhua",
""
],
[
"Li",
"Xinfeng",
""
],
[
"Li",
"Shengyang",
""
],
[
"Wu",
"Peng",
""
],
[
"Zhang",
"Jiuxing",
""
],
[
"Wang",
"Bo",
""
],
[
"Bai",
"Jinming",
""
],
[
"Cai",
"Yi-Fu",
""
],
[
"Cai",
"Zheng",
""
],
[
"Cao",
"Jie",
""
],
[
"Chan",
"Kwan Chuen",
""
],
[
"Chang",
"Jin",
""
],
[
"Chen",
"Xiaodian",
""
],
[
"Chen",
"Xuelei",
""
],
[
"Chen",
"Yuqin",
""
],
[
"Chen",
"Yun",
""
],
[
"Cui",
"Wei",
""
],
[
"Dong",
"Subo",
""
],
[
"Du",
"Pu",
""
],
[
"Duan",
"Wenying",
""
],
[
"Fan",
"Junhui",
""
],
[
"Fan",
"LuLu",
""
],
[
"Fan",
"Zhou",
""
],
[
"Fan",
"Zuhui",
""
],
[
"Fang",
"Taotao",
""
],
[
"Fu",
"Jianning",
""
],
[
"Fu",
"Liping",
""
],
[
"Fu",
"Zhensen",
""
],
[
"Gao",
"Jian",
""
],
[
"Gu",
"Shenghong",
""
],
[
"Gu",
"Yidong",
""
],
[
"Guo",
"Qi",
""
],
[
"Han",
"Zhanwen",
""
],
[
"Hu",
"Bin",
""
],
[
"Huang",
"Zhiqi",
""
],
[
"Ho",
"Luis C.",
""
],
[
"Jiang",
"Linhua",
""
],
[
"Jiang",
"Ning",
""
],
[
"Jing",
"Yipeng",
""
],
[
"Kang",
"Xi",
""
],
[
"Kong",
"Xu",
""
],
[
"Li",
"Cheng",
""
],
[
"Li",
"Chengyuan",
""
],
[
"Li",
"Di",
""
],
[
"Li",
"Jing",
""
],
[
"Li",
"Nan",
""
],
[
"Li",
"Yang A.",
""
],
[
"Liao",
"Shilong",
""
],
[
"Lin",
"Weipeng",
""
],
[
"Liu",
"Fengshan",
""
],
[
"Liu",
"Jifeng",
""
],
[
"Liu",
"Xiangkun",
""
],
[
"Liu",
"Zhuokai",
""
],
[
"Mao",
"Ruiqing",
""
],
[
"Mao",
"Shude",
""
],
[
"Meng",
"Xianmin",
""
],
[
"Pang",
"Xiaoying",
""
],
[
"Peng",
"Xiyan",
""
],
[
"Peng",
"Yingjie",
""
],
[
"Shan",
"Huanyuan",
""
],
[
"Shen",
"Juntai",
""
],
[
"Shen",
"Shiyin",
""
],
[
"Shen",
"Zhiqiang",
""
],
[
"Shi",
"Sheng-Cai",
""
],
[
"Shi",
"Yong",
""
],
[
"Tan",
"Siyuan",
""
],
[
"Tian",
"Hao",
""
],
[
"Wang",
"Jianmin",
""
],
[
"Wang",
"Jun-Xian",
""
],
[
"Wang",
"Xin",
""
],
[
"Wang",
"Yuting",
""
],
[
"Wu",
"Hong",
""
],
[
"Wu",
"Jingwen",
""
],
[
"Wu",
"Xuebing",
""
],
[
"Xu",
"Chun",
""
],
[
"Xue",
"Xiang-Xiang",
""
],
[
"Xue",
"Yongquan",
""
],
[
"Yang",
"Ji",
""
],
[
"Yang",
"Xiaohu",
""
],
[
"Yao",
"Qijun",
""
],
[
"Yuan",
"Fangting",
""
],
[
"Yuan",
"Zhen",
""
],
[
"Zhang",
"Jun",
""
],
[
"Zhang",
"Pengjie",
""
],
[
"Zhang",
"Tianmeng",
""
],
[
"Zhang",
"Wei",
""
],
[
"Zhang",
"Xin",
""
],
[
"Zhao",
"Gang",
""
],
[
"Zhao",
"Gongbo",
""
],
[
"Zhong",
"Hongen",
""
],
[
"Zhong",
"Jing",
""
],
[
"Zhou",
"Liyong",
""
],
[
"Zhu",
"Wei",
""
],
[
"Zu",
"Ying",
""
]
] |
2507.05406
|
Andrei Mironov
|
A. Mironov, A. Oreshina, A. Popolitov
|
Phases in WLZZ Matrix Models
|
15 pages, LaTeX
|
Phys. Lett. B869 (2025) 139845
|
10.1016/j.physletb.2025.139845
|
FIAN/TD-13/25; IITP/TH-21/25; ITEP/TH-24/25; MIPT/TH-17/25
|
hep-th math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We discuss the space of solutions to the Ward identities associated with the WLZZ models. We mostly concentrate on the case of these models described by a two-matrix model with the cubic potential in one of the matrices. We study how this space of solutions can be described by the freedom in choosing integration contours in the matrix integral.
|
[
{
"version": "v1",
"created": "Mon, 07 Jul 2025 18:45:57 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Mironov",
"A.",
""
],
[
"Oreshina",
"A.",
""
],
[
"Popolitov",
"A.",
""
]
] |
2507.06320
|
Kaifei Ning
|
Kaifei Ning, Michael Ramsey-Musolf
|
Revisiting the Electron EDM in the NMSSM
|
8 pages, 6 figures, 1 table
| null | null | null |
hep-ph hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
The Next-to-Minimal Supersymmetric Standard Model (NMSSM) with explicit CP violation offers a promising framework for explaining the observed baryon asymmetry while remaining consistent with stringent electric dipole moment (EDM) bounds. In this work, we identify a previously overlooked set of two-loop diagrams that contribute to fermion EDMs in the NMSSM. Our analysis of the electron EDM shows that, for certain NMSSM specific CP-violating phases, these diagrams partially cancel existing contributions, thereby relaxing the associated constraints. In other phases, the diagrams dominate and tighten the bounds, which in turn necessitates more finely tuned parameters to reconcile successful baryogenesis with current EDM limits.
|
[
{
"version": "v1",
"created": "Tue, 08 Jul 2025 18:14:16 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Jul 2025 01:53:05 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 21:23:52 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ning",
"Kaifei",
""
],
[
"Ramsey-Musolf",
"Michael",
""
]
] |
2507.06463
|
David Bader
|
Atieh Barati Nia, Mohammad Dindoost, David A. Bader
|
Evaluating Efficiency and Novelty of LLM-Generated Code for Graph Analysis
|
7 pages, v2: minor revision to match final paper published in the The 29th Annual IEEE High Performance Extreme Computing Conference (HPEC), Virtual, September 15-19, 2025. Outstanding Student Paper Award
| null | null | null |
cs.SE
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) are increasingly used to automate software development, yet most prior evaluations focus on functional correctness or high-level languages such as Python. As one of the first systematic explorations of LLM-assisted software performance engineering, we present a comprehensive study of LLMs' ability to generate efficient C implementations of graph-analysis routines -- code that must satisfy stringent runtime and memory constraints. This emerging field of LLM-assisted algorithm engineering holds significant promise, as these models may possess the capability to design novel approaches that improve existing algorithms and their implementations. Eight state-of-the-art models (OpenAI ChatGPT o3 and o4-mini-high, Anthropic Claude 4 Sonnet and Sonnet Extended, Google Gemini 2.5 Flash and Pro, xAI Grok 3-Think, and DeepSeek DeepThink R1) are benchmarked using two distinct approaches. The first approach evaluates the ability of LLMs to generate algorithms that outperform existing benchmarks. The second approach assesses their capability to generate graph algorithms for integration into performance-critical systems. The results show that Claude Sonnet 4 Extended achieves superior performance in ready-to-use code generation and efficiency, outperforming human-written baselines in triangle counting. Although our findings demonstrate that contemporary LLMs excel in optimizing and integrating established algorithms, the potential for these models to eventually invent transformative algorithmic techniques represents a compelling frontier for future research. We provide prompts, generated code, and measurement scripts to promote reproducible research in this rapidly evolving domain. All of the source code is available on GitHub at https://github.com/Bader-Research/LLM-triangle-counting/.
|
[
{
"version": "v1",
"created": "Wed, 09 Jul 2025 00:46:30 GMT"
},
{
"version": "v2",
"created": "Thu, 28 Aug 2025 14:22:23 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Nia",
"Atieh Barati",
""
],
[
"Dindoost",
"Mohammad",
""
],
[
"Bader",
"David A.",
""
]
] |
2507.06656
|
Hongjie Wu
|
Hongjie Wu, Mingqin Zhang, Linchao He, Ji-Zhe Zhou, Jiancheng Lv
|
Enhancing Diffusion Model Stability for Image Restoration via Gradient Management
|
Accepted to ACM Multimedia 2025
| null |
10.1145/3746027.3754582
| null |
cs.CV cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Diffusion models have shown remarkable promise for image restoration by leveraging powerful priors. Prominent methods typically frame the restoration problem within a Bayesian inference framework, which iteratively combines a denoising step with a likelihood guidance step. However, the interactions between these two components in the generation process remain underexplored. In this paper, we analyze the underlying gradient dynamics of these components and identify significant instabilities. Specifically, we demonstrate conflicts between the prior and likelihood gradient directions, alongside temporal fluctuations in the likelihood gradient itself. We show that these instabilities disrupt the generative process and compromise restoration performance. To address these issues, we propose Stabilized Progressive Gradient Diffusion (SPGD), a novel gradient management technique. SPGD integrates two synergistic components: (1) a progressive likelihood warm-up strategy to mitigate gradient conflicts; and (2) adaptive directional momentum (ADM) smoothing to reduce fluctuations in the likelihood gradient. Extensive experiments across diverse restoration tasks demonstrate that SPGD significantly enhances generation stability, leading to state-of-the-art performance in quantitative metrics and visually superior results. Code is available at https://github.com/74587887/SPGD.
|
[
{
"version": "v1",
"created": "Wed, 09 Jul 2025 08:40:46 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 05:42:46 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wu",
"Hongjie",
""
],
[
"Zhang",
"Mingqin",
""
],
[
"He",
"Linchao",
""
],
[
"Zhou",
"Ji-Zhe",
""
],
[
"Lv",
"Jiancheng",
""
]
] |
2507.08406
|
Yijia Zhang
|
Weigang Feng, Yijia Zhang, Zekun Wang, Zhengyang Wang, Yi Wang, Peijun Ma, Ningyi Xu
|
CCSS: Hardware-Accelerated RTL Simulation with Fast Combinational Logic Computing and Sequential Logic Synchronization
|
We plan to add more experiments and refine the figures in the paper. In addition, the overall structure needs significant revision to improve its readability
| null | null | null |
cs.AR cs.DC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As transistor counts in a single chip exceed tens of billions, the complexity of RTL-level simulation and verification has grown exponentially, often extending simulation campaigns to several months. In industry practice, RTL simulation is divided into two phases: functional debug and system validation. While system validation demands high simulation speed and is typically accelerated using FPGAs, functional debug relies on rapid compilation-rendering multi-core CPUs the primary choice. However, the limited simulation speed of CPUs has become a major bottleneck. To address this challenge, we propose CCSS, a scalable multi-core RTL simulation platform that achieves both fast compilation and high simulation throughput. CCSS accelerates combinational logic computation and sequential logic synchronization through specialized architecture and compilation strategies. It employs a balanced DAG partitioning method and efficient boolean computation cores for combinational logic, and adopts a low-latency network-on-chip (NoC) design to synchronize sequential states across cores efficiently. Experimental results show that CCSS delivers up to 12.9x speedup over state-of-the-art multi-core simulators.
|
[
{
"version": "v1",
"created": "Fri, 11 Jul 2025 08:27:32 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 05:58:08 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Feng",
"Weigang",
""
],
[
"Zhang",
"Yijia",
""
],
[
"Wang",
"Zekun",
""
],
[
"Wang",
"Zhengyang",
""
],
[
"Wang",
"Yi",
""
],
[
"Ma",
"Peijun",
""
],
[
"Xu",
"Ningyi",
""
]
] |
2507.08414
|
Sergei Ivanov Olegovich
|
Emmanuel Dror Farjoun, Sergei O. Ivanov
|
Bousfield-Kan completion as a codensity $\infty$-monad
| null | null | null | null |
math.AT
|
http://creativecommons.org/publicdomain/zero/1.0/
|
We develop a theory of codensity monads associated with full subcategories in the setting of $\infty$-categories. For an $\infty$-monad $\mathcal{M}$ on an $\infty$-category $\mathcal{C}$ that admits totalizations of cosimplicial objects, we consider the $\mathcal{M}$-completion functor defined as the totalization of the canonical cosimplicial functor associated with $\mathcal{M}$. We prove that the $\mathcal{M}$-completion functor is the codensity $\infty$-monad of a full $\infty$-subcategory $\mathcal{A}(\mathcal{M})\subseteq \mathcal{C}$ spanned by objects that admit a structure of $\mathcal{M}$-algebra. This implies that the $\mathcal{M}$-completion functor is the terminal object in the $\infty$-category of $\infty$-monads preserving all objects of $\mathcal{A}(\mathcal{M})$. As an application, we prove that the classical Bousfield-Kan $R$-completion functor is the codensity $\infty$-monad of the full subcategory in the $\infty$-category of spaces $\mathcal{K}(R) \subseteq \mathsf{Spc}$ spanned by the empty space and the products of Eilenberg-MacLane spaces of $R$-modules. As a corollary, we obtain two universal properties of the Bousfield-Kan $R$-completion: it is the terminal $\mathcal{K}(R)$-preserving coaugmented functor, and the terminal $\mathcal{K}(R)$-preserving $\infty$-monad.
|
[
{
"version": "v1",
"created": "Fri, 11 Jul 2025 08:54:57 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 08:03:44 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Farjoun",
"Emmanuel Dror",
""
],
[
"Ivanov",
"Sergei O.",
""
]
] |
2507.09879
|
Pooja Kulkarni
|
Tanvi Bajpai, Chandra Chekuri, Pooja Kulkarni
|
Covering a Few Submodular Constraints and Applications
|
34 pages. Accepted to APPROX 2025
| null | null | null |
cs.DS cs.AI cs.GT
|
http://creativecommons.org/licenses/by/4.0/
|
We consider the problem of covering multiple submodular constraints. Given a finite ground set $N$, a cost function $c: N \rightarrow \mathbb{R}_+$, $r$ monotone submodular functions $f_1,f_2,\ldots,f_r$ over $N$ and requirements $b_1,b_2,\ldots,b_r$ the goal is to find a minimum cost subset $S \subseteq N$ such that $f_i(S) \ge b_i$ for $1 \le i \le r$. When $r=1$ this is the well-known Submodular Set Cover problem. Previous work \cite{chekuri2022covering} considered the setting when $r$ is large and developed bi-criteria approximation algorithms, and approximation algorithms for the important special case when each $f_i$ is a weighted coverage function. These are fairly general models and capture several concrete and interesting problems as special cases. The approximation ratios for these problem are at least $\Omega(\log r)$ which is unavoidable when $r$ is part of the input. In this paper, motivated by some recent applications, we consider the problem when $r$ is a \emph{fixed constant} and obtain two main results. For covering multiple submodular constraints we obtain a randomized bi-criteria approximation algorithm that for any given integer $\alpha \ge 1$ outputs a set $S$ such that $f_i(S) \ge$ $(1-1/e^\alpha -\epsilon)b_i$ for each $i \in [r]$ and $\mathbb{E}[c(S)] \le (1+\epsilon)\alpha \cdot \sf{OPT}$. Second, when the $f_i$ are weighted coverage functions from a deletion-closed set system we obtain a $(1+\epsilon)$ $(\frac{e}{e-1})$ $(1+\beta)$-approximation where $\beta$ is the approximation ratio for the underlying set cover instances via the natural LP. These results show that one can obtain nearly as good an approximation for any fixed $r$ as what one would achieve for $r=1$. We mention some applications that follow easily from these general results and anticipate more in the future.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 03:32:42 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 15:05:05 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bajpai",
"Tanvi",
""
],
[
"Chekuri",
"Chandra",
""
],
[
"Kulkarni",
"Pooja",
""
]
] |
2507.10383
|
Uri Cohen
|
Uri Cohen, M\'at\'e Lengyel
|
Dynamical stability for dense patterns in discrete attractor neural networks
| null | null | null | null |
cond-mat.dis-nn cond-mat.stat-mech cs.LG cs.NE q-bio.NC
|
http://creativecommons.org/licenses/by/4.0/
|
Neural networks storing multiple discrete attractors are canonical models of biological memory. Previously, the dynamical stability of such networks could only be guaranteed under highly restrictive conditions. Here, we derive a theory of the local stability of discrete fixed points in a broad class of networks with graded neural activities and in the presence of noise. By directly analyzing the bulk and outliers of the Jacobian spectrum, we show that all fixed points are stable below a critical load that is distinct from the classical \textit{critical capacity} and depends on the statistics of neural activities in the fixed points as well as the single-neuron activation function. Our analysis highlights the computational benefits of threshold-linear activation and sparse-like patterns.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 15:23:24 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 11:39:43 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Cohen",
"Uri",
""
],
[
"Lengyel",
"Máté",
""
]
] |
2507.10414
|
Zhiwei Zheng
|
Zhiwei Zheng
|
A Lemma on Leech-like Lattices
|
7 pages, comments welcome
| null | null | null |
math.AG math.GR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A Leech pair is defined as a pair $(G,S)$, where $S$ is a positive definite even lattice without roots, equipped with a faithful action of a finite group $G$, such that the invariant sublattice of $S$ under the action of $G$ is trivial, and the induced action of $G$ on the discriminant group of $S$ is also trivial. This structure appears naturally when investigating hyperk\"ahler manifolds and the symplectic automorphisms acting on them. An important lemma due to Gaberdiel--Hohenegger--Volpato asserts that a Leech pair $(G,S)$ admits a primitive embedding into the Leech lattice if $rank(S)+\ell(A_S)\le 24$. However, the original proof is incomplete, as demonstrated by a counterexample provided by Marquand and Muller. They also presented a computer-assisted proof of the lemma for cases where $rank(S) \le 21$. In this paper, we modify the original approach to provide a complete and conceptual proof of the lemma.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 15:57:00 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 05:56:26 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zheng",
"Zhiwei",
""
]
] |
2507.10578
|
Jeremy Styborski
|
Jeremy Styborski, Mingzhi Lyu, Jiayou Lu, Nupur Kapur, Adams Kong
|
When and Where do Data Poisons Attack Textual Inversion?
|
Accepted to ICCV 2025
| null | null | null |
cs.CR cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Poisoning attacks pose significant challenges to the robustness of diffusion models (DMs). In this paper, we systematically analyze when and where poisoning attacks textual inversion (TI), a widely used personalization technique for DMs. We first introduce Semantic Sensitivity Maps, a novel method for visualizing the influence of poisoning on text embeddings. Second, we identify and experimentally verify that DMs exhibit non-uniform learning behavior across timesteps, focusing on lower-noise samples. Poisoning attacks inherit this bias and inject adversarial signals predominantly at lower timesteps. Lastly, we observe that adversarial signals distract learning away from relevant concept regions within training data, corrupting the TI process. Based on these insights, we propose Safe-Zone Training (SZT), a novel defense mechanism comprised of 3 key components: (1) JPEG compression to weaken high-frequency poison signals, (2) restriction to high timesteps during TI training to avoid adversarial signals at lower timesteps, and (3) loss masking to constrain learning to relevant regions. Extensive experiments across multiple poisoning methods demonstrate that SZT greatly enhances the robustness of TI against all poisoning attacks, improving generative quality beyond prior published defenses. Code: www.github.com/JStyborski/Diff_Lab Data: www.github.com/JStyborski/NC10
|
[
{
"version": "v1",
"created": "Fri, 11 Jul 2025 10:35:52 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Jul 2025 04:27:33 GMT"
},
{
"version": "v3",
"created": "Mon, 28 Jul 2025 05:07:59 GMT"
},
{
"version": "v4",
"created": "Wed, 03 Sep 2025 05:48:05 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Styborski",
"Jeremy",
""
],
[
"Lyu",
"Mingzhi",
""
],
[
"Lu",
"Jiayou",
""
],
[
"Kapur",
"Nupur",
""
],
[
"Kong",
"Adams",
""
]
] |
2507.10660
|
Navdeep Singh Dhindsa
|
Navdeep Singh Dhindsa, Nilmani Mathur, M. Padmanath
|
Lattice QCD Study of Positive Parity Dibaryons with Maximal Charm and Strangeness
|
v2: 15 pages, 8 tables, 10 main text figures, and 2 appendices figures. No change in results and version consistent with Published PRD version
|
Phys. Rev. D 112, 054501 (2025)
|
10.1103/gnkg-8lyz
|
TIFR/TH/25-14
|
hep-lat hep-ex hep-ph nucl-th
|
http://creativecommons.org/licenses/by/4.0/
|
We present the ground-state energy spectra of dibaryons composed of single-flavor quarks, specifically systems with strangeness $\mathcal{S} = -6$ and charm $\mathcal{C} = 6$. Our lattice QCD study is based on $N_f=2+1+1$ MILC ensembles with highly improved staggered quark (HISQ) sea quarks, spanning four lattice spacings and two spatial volumes. We employ valence quark propagators realized using a relativistic overlap action, evaluate correlation matrices with carefully designed operator bases, and extract reliable ground-state energy estimates in the $S = 0$ and $S = 2$ spin channels. We explore their binding characteristics and interaction dynamics by examining the energy separation between the dibaryon states and the corresponding two-baryon thresholds. These results contribute to a deeper understanding of single-flavor dibaryon states as a function of the quark masses. In the $S=0$ channel, the $\Omega_{ccc}$-$\Omega_{ccc}$ system exhibits a clear signal of a bound state, while the $\Omega$-$\Omega$ system lies very close to the threshold, making it difficult to draw definitive conclusions. For $S=2$, both systems are found to be unbound.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 18:00:02 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 06:15:28 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Dhindsa",
"Navdeep Singh",
""
],
[
"Mathur",
"Nilmani",
""
],
[
"Padmanath",
"M.",
""
]
] |
2507.10863
|
Hesam Arabzadeh
|
Hesam Arabzadeh, Carol Griswold Hoover, William Graham Hoover, Brad Lee Holian
|
Chaos in Nonequilibrium Two-Temperature $(T_x, T_y)$ Nos\'e-Hoover Cell Models
| null | null | null | null |
cond-mat.stat-mech nlin.CD
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We revisit a two-temperature Nos\'e-Hoover wanderer particle embedded in a two-dimensional periodic 2x2 cell with four smooth repulsive corners at $(x,y) = (\pm 1, \pm 1)$ to explore chaos with anisotropic thermostatting. The model employs separate thermostats in the x and y directions, enabling controlled deviations from equilibrium. By integrating the full six-dimensional equations of motion and computing the complete Lyapunov spectrum, we confirm chaos and quantify phase-space contraction with high numerical precision. The total contraction rate, interpreted as entropy production, increases nonlinearly with the thermostat anisotropy, deviating from the quadratic dependence expected from linear-response theory, $\Lambda\propto -\delta^{2}$. We compare two fits for $\Lambda$ as a function of $\delta = 0.5 -T_y$: 1) a power law, $\Lambda\propto -\delta^{2.44}$, 2) a quadratic-plus-quartic expansion. While the former captures low-driving behavior slightly better, the latter more accurately describes the strongly driven regime and remains consistent with linear response theory near equilibrium. An empirical linear relation between dissipation and phase-space dimensionality loss is also identified, $\Lambda\propto (D_{KY}-6) / 3$, where $D_{KY}$ is the approximate Kaplan-Yorke dimension. Our results demonstrate that nonlinear dissipation scaling emerges naturally even in minimal driven systems. Momentum statistics show significant non-Gaussian behavior under strong driving. Despite its dissipative nature, the model remains strictly time-reversible, offering a pedagogically rich example of microscopic reversibility coexisting with macroscopic entropy production.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 23:36:16 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Jul 2025 15:30:27 GMT"
},
{
"version": "v3",
"created": "Fri, 18 Jul 2025 16:27:08 GMT"
},
{
"version": "v4",
"created": "Tue, 05 Aug 2025 17:55:15 GMT"
},
{
"version": "v5",
"created": "Fri, 08 Aug 2025 17:49:09 GMT"
},
{
"version": "v6",
"created": "Sun, 17 Aug 2025 05:18:13 GMT"
},
{
"version": "v7",
"created": "Tue, 02 Sep 2025 19:30:45 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Arabzadeh",
"Hesam",
""
],
[
"Hoover",
"Carol Griswold",
""
],
[
"Hoover",
"William Graham",
""
],
[
"Holian",
"Brad Lee",
""
]
] |
2507.11066
|
Tianqi Cang
|
Xueying Hu, Tianqi Cang, Jian-Ning Fu, Xuan Wang, Keyu Xing, Haotian Wang, Pascal Petit, Jiaxin Wang, Yong Yang, He Zhao
|
Inactive longitude and superflare in the active single-lined pre-main sequence binary V2279 Cyg
|
23 pages, 16 figures. Accepted by ApJ
|
ApJ 990 27 (2025)
|
10.3847/1538-4357/adeb69
| null |
astro-ph.SR
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Young, solar-like stars in the pre-main sequence (PMS) stage exhibit vigorous magnetic activity that significantly influences their circumstellar environments and the processes of planetary formation and evolution. In binary systems, tidal forces and magnetic interactions can further shape the magnetic geometry. We report a longitudinal preference of star spots, chromospheric activities, and flares in the active single-lined spectroscopic PMS binary system V2279 Cyg, based on long-term photometric observations from \textit{Kepler} and \textit{TESS} alongside spectroscopic data from LAMOST. The system is classified as a weak-line T Tauri binary, with component masses estimated at 0.86 $M_\odot$ and 0.27 $M_\odot$. V2279 Cyg's nearly circular orbit is synchronized with its 4.126-day rotational period. Observations reveal large star spot regions clustered near the far-side hemisphere. Spectroscopic data show strong, double-peak H$\alpha$ emission, the strength of which is highly correlated with star spot distribution, indicating the presence of an active longitude on the primary star. We also mapped the prominence structure co-rotating with the primary star, suggesting a dense structure close to the near-side hemisphere. Furthermore, we identify an inactive longitude of flares during the 4-year \textit{Kepler} observations, where the frequency of flare activity is significantly reduced after the superior conjunction, marking the first such identification in active binary systems. Additionally, a white light superflare, releasing energy of $2.5 \times 10^{37}$ erg, was detected in \textit{TESS} observations. These findings provide valuable insights into the magnetic field geometry and dynamo processes in PMS binaries, underscoring the critical role of tidal interactions in shaping magnetic activities.
|
[
{
"version": "v1",
"created": "Tue, 15 Jul 2025 07:59:57 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Hu",
"Xueying",
""
],
[
"Cang",
"Tianqi",
""
],
[
"Fu",
"Jian-Ning",
""
],
[
"Wang",
"Xuan",
""
],
[
"Xing",
"Keyu",
""
],
[
"Wang",
"Haotian",
""
],
[
"Petit",
"Pascal",
""
],
[
"Wang",
"Jiaxin",
""
],
[
"Yang",
"Yong",
""
],
[
"Zhao",
"He",
""
]
] |
2507.11110
|
Lin Lan
|
Lin Lan, He Gao, Shunke Ai, Wen-Jin Xie, Yong Yuan, Long Li, Li-Ping Xin, Jian-Yan Wei
|
Constrain magnetar parameters by taking into account the evolutionary effects of radius and moment of inertia with \emph{Swift}/XRT data
|
68 pages, 27 figures, and 14 tables, accepted for publication in ApJS, and matched with the published verison
| null | null | null |
astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A newly born millisecond magnetar has been proposed as one possible central engine of some GRBs with X-ray plateau emission. In this work, we systematically analyzed the Swift/XRT data of long GRBs with plateau emission that were detected before 2023 December, and estimated the physical parameters by considering the $R/I$ evolutionary effects. We found that neglecting the $R/I$ evolutionary effects can lead to systematic overestimation or underestimation of magnetar parameters such as $B_p$, $P_0$, and $\epsilon$ from 20\% to 50\%. We also found that some tight correlations, which can be approximately expressed as $\epsilon\propto P_0^{1.57\pm0.22}$, $\epsilon\propto B_p^{0.97\pm0.13}$, $B_p\propto P_0^{1.30\pm0.16}$, $E_{\rm wind}\propto E_{\rm jet,iso}^{0.83\pm0.07}(E_{\rm jet}^{0.76\pm0.06})$, $P_0\propto E_{\rm jet,iso}^{-0.29\pm0.03}(E_{\rm jet}^{-0.26\pm0.02})$, $B_p\propto E_{\rm jet,iso}^{-0.58\pm0.06}(E_{\rm jet}^{-0.55\pm0.05})$, and $\epsilon\propto E_{\rm jet,iso}^{-0.55\pm0.07}(E_{\rm jet}^{-0.52\pm0.06})$ for our selected EoSs. The universal correlations suggest that a nascent magnetar with the faster $P_0$, lower $B_p$, and lower $\epsilon$ are more inclined to power a more energetic GRB jet, and the $\epsilon$ and $P_0$ of newborn magnetar are likely to originate from the magnetically induced distortion and correspond to the equilibrium spin period as a result of interaction between the magnetar and its accretion disk, respectively. Finally, we found that the GW signals from the remnants of those GW-dominated GRBs with redshift measurements cannot reach aLIGO sensitivity threshold, and only two cases (GRBs 150323A and 170607A) can reach ET sensitivity threshold. Future GW observations could not only offer the first smoking gun that a protomagnetar can serve as the central engine of GRBs but also play a crucial role in precisely constraining the neutron star EoS.
|
[
{
"version": "v1",
"created": "Tue, 15 Jul 2025 09:03:30 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 03:05:41 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Lan",
"Lin",
""
],
[
"Gao",
"He",
""
],
[
"Ai",
"Shunke",
""
],
[
"Xie",
"Wen-Jin",
""
],
[
"Yuan",
"Yong",
""
],
[
"Li",
"Long",
""
],
[
"Xin",
"Li-Ping",
""
],
[
"Wei",
"Jian-Yan",
""
]
] |
2507.12161
|
Vladimir Dergachev Ph.D.
|
Vladimir Dergachev and Maria Alessandra Papa
|
Early release of low-frequency atlas of continuous gravitational waves
|
Added data from new injection study of all-sky population average upper limit proxy
| null | null | null |
gr-qc astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present the public release of the low-frequency atlas of continuous gravitational waves, covering signals with frequencies from 20 Hz to 200 Hz and frequency derivatives from -5e-11 to 5e-11 Hz/s. Compared to the previous atlas releases, this version demonstrates significant improvements in sensitivity and sky resolution. In the most sensitive region even the worst-case upper limits on gravitational wave strain are below 1e-25. The atlas data is being released ahead of the completion of the full follow-up analysis.
|
[
{
"version": "v1",
"created": "Wed, 16 Jul 2025 11:41:11 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 15:48:45 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Dergachev",
"Vladimir",
""
],
[
"Papa",
"Maria Alessandra",
""
]
] |
2507.13429
|
Marie Cornelius
|
Marie Cornelius (1), Irene Tamborra (1), Malte Heinlein (2,3), Shashank Shalgar (1), Hans-Thomas Janka (2) ((1) Niels Bohr Institute, (2) MPI Astrophysics, (3) TUM)
|
Electron-neutrino lepton number crossings: Variations with the supernova core physics
|
14 pages, including 8 figures and 1 appendix. Minor changes. Matches version accepted for publication in Phys. Rev. D
|
Phys. Rev. D 112 (2025) 6, 063006
|
10.1103/n2h5-3qwz
| null |
astro-ph.HE hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
A crucial ingredient affecting fast neutrino flavor conversion in core-collapse supernovae (SNe) is the shape of the angular distribution of the electron-neutrino lepton number (ELN). The presence of an ELN crossing signals favorable conditions for flavor conversion. However, the dependence of ELN crossings on the SN properties is only partially understood. We investigate a suite of 12 spherically symmetric neutrino-hydrodynamics simulations of the core collapse of a SN with a mass of $18.6 M_\odot$; each model employs different microphysics (i.e., three different nuclear equations of state, with and without muon creation) and includes or not a mixing-length treatment for proto-neutron star convection. We solve the Boltzmann equations to compute the neutrino angular distributions relying on static fluid properties extracted from each of the SN simulations in our suite for six selected post-bounce times. We explore the dependence of the ELN distributions on the SN microphysics and proto-neutron star convection. We find that the latter shifts the proto-neutron star radius outwards, favoring the appearance of ELN crossings at larger radii. On the other hand, muon creation causes proto-neutron star contraction, facilitating the occurrence of ELN crossings at smaller radii. These effects mildly depend on the nuclear equation of state. Our findings highlight the subtle impact of the SN microphysics, proto-neutron star convection, and neutrino transport on the ELN angular distributions.
|
[
{
"version": "v1",
"created": "Thu, 17 Jul 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 15:30:02 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Cornelius",
"Marie",
"",
"Niels Bohr Institute"
],
[
"Tamborra",
"Irene",
"",
"Niels Bohr Institute"
],
[
"Heinlein",
"Malte",
"",
"MPI Astrophysics",
"TUM"
],
[
"Shalgar",
"Shashank",
"",
"Niels Bohr Institute"
],
[
"Janka",
"Hans-Thomas",
"",
"MPI Astrophysics"
]
] |
2507.14048
|
Elisa Fazzari
|
Giovanni Montani, Elisa Fazzari, Nakia Carlevaro, and Maria Giovanna Dainotti
|
Two Dynamical Scenarios for Binned Master Sample Interpretation
|
16 pages, 2 figures; references added; matches the published version
|
Entropy 2025, 27, 895
|
10.3390/e27090895
| null |
astro-ph.CO gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We analyze two different scenarios for the late Universe dynamics, resulting into Hubble parameters deviating from the $\Lambda$CDM, mainly for the presence of an additional free parameter, which is the dark energy parameter. The first model consists of a pure evolutionary dark energy paradigm, as result of its creation by the gravitational field of the expanding Universe. The second model also considers an interaction of the evolutionary dark energy with the matter component, postulated via the conservation of the sum of their ideal energy-momentum tensors. These two models are then compared \textit{via} the diagnostic tool of the effective running Hubble constant, with the binned data of the so-called ``Master sample'' for the Type Ia Supernovae. The comparison procedures, based on a standard MCMC analysis, led to a clear preference of data for the dark energy - matter interaction model, which is associated to a phantom matter equation of state parameter (very close to $-1$) when, being left free by data (it has a flat posterior), it is fixed in order to reproduce the decreasing power-law behavior of the effective running Hubble constant, already discussed in literature.
|
[
{
"version": "v1",
"created": "Fri, 18 Jul 2025 16:14:28 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 09:55:04 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Montani",
"Giovanni",
""
],
[
"Fazzari",
"Elisa",
""
],
[
"Carlevaro",
"Nakia",
""
],
[
"Dainotti",
"Maria Giovanna",
""
]
] |
2507.15976
|
Aaron Ho
|
Aaron Ho (1), Lorenzo Zanisi (2), Bram de Leeuw (3), Vincent Galvan (1), Pablo Rodriguez-Fernandez (1), Nathaniel T. Howard (1) ((1) MIT Plasma Science and Fusion Center, Cambridge, USA, (2) UKAEA Culham Centre for Fusion Energy, Abingdon, UK, (3) Radboud University, Nijmegen, Netherlands)
|
Efficient dataset construction using active learning and uncertainty-aware neural networks for plasma turbulent transport surrogate models
| null | null |
10.1063/5.0276455
| null |
physics.plasm-ph cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
This work demonstrates a proof-of-principle for using uncertainty-aware architectures, in combination with active learning techniques and an in-the-loop physics simulation code as a data labeller, to construct efficient datasets for data-driven surrogate model generation. Building off of a previous proof-of-principle successfully demonstrating training set reduction on static pre-labelled datasets, using the ADEPT framework, this strategy was applied again to the plasma turbulent transport problem within tokamak fusion plasmas, specifically the QuaLiKiz quasilinear electrostatic gyrokinetic turbulent transport code. While QuaLiKiz provides relatively fast evaluations, this study specifically targeted small datasets to serve as a proxy for more expensive codes, such as CGYRO or GENE. The newly implemented algorithm uses the SNGP architecture for the classification component of the problem and the BNN-NCP architecture for the regression component, training models for all turbulent modes (ITG, TEM, ETG) and all transport fluxes ($Q_e$, $Q_i$, $\Gamma_e$, $\Gamma_i$, and $\Pi_i$) described by the general QuaLiKiz output. With 45 active learning iterations, moving from a small initial training set of $10^{2}$ to a final set of $10^{4}$, the resulting models reached a $F_1$ classification performance of ~0.8 and a $R^2$ regression performance of ~0.75 on an independent test set across all outputs. This extrapolates to reaching the same performance and efficiency as the previous ADEPT pipeline, although on a problem with 1 extra input dimension. While the improvement rate achieved in this implementation diminishes faster than expected, the overall technique is formulated with components that can be upgraded and generalized to many surrogate modeling applications beyond plasma turbulent transport predictions.
|
[
{
"version": "v1",
"created": "Mon, 21 Jul 2025 18:15:12 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ho",
"Aaron",
"",
"MIT Plasma Science and Fusion Center, Cambridge, USA"
],
[
"Zanisi",
"Lorenzo",
"",
"UKAEA Culham Centre for Fusion Energy, Abingdon, UK"
],
[
"de Leeuw",
"Bram",
"",
"Radboud University, Nijmegen, Netherlands"
],
[
"Galvan",
"Vincent",
"",
"MIT Plasma Science and Fusion Center, Cambridge, USA"
],
[
"Rodriguez-Fernandez",
"Pablo",
"",
"MIT Plasma Science and Fusion Center, Cambridge, USA"
],
[
"Howard",
"Nathaniel T.",
"",
"MIT Plasma Science and Fusion Center, Cambridge, USA"
]
] |
2507.16687
|
Lennart Van Sluijs
|
Lennart van Sluijs, Hayley Beltz, Isaac Malsky, Genevieve H. Pereira, L. Cinque, Emily Rauscher, and Jayne Birkby
|
Assessing robustness and bias in 1D retrievals of 3D Global Circulation Models at high spectral resolution: a WASP-76 b simulation case study in emission
|
24 pages, 11 figures, accepted to ApJ
|
ApJ 990 106 (2025)
|
10.3847/1538-4357/adf328
| null |
astro-ph.EP
|
http://creativecommons.org/licenses/by/4.0/
|
High-resolution spectroscopy (HRS) of exoplanet atmospheres has successfully detected many chemical species and is quickly moving toward detailed characterization of the chemical abundances and dynamics. HRS is highly sensitive to the line shape and position, thus, it can detect three-dimensional (3D) effects such as winds, rotation, and spatial variation of atmospheric conditions. At the same time, retrieval frameworks are increasingly deployed to constrain chemical abundances, pressure-temperature (P-T) structures, orbital parameters, and rotational broadening. To explore the multidimensional parameter space, they need computationally fast models that are consequently mostly one-dimensional (1D). However, this approach risks introducing interpretation bias since the planet's true nature is 3D. We investigate the robustness of this methodology at high spectral resolution by running 1D retrievals on simulated observations in emission within an observational framework using 3D Global Circulation Models of the quintessential HJ WASP-76 b. We find that the retrieval broadly recovers conditions present in the atmosphere, but that the retrieved P-T and chemical profiles are not a homogeneous average of all spatial and phase-dependent information. Instead, they are most sensitive to spatial regions with large thermal gradients, which do not necessarily coincide with the strongest emitting regions. Our results further suggest that the choice of parameterization for the P-T and chemical profiles, as well as Doppler offsets among opacity sources, impact retrieval results. These factors should be carefully considered in future retrieval analyses.
|
[
{
"version": "v1",
"created": "Tue, 22 Jul 2025 15:19:56 GMT"
},
{
"version": "v2",
"created": "Wed, 23 Jul 2025 15:06:45 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"van Sluijs",
"Lennart",
""
],
[
"Beltz",
"Hayley",
""
],
[
"Malsky",
"Isaac",
""
],
[
"Pereira",
"Genevieve H.",
""
],
[
"Cinque",
"L.",
""
],
[
"Rauscher",
"Emily",
""
],
[
"Birkby",
"Jayne",
""
]
] |
2507.17621
|
Alexander Bershadskii
|
A. Bershadskii
|
Chaotic/turbulent cross-helical MHD dynamo
|
Extended (new data have been added)
| null | null | null |
physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Using results of laboratory experiments and direct numerical simulations, it is shown that cross-helicity can dominate power spectra of the magnetic field generated by magnetohydrodynamic (MHD) dynamo in a chaotic/turbulent von K\'{a}rm\'{a}n flow and its numerical analogue (simulation). The theoretical consideration is based on a Kolmogorov-like phenomenology within the framework of the distributed chaos concept. The observations of the dynamics of the solar magnetic field (solar activity) have been briefly discussed in this context.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 15:52:39 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Aug 2025 16:16:47 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 15:13:00 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bershadskii",
"A.",
""
]
] |
2507.17686
|
Takashi Hayakawa Dr.
|
Takashi Hayakawa and Satoshi Asai
|
Debiased maximum-likelihood estimators for hazard ratios under kernel-based machine-learning adjustment
|
Proposition 3 of the first version was wrong. This was fixed by introducing new theoretical results in the second version so that all of the claims of the first version are valid For some reason, in the uploading of the second version, a duplicate of the first version was wrongly publicized, which we think is not our mistake. We therefore upload the updated version as version 3
| null | null | null |
stat.ML cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Previous studies have shown that hazard ratios between treatment groups estimated with the Cox model are uninterpretable because the unspecified baseline hazard of the model fails to identify temporal change in the risk set composition due to treatment assignment and unobserved factors among multiple, contradictory scenarios. To alleviate this problem, especially in studies based on observational data with uncontrolled dynamic treatment and real-time measurement of many covariates, we propose abandoning the baseline hazard and using kernel-based machine learning to explicitly model the change in the risk set with or without latent variables. For this framework, we clarify the context in which hazard ratios can be causally interpreted, and then develop a method based on Neyman orthogonality to compute debiased maximum-likelihood estimators of hazard ratios, proving necessary convergence results. Numerical simulations confirm that the proposed method identifies the true hazard ratios with minimal bias. These results lay the foundation for developing a useful, alternative method for causal inference with uncontrolled, observational data in modern epidemiology.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 16:51:09 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 22:49:14 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 03:13:44 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Hayakawa",
"Takashi",
""
],
[
"Asai",
"Satoshi",
""
]
] |
2507.17695
|
Ilias Chatzistefanidis
|
Ilias Chatzistefanidis, Navid Nikaein
|
Symbiotic Agents: A Novel Paradigm for Trustworthy AGI-driven Networks
|
Submitted to Computer Networks AI for 6G
| null | null | null |
cs.AI cs.NI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large Language Model (LLM)-based autonomous agents are expected to play a vital role in the evolution of 6G networks, by empowering real-time decision-making related to management and service provisioning to end-users. This shift facilitates the transition from a specialized intelligence approach, where artificial intelligence (AI) algorithms handle isolated tasks, to artificial general intelligence (AGI)-driven networks, where agents possess broader reasoning capabilities and can manage diverse network functions. In this paper, we introduce a novel agentic paradigm that combines LLMs with real-time optimization algorithms towards Trustworthy AI, defined as symbiotic agents. Optimizers at the LLM's input-level provide bounded uncertainty steering for numerically precise tasks, whereas output-level optimizers supervised by the LLM enable adaptive real-time control. We design and implement two novel agent types including: (i) Radio Access Network optimizers, and (ii) multi-agent negotiators for Service-Level Agreements (SLAs). We further propose an end-to-end architecture for AGI networks and evaluate it on a 5G testbed capturing channel fluctuations from moving vehicles. Results show that symbiotic agents reduce decision errors fivefold compared to standalone LLM-based agents, while smaller language models (SLM) achieve similar accuracy with a 99.9% reduction in GPU resource overhead and in near-real-time loops of 82 ms. A multi-agent demonstration for collaborative RAN on the real-world testbed highlights significant flexibility in service-level agreement and resource allocation, reducing RAN over-utilization by approximately 44%. Drawing on our findings and open-source implementations, we introduce the symbiotic paradigm as the foundation for next-generation, AGI-driven networks-systems designed to remain adaptable, efficient, and trustworthy even as LLMs advance.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 17:01:23 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 09:49:56 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Chatzistefanidis",
"Ilias",
""
],
[
"Nikaein",
"Navid",
""
]
] |
2507.17736
|
Shreya Meel Ms
|
Shreya Meel and Sennur Ulukus
|
Symmetric Private Information Retrieval (SPIR) on Graph-Based Replicated Systems
| null | null | null | null |
cs.IT cs.CR cs.DB cs.NI eess.SP math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce the problem of symmetric private information retrieval (SPIR) on replicated databases modeled by a simple graph. In this model, each vertex corresponds to a server, and a message is replicated on two servers if and only if there is an edge between them. We consider the setting where the server-side common randomness necessary to accomplish SPIR is also replicated at the servers according to the graph, and we call this as message-specific common randomness. In this setting, we establish a lower bound on the SPIR capacity, i.e., the maximum download rate, for general graphs, by proposing an achievable SPIR scheme. Next, we prove that, for any SPIR scheme to be feasible, the minimum size of message-specific randomness should be equal to the size of a message. Finally, by providing matching upper bounds, we derive the exact SPIR capacity for the class of path and regular graphs.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 17:51:08 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 16:08:01 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Meel",
"Shreya",
""
],
[
"Ulukus",
"Sennur",
""
]
] |
2507.18137
|
Hiroyasu Satoh
|
Hiroyasu Satoh and Hemangi Madhusudan Shah
|
Conformal Vector Fields on Damek-Ricci Space
|
18 pages
| null | null | null |
math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this article, we show that every conformal vector field on a Damek-Ricci space is necessarily a Killing vector field; in other words, there exists no non-trivial conformal vector field on such spaces. This result generalizes the corresponding theorem for complex hyperbolic spaces established in arXiv:2506.09710.
|
[
{
"version": "v1",
"created": "Thu, 24 Jul 2025 06:57:20 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 23:29:16 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Satoh",
"Hiroyasu",
""
],
[
"Shah",
"Hemangi Madhusudan",
""
]
] |
2507.19191
|
Fernando Camacho-Cadena
|
Fernando Camacho-Cadena
|
Self-intersecting curves on a pair of pants and periodic orbits of Hamiltonian flows
|
Updated introduction and references. Comments welcome!
| null | null | null |
math.GT math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The character variety $\mathscr{X}(S,G)$ associated to an oriented compact surface $S$ with boundary and a real reductive Lie group $G$ admits a Poisson structure and is foliated by symplectic leaves. When $G$ is a matrix group, any closed curve $c\in\pi_1(S)$ induces a trace function $\mathsf{tr}_c\colon[\rho]\mapsto \mathsf{tr}(\rho(c))$ on $\mathscr{X}(S,G)$. In this article, we study the Hamiltonian flows of trace functions associated to self-intersecting curves. We prove that when $G=\mathsf{PSL}(3,\mathbb{R})$ and $S$ is the pair of pants, every orbit of the Hamiltonian flow of the trace of a figure eight curve on $S$ is periodic and has a unique fixed point. The proof uses explicit computations in Fock-Goncharov coordinates. As an application, we prove a similar statement for the trace of the $\Theta$-web. Finally, we focus on the symplectic leaf corresponding to the unipotent locus, and derive similar results for two more self-intersecting curves: the commutator, and a curve going $k$ times around a boundary component.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 11:59:22 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 12:09:34 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Camacho-Cadena",
"Fernando",
""
]
] |
2507.19369
|
Yoav Ellinson
|
Yoav Ellinson and Sharon Gannot
|
Binaural Target Speaker Extraction using HRTFs
| null | null | null | null |
eess.AS cs.SD
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
In this work, we aim to imitate the human ability to selectively attend to a single speaker, even in the presence of multiple simultaneous talkers. To achieve this, we propose a novel approach for binaural target speaker extraction that leverages the listener's Head-Related Transfer Function (HRTF) to isolate the desired speaker. Notably, our method does not rely on speaker embeddings, making it speaker-independent and enabling strong generalization across multiple speech datasets and languages. We employ a fully complex-valued neural network that operates directly on the complex-valued Short-Time Fourier transform (STFT) of the mixed audio signals, and compare it to a Real-Imaginary (RI)-based neural network, demonstrating the advantages of the former. We first evaluate the method in an anechoic, noise-free scenario, achieving excellent extraction performance while preserving the binaural cues of the target signal. We then extend the evaluation to reverberant conditions. Our method proves robust, maintaining speech clarity and source directionality while simultaneously reducing reverberation. A comparative analysis with existing binaural Target Speaker Extraction (TSE) methods demonstrates that our approach attains performance on par with competing techniques in terms of noise reduction and perceptual quality, while offering a clear advantage in preserving binaural cues.Demo-page: https://bi-ctse-hrtf.github.io
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 15:20:49 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 06:46:45 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ellinson",
"Yoav",
""
],
[
"Gannot",
"Sharon",
""
]
] |
2507.19422
|
Marta Piscitelli
|
Marta Piscitelli, Ot\'on V\'azquez Doce
|
Adressing the p{\Omega}- interaction and di-baryonic states via femtoscopy
|
Contribution to: HADRON 2025, e-Print: 2507.19422 [nucl-th]
| null | null | null |
nucl-th nucl-ex
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Motivated by recent experimental measurements of the p$\Omega^-$ correlation function and the concurrent theoretical efforts to describe the strong interaction among hadrons in the strangeness sector, we present a data-driven approach for fine tuning of a meson-exchanges potential for the p$\Omega^-$ system. Using femtoscopy data from the ALICE and STAR collaborations, we constrain the strength of the interaction, encoded in the tunable short-range parameter introduced in the potential. The resulting model provides a good description of the measured correlation functions and favors the existence of a bound state in the \({}^5S_2\) channel with a binding energy of approximately \(0.5\,\mathrm{MeV}\). The role of the \({}^3S_1\) channel, however, remains poorly constrained due to the absence of an accurate model accounting for its inelastic contributions.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 16:41:25 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 12:25:24 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Piscitelli",
"Marta",
""
],
[
"Doce",
"Otón Vázquez",
""
]
] |
2507.19733
|
Alec Scully
|
Forrest Hare Alec Sculley, and Cameron Stockton
|
Integrating Activity Predictions in Knowledge Graphs
|
21 pages. 18 figures. Conference: Semantic Technology for Intelligence, Defense, and Security (STIDS 2024)
| null | null | null |
cs.AI cs.DB
|
http://creativecommons.org/licenses/by/4.0/
|
We argue that ontology-structured knowledge graphs can play a crucial role in generating predictions about future events. By leveraging the semantic framework provided by Basic Formal Ontology (BFO) and Common Core Ontologies (CCO), we demonstrate how data such as the movements of a fishing vessel can be organized in and retrieved from a knowledge graph. These query results are then used to create Markov chain models, allowing us to predict future states based on the vessel's history. To fully support this process, we introduce the term `spatiotemporal instant' to complete the necessary structural semantics. Additionally, we critique the prevailing ontological model of probability, according to which probabilities are about the future. We propose an alternative view, where at least some probabilities are treated as being about actual process profiles, which better captures the dynamics of real-world phenomena. Finally, we demonstrate how our Markov chain-based probability calculations can be seamlessly integrated back into the knowledge graph, enabling further analysis and decision-making.
|
[
{
"version": "v1",
"created": "Sat, 26 Jul 2025 01:22:06 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 18:49:31 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Sculley",
"Forrest Hare Alec",
""
],
[
"Stockton",
"Cameron",
""
]
] |
2507.19820
|
Chilin Zhang
|
Chilin Zhang
|
A further remark on the density estimate for degenerate Allen-Cahn equations: $\Delta_{p}$-type equations for $1<p<\frac{n}{n-1}$ with rough coefficients
| null | null | null | null |
math.AP math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this short remark on a previous paper \cite{SZ25}, we continue the study of Allen-Cahn equations associated with Ginzburg-Landau energies \begin{equation*}
J(v,\Omega)=\int_{\Omega}\Big\{F(\nabla v,v,x)+W(v,x)\Big\}dx, \end{equation*} involving a Dirichlet energy $F(\vec{\xi},\tau,x)\sim|\vec{\xi}|^{p}$ and a degenerate double-well potential $W(\tau,x)\sim(1-\tau^{2})^{m}$. In contrast to \cite{SZ25}, we remove all regularity assumptions on the Ginzburg-Landau energy. Then, with further assumptions that $1<p<\frac{n}{n-1}$ and that $W(\tau,x)$ is monotone in $\tau$ on both sides of $0$, we establish a density estimate for the level sets of nontrivial minimizers $|u|\leq1$.
|
[
{
"version": "v1",
"created": "Sat, 26 Jul 2025 06:33:39 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zhang",
"Chilin",
""
]
] |
2507.20301
|
Abdullah Alabdullah
|
Abdullah Alabdullah, Lifeng Han and Chenghua Lin
|
Advancing Dialectal Arabic to Modern Standard Arabic Machine Translation
| null | null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Dialectal Arabic (DA) poses a persistent challenge for natural language processing (NLP), as most everyday communication in the Arab world occurs in dialects that diverge significantly from Modern Standard Arabic (MSA). This linguistic divide impedes progress in Arabic machine translation. This paper presents two core contributions to advancing DA-MSA translation for the Levantine, Egyptian, and Gulf dialects, particularly in low-resource and computationally constrained settings: (i) a comprehensive evaluation of training-free prompting techniques, and (ii) the development of a resource-efficient fine-tuning pipeline. Our evaluation of prompting strategies across six large language models (LLMs) found that few-shot prompting consistently outperformed zero-shot, chain-of-thought, and our proposed Ara-TEaR method. Ara-TEaR is designed as a three-stage self-refinement prompting process, targeting frequent meaning-transfer and adaptation errors in DA-MSA translation. In this evaluation, GPT-4o achieved the highest performance across all prompting settings. For fine-tuning LLMs, a quantized Gemma2-9B model achieved a chrF++ score of 49.88, outperforming zero-shot GPT-4o (44.58). Joint multi-dialect trained models outperformed single-dialect counterparts by over 10% chrF++, and 4-bit quantization reduced memory usage by 60% with less than 1% performance loss. The results and insights of our experiments offer a practical blueprint for improving dialectal inclusion in Arabic NLP, showing that high-quality DA-MSA machine translation is achievable even with limited resources and paving the way for more inclusive language technologies.
|
[
{
"version": "v1",
"created": "Sun, 27 Jul 2025 14:37:53 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 20:29:50 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Alabdullah",
"Abdullah",
""
],
[
"Han",
"Lifeng",
""
],
[
"Lin",
"Chenghua",
""
]
] |
2507.20800
|
Vinil Polepalli
|
Vinil Polepalli
|
LanternNet: A Hub-and-Spoke System to Seek and Suppress Spotted Lanternfly Populations
| null | null | null | null |
cs.RO cs.AI cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The invasive spotted lanternfly (SLF) poses a significant threat to agriculture and ecosystems, causing widespread damage. Current control methods, such as egg scraping, pesticides, and quarantines, prove labor-intensive, environmentally hazardous, and inadequate for long-term SLF suppression. This research introduces LanternNet, a novel autonomous robotic Hub-and-Spoke system designed for scalable detection and suppression of SLF populations. A central, tree-mimicking hub utilizes a YOLOv8 computer vision model for precise SLF identification. Three specialized robotic spokes perform targeted tasks: pest neutralization, environmental monitoring, and navigation/mapping. Field deployment across multiple infested sites over 5 weeks demonstrated LanternNet's efficacy. Quantitative analysis revealed significant reductions (p < 0.01, paired t-tests) in SLF populations and corresponding improvements in tree health indicators across the majority of test sites. Compared to conventional methods, LanternNet offers substantial cost advantages and improved scalability. Furthermore, the system's adaptability for enhanced autonomy and targeting of other invasive species presents significant potential for broader ecological impact. LanternNet demonstrates the transformative potential of integrating robotics and AI for advanced invasive species management and improved environmental outcomes.
|
[
{
"version": "v1",
"created": "Mon, 28 Jul 2025 13:08:33 GMT"
},
{
"version": "v2",
"created": "Sun, 03 Aug 2025 12:10:39 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 17:09:49 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Polepalli",
"Vinil",
""
]
] |
2507.21380
|
Dylan Jow
|
Dylan L. Jow, Anna Tsai, Ue-Li Pen
|
How many pixels are there in a polarized pulsar timing array map?
| null | null | null | null |
astro-ph.IM astro-ph.CO gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
The standard approach to searching for gravitational wave signatures in pulsar timing array (PTA) data has been to compare the theoretical Hellings and Downs (HD) curve with the observed correlations in pulsar timing residuals as a function of angular separation on the sky between pulsar pairs. While this approach has successfully produced evidence for the presence of nanohertz-wavelength gravitational waves, it does not, on its own, produce any directional information. It is also insensitive to the polarization of the gravitational waves. An alternative approach is to construct maps of the gravitational wave distribution on the sky. In this paper, we present a simple quadratic estimator of the gravitational wave power as a function of direction on the sky that is sensitive to the polarization state of the wave. In this way, we describe the full, $S_2 \times S_2$, state-space of a polarized gravitational wave background across the sky and the Poincar\'e sphere describing polarization. A natural question arises from this perspective: what is the resolution of a polarized sky-map, i.e. effectively how many independent pixels can a such a map contain? In other words, how many distinct gravitational waves can a PTA, in principle, distinguish? It turns out the answer is finite, and is approximately $N_{\rm res} = 16 \times 2 = 32$ per frequency, where 16 is the number of resolvable sky-positions and 2 is the number of distinct polarization states. This corresponds to an angular resolution of $58^\circ$, which can be achieved by a PTA with more than $N_{\rm pulsar} \gtrsim 20$ pulsars. We demonstrate that the variance of the map is equivalent to the HD significance, while for a single point source, a 3-$\sigma$ HD signal corresponds to a 5.2-$\sigma$ map significance.
|
[
{
"version": "v1",
"created": "Mon, 28 Jul 2025 23:05:56 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 04:57:16 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Jow",
"Dylan L.",
""
],
[
"Tsai",
"Anna",
""
],
[
"Pen",
"Ue-Li",
""
]
] |
2507.21647
|
Muhammad Shahzad Anjum
|
Muhammad S. Anjum, Shu-Jin Hou, Liang Chen, Zhigang Li, Minfeng Gu
|
From Equipartition to Curvature: The Spectral Evolution of 4FGL Blazars
|
Accepted for publication in ApJS
| null |
10.3847/1538-4365/adf4dc
| null |
astro-ph.HE astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the evolution of spectral energy distribution (SED) and underlying electron energy distribution (EED) by modeling the nearly simultaneous broadband spectra of selected bright 4FGL blazars, in the context of a combined cooling and stochastic acceleration scenario. We find that one-zone leptonic model with log-parabolic (LP) EED can successfully fit the GeV-TeV emission of blazars. The synchrotron frequency $\nu_s$ of blazars mainly evolves due to variation of electron peak energy $\gamma_{3p}$. The BL Lac objects (BL Lacs) show a negative trend in the $\nu_s- \nu_s L_s$ SED plane, known as blazar sequence, that does not seem to be an artifact of Doppler boosting, but driven by the equipartition constraints. A positive correlation is found between the derived magnetic field $B$ and electron density $n_e$, whereas $n_e$ and $\gamma_{3p}$ negatively relate, as expected in an equipartition scenario. The flat spectrum radio quasars (FSRQs) deviate significantly from such a scenario, indicating their jet parameters should be varying independently. The synchrotron peak frequency $\nu_s$ and its spectral curvature $b_s$ negatively correlate for all blazars, confirming the stochastic particle acceleration in blazar jets. However, blazars do not show the signature of hard-sphere acceleration, indicating that magnetic turbulence in the jets might be soft and physical conditions might be near to steady state, consistent with equipartition. Furthermore, for BL Lacs, the SED curvature $b_s$ and the EED curvature $r$ and nearly meet the theoretical relationship $r=5b_s$, whereas the FSRQs show large deviation due to poor constrain on $b_s$ due to presence of thermal component.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 10:04:50 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Jul 2025 07:50:14 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Anjum",
"Muhammad S.",
""
],
[
"Hou",
"Shu-Jin",
""
],
[
"Chen",
"Liang",
""
],
[
"Li",
"Zhigang",
""
],
[
"Gu",
"Minfeng",
""
]
] |
2507.22728
|
Shi-Lei Su
|
Bing-Bing Liu and Shi-Lei Su
|
Gain on ground state of quantum system for truly $\mathcal{PT}$ symmetry
|
7 pages, 3 figures
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
For a truly $\mathcal{PT}$-symmetric quantum system, the conventional non-Hermitian Hamiltonian is $H = \Omega\sigma_x -i\gamma|1\rangle\langle1| + i\gamma|0\rangle\langle0|$, where $\Omega$ and $\gamma$ are real parameters and $\sigma_x$ denotes Pauli X operator. These three terms represent coherent coupling, loss (on state $|1\rangle$), and gain (on state $|0\rangle$), respectively. Although the works in [Phys. Rev. Lett. \textbf{101}, 230404 (2008); Phys. Rev. Lett. \textbf{119}, 190401 (2017); Science \textbf{364}, 878 (2019)] proposed theoretically and/or demonstrate dilation methods for a truly parity-time($\mathcal{PT}$)-symmetric Hamiltonian by embedding into larger Hermitian space, directly realizing the gain term $+i\gamma|0\rangle\langle0|$ has still remained an outstanding challenge for quantum system. While systems omitting this gain term can exhibit a passively $\mathcal{PT}$-symmetric energy spectrum (featuring a parallel imaginary shift) and display related phenomena, they fail to capture the full physical behavior and unique properties inherent to truly $\mathcal{PT}$-symmetric systems. In this manuscript, we propose a method to achieve effective gain on the ground state $|0\rangle$ ($+i\gamma|0\rangle\langle0|$) after averaging all trajectories, by integrating the S{\o}rensen-Reiter effective operator method with the Wiseman-Milburn master equation for continuous measurement and instantaneous feedback control after averaging the evolution over all trajectories. This approach provides a possible pathway to efficiently construct truly $\mathcal{PT}$-symmetric quantum devices, offering a powerful platform for engineering quantum resources vital for quantum information technology applications.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 14:47:50 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 12:03:54 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 03:43:58 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Liu",
"Bing-Bing",
""
],
[
"Su",
"Shi-Lei",
""
]
] |
2507.22751
|
Elspeth Cudmore
|
Antoine Armatol, Corinne Augier, Louis Bailly-Salins, Guillaume Baulieu, Laurent Berg\'e, Julien Billard, Juliette Bl\'e, Guillaume Bres, Jean-Louis Bret, Alexandre Broniatowski, Martino Calvo, Antonella Cavanna, Antoine Cazes, Emanuela Celi, David Chaize, Mohammed Chala, Maurice Chappellier, Luke Chaplinsky, Guillaume Chemin, Ran Chen, Jules Colas, Laurent Couraud, Elspeth Cudmore, Maryvonne De Jesus, Nicole Dombrowski, Louis Dumoulin, Alan Durnez, Olivier Exshaw, Sylvain Ferriol, Enectali Figueroa-Feliciano, Joseph A. Formaggio, Stephane Fuard, Jules Gascon, Andrea Giuliani, Corinne Goy, Cyrille Guerin, Elsa Guy, Le\"ila Haegel, Scott A. Hertel, Christophe Hoarau, Ziqing Hong, Jean-Christophe Ianigro, Yong Jin, Alexandre Juillard, Temirlan Khussainov, Andrew Kubik, Jacob Lamblin, Hugues Lattaud, Tatiana Le-Bellec, Laetitia Leroy, Mingyu Li, Alexey Lubashevskiy, Stefanos Marnieros, Nicolas Martini, Julien Minet, Alessandro Monfardini, Franck Mounier, Valentina Novati, Emiliano Olivieri, Pratyush K. Patel, Eric Perbet, Harold Douglas Pinckney, Denys V. Poda, Dmitrii Ponomarev, Wouter Van De Pontseele, Jean-S\'ebastien Real, Faith C. Reyes, Alejandro Rodriguez, Murielle Rousseau, Sergey Rozov, Irina Rozova, Brianna Ryan, Deeksha Sabhari, Silvia Scorza, Renaud Serra, Yegor Shevchik, Torsten Soldner, Anne Stutz, Christian Ulysse, Lionel Vagneron, Sergey Vasilyev, Francis Vezzu, Paul Vittaz, Evgeny Yakushev, Jiatong Yang, Daniya Zinatulina
|
Characterization of mini-CryoCube detectors from the Ricochet experiment commissioning at the Institut Laue-Langevin
|
16 pages, 9 figures
| null | null | null |
hep-ex astro-ph.IM physics.ins-det
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The Ricochet experiment aims to measure the coherent elastic neutrino-nucleus scattering process from antineutrinos emitted by a research nuclear reactor operated by the Institut Laue-Langevin (Grenoble, France). This article presents a description of the Ricochet experimental installation and the detector performance achieved during its commissioning with a mini-CryoCube module consisting of three 42-gram germanium cryogenic calorimeters. The baseline resolutions and background levels are reported both during reactor-on and reactor-off periods, and as noise mitigation techniques were improved. A baseline resolution of 40 eV electron equivalent was achieved for the ionization channel after setup improvements, and the phonon channel resolutions ranged from 50 to 80 eV of total phonon energy. In the energy region from 2 to 7 keV, a nuclear recoil rate of 15(2) events/(kg day keV) is measured during the reactor-off period selecting events in coincidence with muon veto signals. This rate is in agreement with the cosmogenic neutron rate calculated from GEANT4 simulations. After the rejection of events in coincidence with signals in the muon veto detectors, a combined 90% C.L. limit on the nuclear recoil background of < 9 events/(kg day keV) is obtained in that energy region during the reactor-on period, which is compatible with our GEANT4 model calculation corresponding to a total rate of 5 events/(kg day keV). The sensitivity of this analysis was however found to be limited by a surface event contamination which is currently being addressed by the Ricochet Collaboration with upgraded detectors.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 15:08:43 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Armatol",
"Antoine",
""
],
[
"Augier",
"Corinne",
""
],
[
"Bailly-Salins",
"Louis",
""
],
[
"Baulieu",
"Guillaume",
""
],
[
"Bergé",
"Laurent",
""
],
[
"Billard",
"Julien",
""
],
[
"Blé",
"Juliette",
""
],
[
"Bres",
"Guillaume",
""
],
[
"Bret",
"Jean-Louis",
""
],
[
"Broniatowski",
"Alexandre",
""
],
[
"Calvo",
"Martino",
""
],
[
"Cavanna",
"Antonella",
""
],
[
"Cazes",
"Antoine",
""
],
[
"Celi",
"Emanuela",
""
],
[
"Chaize",
"David",
""
],
[
"Chala",
"Mohammed",
""
],
[
"Chappellier",
"Maurice",
""
],
[
"Chaplinsky",
"Luke",
""
],
[
"Chemin",
"Guillaume",
""
],
[
"Chen",
"Ran",
""
],
[
"Colas",
"Jules",
""
],
[
"Couraud",
"Laurent",
""
],
[
"Cudmore",
"Elspeth",
""
],
[
"De Jesus",
"Maryvonne",
""
],
[
"Dombrowski",
"Nicole",
""
],
[
"Dumoulin",
"Louis",
""
],
[
"Durnez",
"Alan",
""
],
[
"Exshaw",
"Olivier",
""
],
[
"Ferriol",
"Sylvain",
""
],
[
"Figueroa-Feliciano",
"Enectali",
""
],
[
"Formaggio",
"Joseph A.",
""
],
[
"Fuard",
"Stephane",
""
],
[
"Gascon",
"Jules",
""
],
[
"Giuliani",
"Andrea",
""
],
[
"Goy",
"Corinne",
""
],
[
"Guerin",
"Cyrille",
""
],
[
"Guy",
"Elsa",
""
],
[
"Haegel",
"Leïla",
""
],
[
"Hertel",
"Scott A.",
""
],
[
"Hoarau",
"Christophe",
""
],
[
"Hong",
"Ziqing",
""
],
[
"Ianigro",
"Jean-Christophe",
""
],
[
"Jin",
"Yong",
""
],
[
"Juillard",
"Alexandre",
""
],
[
"Khussainov",
"Temirlan",
""
],
[
"Kubik",
"Andrew",
""
],
[
"Lamblin",
"Jacob",
""
],
[
"Lattaud",
"Hugues",
""
],
[
"Le-Bellec",
"Tatiana",
""
],
[
"Leroy",
"Laetitia",
""
],
[
"Li",
"Mingyu",
""
],
[
"Lubashevskiy",
"Alexey",
""
],
[
"Marnieros",
"Stefanos",
""
],
[
"Martini",
"Nicolas",
""
],
[
"Minet",
"Julien",
""
],
[
"Monfardini",
"Alessandro",
""
],
[
"Mounier",
"Franck",
""
],
[
"Novati",
"Valentina",
""
],
[
"Olivieri",
"Emiliano",
""
],
[
"Patel",
"Pratyush K.",
""
],
[
"Perbet",
"Eric",
""
],
[
"Pinckney",
"Harold Douglas",
""
],
[
"Poda",
"Denys V.",
""
],
[
"Ponomarev",
"Dmitrii",
""
],
[
"Van De Pontseele",
"Wouter",
""
],
[
"Real",
"Jean-Sébastien",
""
],
[
"Reyes",
"Faith C.",
""
],
[
"Rodriguez",
"Alejandro",
""
],
[
"Rousseau",
"Murielle",
""
],
[
"Rozov",
"Sergey",
""
],
[
"Rozova",
"Irina",
""
],
[
"Ryan",
"Brianna",
""
],
[
"Sabhari",
"Deeksha",
""
],
[
"Scorza",
"Silvia",
""
],
[
"Serra",
"Renaud",
""
],
[
"Shevchik",
"Yegor",
""
],
[
"Soldner",
"Torsten",
""
],
[
"Stutz",
"Anne",
""
],
[
"Ulysse",
"Christian",
""
],
[
"Vagneron",
"Lionel",
""
],
[
"Vasilyev",
"Sergey",
""
],
[
"Vezzu",
"Francis",
""
],
[
"Vittaz",
"Paul",
""
],
[
"Yakushev",
"Evgeny",
""
],
[
"Yang",
"Jiatong",
""
],
[
"Zinatulina",
"Daniya",
""
]
] |
2507.23751
|
Jason Weston
|
Ping Yu, Jack Lanchantin, Tianlu Wang, Weizhe Yuan, Olga Golovneva, Ilia Kulikov, Sainbayar Sukhbaatar, Jason Weston, Jing Xu
|
CoT-Self-Instruct: Building high-quality synthetic prompts for reasoning and non-reasoning tasks
| null | null | null | null |
cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We propose CoT-Self-Instruct, a synthetic data generation method that instructs LLMs to first reason and plan via Chain-of-Thought (CoT) based on given seed tasks, and then generate a new synthetic example of similar quality and complexity. This is followed by a filtering step to select high-quality data using automatic metrics, which are then used for LLM training. In verifiable reasoning, our synthetic data significantly outperforms existing training datasets, such as s1k and OpenMathReasoning, when evaluated on MATH500, AMC23, AIME24, and GPQA-Diamond. For non-verifiable instruction-following tasks, our method surpasses the performance of both human and standard Self-Instruct training data on the AlpacaEval 2.0 and Arena-Hard benchmarks.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 17:38:50 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 14:36:00 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Yu",
"Ping",
""
],
[
"Lanchantin",
"Jack",
""
],
[
"Wang",
"Tianlu",
""
],
[
"Yuan",
"Weizhe",
""
],
[
"Golovneva",
"Olga",
""
],
[
"Kulikov",
"Ilia",
""
],
[
"Sukhbaatar",
"Sainbayar",
""
],
[
"Weston",
"Jason",
""
],
[
"Xu",
"Jing",
""
]
] |
2508.00404
|
Xian-Hui Ge
|
Zi-Yu Hou, Yu-Qi Lei, Xian-Hui Ge
|
Quasi-Normal Modes and Nonlinear Electrodynamics in Black Hole Phase Transitions
|
18 pages, 11 figures, references added, appendix added
| null | null | null |
hep-th astro-ph.HE gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the connection between thermodynamic phase transitions and quasi-normal modes (QNMs) in charged black holes with a positive curvature constant, within the framework of $F(R)$-Euler-Heisenberg gravity. Nonlinear electromagnetic fields lead to rich thermodynamic phase structures and significantly affect the QNMs of massless scalar fields. By analyzing the QNMs spectrum, we find that the transition point marking the disappearance of the divergence in the QNMs slope parameter $K$ aligns with the change of the thermodynamic phase structure described by the heat capacity, within the bounds of computational uncertainty. This precise matching holds under variations of the curvature parameter and charge. Furthermore, we show that larger angular quantum number $l$ diminishes this correspondence, while higher overtone number $n$ restores it beyond a threshold. These findings demonstrate that thermodynamic phase transitions of black holes carry embedded dynamical information, uncovering a fundamental link between black hole thermodynamic and dynamical properties.
|
[
{
"version": "v1",
"created": "Fri, 01 Aug 2025 08:05:46 GMT"
},
{
"version": "v2",
"created": "Sun, 10 Aug 2025 07:03:09 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 08:09:53 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Hou",
"Zi-Yu",
""
],
[
"Lei",
"Yu-Qi",
""
],
[
"Ge",
"Xian-Hui",
""
]
] |
2508.01197
|
Zhan Shi
|
Zhan Shi, Song Wang, Junbo Chen, Jianke Zhu
|
A Coarse-to-Fine Approach to Multi-Modality 3D Occupancy Grounding
| null | null | null | null |
cs.CV cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Visual grounding aims to identify objects or regions in a scene based on natural language descriptions, essential for spatially aware perception in autonomous driving. However, existing visual grounding tasks typically depend on bounding boxes that often fail to capture fine-grained details. Not all voxels within a bounding box are occupied, resulting in inaccurate object representations. To address this, we introduce a benchmark for 3D occupancy grounding in challenging outdoor scenes. Built on the nuScenes dataset, it integrates natural language with voxel-level occupancy annotations, offering more precise object perception compared to the traditional grounding task. Moreover, we propose GroundingOcc, an end-to-end model designed for 3D occupancy grounding through multi-modal learning. It combines visual, textual, and point cloud features to predict object location and occupancy information from coarse to fine. Specifically, GroundingOcc comprises a multimodal encoder for feature extraction, an occupancy head for voxel-wise predictions, and a grounding head to refine localization. Additionally, a 2D grounding module and a depth estimation module enhance geometric understanding, thereby boosting model performance. Extensive experiments on the benchmark demonstrate that our method outperforms existing baselines on 3D occupancy grounding. The dataset is available at https://github.com/RONINGOD/GroundingOcc.
|
[
{
"version": "v1",
"created": "Sat, 02 Aug 2025 05:05:50 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 12:05:59 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Shi",
"Zhan",
""
],
[
"Wang",
"Song",
""
],
[
"Chen",
"Junbo",
""
],
[
"Zhu",
"Jianke",
""
]
] |
2508.01415
|
Mingcong Lei
|
Mingcong Lei, Honghao Cai, Binbin Que, Zezhou Cui, Liangchen Tan, Junkun Hong, Gehan Hu, Shuangyu Zhu, Yimou Wu, Shaohan Jiang, Ge Wang, Zhen Li, Shuguang Cui, Yiming Zhao, Yatong Han
|
RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Lifelong Learning in Physical Embodied Systems
| null | null | null | null |
cs.RO cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We present RoboMemory, a brain-inspired multi-memory framework for lifelong learning in physical embodied systems, addressing critical challenges in real-world environments: continuous learning, multi-module memory latency, task correlation capture, and infinite-loop mitigation in closed-loop planning. Grounded in cognitive neuroscience, it integrates four core modules: the Information Preprocessor (thalamus-like), the Lifelong Embodied Memory System (hippocampus-like), the Closed-Loop Planning Module (prefrontal lobe-like), and the Low-Level Executer (cerebellum-like) to enable long-term planning and cumulative learning. The Lifelong Embodied Memory System, central to the framework, alleviates inference speed issues in complex memory frameworks via parallelized updates/retrieval across Spatial, Temporal, Episodic, and Semantic submodules. It incorporates a dynamic Knowledge Graph (KG) and consistent architectural design to enhance memory consistency and scalability. Evaluations on EmbodiedBench show RoboMemory outperforms the open-source baseline (Qwen2.5-VL-72B-Ins) by 25% in average success rate and surpasses the closed-source State-of-the-Art (SOTA) (Claude3.5-Sonnet) by 5%, establishing new SOTA. Ablation studies validate key components (critic, spatial memory, long-term memory), while real-world deployment confirms its lifelong learning capability with significantly improved success rates across repeated tasks. RoboMemory alleviates high latency challenges with scalability, serving as a foundational reference for integrating multi-modal memory systems in physical robots.
|
[
{
"version": "v1",
"created": "Sat, 02 Aug 2025 15:39:42 GMT"
},
{
"version": "v2",
"created": "Thu, 07 Aug 2025 08:47:22 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 09:32:27 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Lei",
"Mingcong",
""
],
[
"Cai",
"Honghao",
""
],
[
"Que",
"Binbin",
""
],
[
"Cui",
"Zezhou",
""
],
[
"Tan",
"Liangchen",
""
],
[
"Hong",
"Junkun",
""
],
[
"Hu",
"Gehan",
""
],
[
"Zhu",
"Shuangyu",
""
],
[
"Wu",
"Yimou",
""
],
[
"Jiang",
"Shaohan",
""
],
[
"Wang",
"Ge",
""
],
[
"Li",
"Zhen",
""
],
[
"Cui",
"Shuguang",
""
],
[
"Zhao",
"Yiming",
""
],
[
"Han",
"Yatong",
""
]
] |
2508.01501
|
S. Ishmam Mohtashim
|
Shah Ishmam Mohtashim
|
RinQ: Towards predicting central sites in proteins on current quantum computers
| null | null | null | null |
quant-ph cond-mat.soft physics.bio-ph q-bio.QM
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce RinQ, a hybrid quantum-classical framework for identifying functionally critical residues in proteins by formulating centrality detection as a Quadratic Unconstrained Binary Optimization (QUBO) problem. Protein structures are modeled as residue interaction networks (RINs), and the QUBO formulations are solved using D-Wave's simulated annealing. Applied to a diverse set of proteins, RinQ consistently identifies central residues that closely align with classical benchmarks, demonstrating both the accuracy and robustness of the approach.
|
[
{
"version": "v1",
"created": "Sat, 02 Aug 2025 21:53:09 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 22:05:52 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Mohtashim",
"Shah Ishmam",
""
]
] |
2508.01550
|
Zhilong Chen
|
Zhilong Chen, Chengzong Zhao, Boyuan Chen, Dayi Lin, Yihao Chen, Arthur Leung, Gopi Krishnan Rajbahadur, Gustavo A. Oliva, Haoxiang Zhang, Aaditya Bhatia, Chong Chun Yong, Ahmed E. Hassan
|
RepoForge: Training a SOTA Fast-thinking SWE Agent with an End-to-End Data Curation Pipeline Synergizing SFT and RL at Scale
| null | null | null | null |
cs.SE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Training software engineering (SWE) LLMs is bottlenecked by expensive infrastructure, inefficient evaluation pipelines, scarce training data, and costly quality control. We present RepoForge, an autonomous, end-to-end pipeline that generates, evaluates, and trains SWE agents at scale. Our key contributions include: (1) RepoForge-8B-Agent, achieving 17.4\% on SWE-Bench-Verified~\citep{swebench_verified2024}, establishing new state-of-the-art for $\leq$8B non-thinking LLMs; (2) 7,304 executable environments auto-generated from real GitHub commits with zero manual intervention; (3) 14$\times$ storage reduction (1.4GB $\rightarrow$ 102MB per instance) via intelligent dependency management and image pruning; (4) $>$70\% faster evaluation using a Ray-powered~\citep{ray2018} distributed RepoForge harness; (5) 19,000$\times$ cheaper labeling through our automated SPICE~\citep{spice2024} difficulty assessment technique. By unifying storage-efficient sandboxing, Ray-powered evaluation harness, automated data generation, SPICE-based labeling, and bubble-free RL scaffold, we demonstrate that even $\leq$8B models can reach new state-of-the-art performance on demanding benchmarks like SWE-Bench-Verified. Our approach addresses critical bottlenecks in SWE agent training: high storage costs of container-based evaluation, inefficient sequential reward pipelines, limited availability of high-quality training data, expensive manual labeling, and multi-turn RL pipeline bottlenecks.
|
[
{
"version": "v1",
"created": "Sun, 03 Aug 2025 02:34:16 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 14:56:47 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Chen",
"Zhilong",
""
],
[
"Zhao",
"Chengzong",
""
],
[
"Chen",
"Boyuan",
""
],
[
"Lin",
"Dayi",
""
],
[
"Chen",
"Yihao",
""
],
[
"Leung",
"Arthur",
""
],
[
"Rajbahadur",
"Gopi Krishnan",
""
],
[
"Oliva",
"Gustavo A.",
""
],
[
"Zhang",
"Haoxiang",
""
],
[
"Bhatia",
"Aaditya",
""
],
[
"Yong",
"Chong Chun",
""
],
[
"Hassan",
"Ahmed E.",
""
]
] |
2508.02776
|
Hannah Stacey
|
H. R. Stacey, D. M. Powell, S. Vegetti, J. P. McKean, and D. Wen
|
Investigation of mass substructure in gravitational lens system SDP.81 with ALMA long-baseline observations
|
Accepted for publication in A&A
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
The prevalence and properties of low-mass dark matter haloes serve as a crucial test for understanding the nature of dark matter, and may be constrained through the gravitational deflection of strongly lensed arcs. Previous studies found evidence for the presence of low-mass dark matter haloes in observations of the gravitationally lensed, dusty star-forming galaxy SDP.81, using the Atacama Large Millimetre/sub-millimetre Array (ALMA). In this work, we analyse these observations to assess the robustness of these reported results. While our analysis indicates that the data support additional angular structure in the lensing mass distribution beyond an elliptical power-law density profile, we do not find evidence for two previously reported sub-halo detections. However, we verify with realistic mock data that we could have found evidence in favour of a previously reported $\approx 10^{9}\,{\rm M_{\odot}}$ sub-halo with a log Bayes factor of 29, should it exist in the real data. After testing various systematics, we find that this previous sub-halo inference was most likely spurious and resulted from an inadequate smooth model, specifically, poorly fitting multipoles. While we do not find evidence in favour of any individual sub-halo, we find evidence for similarity in the lensing signatures of multipoles ($m=3,4$) and single massive sub-haloes, consistent with other recent work. We suggest that future searches for low-mass haloes in lensed arcs include lens angular structure in the form of multipoles up to 4th order and require a good-fitting smooth model as a prerequisite. Overall, our findings demonstrate the suitability of ALMA data of this quality to simultaneously constrain the abundance of low-mass haloes and lens angular structure.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 09:25:53 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Stacey",
"H. R.",
""
],
[
"Powell",
"D. M.",
""
],
[
"Vegetti",
"S.",
""
],
[
"McKean",
"J. P.",
""
],
[
"Wen",
"D.",
""
]
] |
2508.03629
|
Matthieu Madera
|
Matthieu Madera
|
Deformations of geometric structures on a subclass of LVM threefolds
|
52 pages
| null | null | null |
math.DG math.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study deformations of geometric structures on some LVM manifolds of complex dimension $3$. More precisely, we study resonant structures, a particular type of $(G,X)$-structures, via the Ehresmann-Thurston principle, and their link with the deformation of the complex structure of the LVM manifold in the sense of Kodaira, Spencer and Kuranishi. We describe the Kuranishi family of these manifolds and construct a family containing all of them and complete at every point.
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 16:47:39 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 13:09:06 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Madera",
"Matthieu",
""
]
] |
2508.03665
|
Claudiu Leoveanu-Condrei Leo
|
Claudiu Leoveanu-Condrei
|
A DbC Inspired Neurosymbolic Layer for Trustworthy Agent Design
|
4 pages, 1 figure
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Generative models, particularly Large Language Models (LLMs), produce fluent outputs yet lack verifiable guarantees. We adapt Design by Contract (DbC) and type-theoretic principles to introduce a contract layer that mediates every LLM call. Contracts stipulate semantic and type requirements on inputs and outputs, coupled with probabilistic remediation to steer generation toward compliance. The layer exposes the dual view of LLMs as semantic parsers and probabilistic black-box components. Contract satisfaction is probabilistic and semantic validation is operationally defined through programmer-specified conditions on well-typed data structures. More broadly, this work postulates that any two agents satisfying the same contracts are \emph{functionally equivalent} with respect to those contracts.
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 17:24:50 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 16:24:59 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 16:15:16 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Leoveanu-Condrei",
"Claudiu",
""
]
] |
2508.03700
|
Shaokang Dong
|
Liujian Tang, Shaokang Dong, Yijia Huang, Minqi Xiang, Hongtao Ruan, Bin Wang, Shuo Li, Zhiheng Xi, Zhihui Cao, Hailiang Pang, Heng Kong, He Yang, Mingxu Chai, Zhilin Gao, Xingyu Liu, Yingnan Fu, Jiaming Liu, Xuanjing Huang, Yu-Gang Jiang, Tao Gui, Qi Zhang, Kang Wang, Yunke Zhang, Yuran Wang
|
MagicGUI: A Foundational Mobile GUI Agent with Scalable Data Pipeline and Reinforcement Fine-tuning
| null | null | null | null |
cs.HC cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This paper presents MagicGUI, a foundational mobile GUI agent designed to address critical challenges in perception, grounding, and reasoning within real-world mobile GUI environments. The framework is underpinned by following six key components: (1) a comprehensive and accurate dataset, constructed via the scalable GUI Data Pipeline, which aggregates the largest and most diverse GUI-centric multimodal data to date from open-source repositories, automated crawling, and targeted manual annotation; (2) enhanced perception and grounding capabilities, facilitating fine-grained multimodal alignment for UI element referencing, grounding, and screen comprehension; (3) a comprehensive and unified action space, encompassing both fundamental UI operations and complex interactive intents to support human-agent interactions; (4) planning-oriented reasoning mechanisms that enable the model to decompose complex user instructions into sequential actions with explicit intermediate meta-paln reasoning; (5) an iterative two-stage training procedure, combining large-scale continue pre-training on 7.8M samples with reinforcement fine-tuning utilizing a spatially enhanced composite reward and dual filtering strategy; and (6) competitive performance on both the proprietary Magic-RICH benchmark and over a dozen public benchmarks, achieving superior performance across GUI perception and agent tasks, while demonstrating robust generalization and real-world deployment potential in practical mobile GUI scenarios, as detailed in Figure 1.
|
[
{
"version": "v1",
"created": "Sat, 19 Jul 2025 12:33:43 GMT"
},
{
"version": "v2",
"created": "Fri, 08 Aug 2025 13:25:14 GMT"
},
{
"version": "v3",
"created": "Sun, 24 Aug 2025 06:53:11 GMT"
},
{
"version": "v4",
"created": "Wed, 03 Sep 2025 15:20:53 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Tang",
"Liujian",
""
],
[
"Dong",
"Shaokang",
""
],
[
"Huang",
"Yijia",
""
],
[
"Xiang",
"Minqi",
""
],
[
"Ruan",
"Hongtao",
""
],
[
"Wang",
"Bin",
""
],
[
"Li",
"Shuo",
""
],
[
"Xi",
"Zhiheng",
""
],
[
"Cao",
"Zhihui",
""
],
[
"Pang",
"Hailiang",
""
],
[
"Kong",
"Heng",
""
],
[
"Yang",
"He",
""
],
[
"Chai",
"Mingxu",
""
],
[
"Gao",
"Zhilin",
""
],
[
"Liu",
"Xingyu",
""
],
[
"Fu",
"Yingnan",
""
],
[
"Liu",
"Jiaming",
""
],
[
"Huang",
"Xuanjing",
""
],
[
"Jiang",
"Yu-Gang",
""
],
[
"Gui",
"Tao",
""
],
[
"Zhang",
"Qi",
""
],
[
"Wang",
"Kang",
""
],
[
"Zhang",
"Yunke",
""
],
[
"Wang",
"Yuran",
""
]
] |
2508.03770
|
Felix Wagner
|
G. Angloher, S. Banik, G. Benato, A. Bento, A. Bertolini, R. Breier, C. Bucci, J. Burkhart, L. Canonica, A. D'Addabbo, S. Di Lorenzo, L. Einfalt, A. Erb, F. v. Feilitzsch, N. Ferreiro Iachellini, S. Fichtinger, D. Fuchs, A. Fuss, A. Garai, V.M. Ghete, P. Gorla, P.V. Guillaumon, S. Gupta, D. Hauff, M. Je\v{s}kovsk\'y, J. Jochum, M. Kaznacheeva, A. Kinast, H. Kluck, S. Kuckuk, H. Kraus, A. Langenk\"amper, M. Mancuso, L. Marini, L. Meyer, V. Mokina, A. Nilima, M. Olmi, T. Ortmann, C. Pagliarone, L. Pattavina, F. Petricca, W. Potzel, P. Povinec, F. Pr\"obst, F. Pucci, F. Reindl, J. Rothe, K. Sch\"affner, J. Schieck, D. Schmiedmayer, S. Sch\"onert, C. Schwertner, M. Stahlberg, L. Stodolsky, C. Strandhagen, R. Strauss, I. Usherov, F. Wagner, M. Willers, and V. Zema (CRESST Collaboration)
|
Description of CRESST-III lithium aluminate data
|
4 pages, 2 figures
| null | null | null |
hep-ex astro-ph.CO physics.ins-det
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Two detector modules with lithium aluminate targets were operated in the CRESST underground setup between February and June 2021. The data collected in this period was used to set the currently strongest cross-section upper limits on the spin-dependent interaction of dark matter (DM) with protons and neutrons for the mass region between 0.25 and 1.5 GeV/c$^2$. The data are available online. In this document, we describe how the data set should be used to reproduce our dark matter results.
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 04:52:36 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 18:35:03 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Angloher",
"G.",
"",
"CRESST Collaboration"
],
[
"Banik",
"S.",
"",
"CRESST Collaboration"
],
[
"Benato",
"G.",
"",
"CRESST Collaboration"
],
[
"Bento",
"A.",
"",
"CRESST Collaboration"
],
[
"Bertolini",
"A.",
"",
"CRESST Collaboration"
],
[
"Breier",
"R.",
"",
"CRESST Collaboration"
],
[
"Bucci",
"C.",
"",
"CRESST Collaboration"
],
[
"Burkhart",
"J.",
"",
"CRESST Collaboration"
],
[
"Canonica",
"L.",
"",
"CRESST Collaboration"
],
[
"D'Addabbo",
"A.",
"",
"CRESST Collaboration"
],
[
"Di Lorenzo",
"S.",
"",
"CRESST Collaboration"
],
[
"Einfalt",
"L.",
"",
"CRESST Collaboration"
],
[
"Erb",
"A.",
"",
"CRESST Collaboration"
],
[
"Feilitzsch",
"F. v.",
"",
"CRESST Collaboration"
],
[
"Iachellini",
"N. Ferreiro",
"",
"CRESST Collaboration"
],
[
"Fichtinger",
"S.",
"",
"CRESST Collaboration"
],
[
"Fuchs",
"D.",
"",
"CRESST Collaboration"
],
[
"Fuss",
"A.",
"",
"CRESST Collaboration"
],
[
"Garai",
"A.",
"",
"CRESST Collaboration"
],
[
"Ghete",
"V. M.",
"",
"CRESST Collaboration"
],
[
"Gorla",
"P.",
"",
"CRESST Collaboration"
],
[
"Guillaumon",
"P. V.",
"",
"CRESST Collaboration"
],
[
"Gupta",
"S.",
"",
"CRESST Collaboration"
],
[
"Hauff",
"D.",
"",
"CRESST Collaboration"
],
[
"Ješkovský",
"M.",
"",
"CRESST Collaboration"
],
[
"Jochum",
"J.",
"",
"CRESST Collaboration"
],
[
"Kaznacheeva",
"M.",
"",
"CRESST Collaboration"
],
[
"Kinast",
"A.",
"",
"CRESST Collaboration"
],
[
"Kluck",
"H.",
"",
"CRESST Collaboration"
],
[
"Kuckuk",
"S.",
"",
"CRESST Collaboration"
],
[
"Kraus",
"H.",
"",
"CRESST Collaboration"
],
[
"Langenkämper",
"A.",
"",
"CRESST Collaboration"
],
[
"Mancuso",
"M.",
"",
"CRESST Collaboration"
],
[
"Marini",
"L.",
"",
"CRESST Collaboration"
],
[
"Meyer",
"L.",
"",
"CRESST Collaboration"
],
[
"Mokina",
"V.",
"",
"CRESST Collaboration"
],
[
"Nilima",
"A.",
"",
"CRESST Collaboration"
],
[
"Olmi",
"M.",
"",
"CRESST Collaboration"
],
[
"Ortmann",
"T.",
"",
"CRESST Collaboration"
],
[
"Pagliarone",
"C.",
"",
"CRESST Collaboration"
],
[
"Pattavina",
"L.",
"",
"CRESST Collaboration"
],
[
"Petricca",
"F.",
"",
"CRESST Collaboration"
],
[
"Potzel",
"W.",
"",
"CRESST Collaboration"
],
[
"Povinec",
"P.",
"",
"CRESST Collaboration"
],
[
"Pröbst",
"F.",
"",
"CRESST Collaboration"
],
[
"Pucci",
"F.",
"",
"CRESST Collaboration"
],
[
"Reindl",
"F.",
"",
"CRESST Collaboration"
],
[
"Rothe",
"J.",
"",
"CRESST Collaboration"
],
[
"Schäffner",
"K.",
"",
"CRESST Collaboration"
],
[
"Schieck",
"J.",
"",
"CRESST Collaboration"
],
[
"Schmiedmayer",
"D.",
"",
"CRESST Collaboration"
],
[
"Schönert",
"S.",
"",
"CRESST Collaboration"
],
[
"Schwertner",
"C.",
"",
"CRESST Collaboration"
],
[
"Stahlberg",
"M.",
"",
"CRESST Collaboration"
],
[
"Stodolsky",
"L.",
"",
"CRESST Collaboration"
],
[
"Strandhagen",
"C.",
"",
"CRESST Collaboration"
],
[
"Strauss",
"R.",
"",
"CRESST Collaboration"
],
[
"Usherov",
"I.",
"",
"CRESST Collaboration"
],
[
"Wagner",
"F.",
"",
"CRESST Collaboration"
],
[
"Willers",
"M.",
"",
"CRESST Collaboration"
],
[
"Zema",
"V.",
"",
"CRESST Collaboration"
]
] |
2508.03797
|
Soobeom Chung
|
Soobeom Chung, Jeff A. Dror
|
Cosmic Axion Background Detection Using Resonant Cavity Arrays
|
20 pages, 4 figures
| null | null | null |
hep-ph astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
The axion is a well-motivated and generic extension of the Standard Model. If produced in the early universe, axions may still be relativistic today, forming a Cosmic Axion Background (C$a$B) potentially detectable in direct detection experiments. A key challenge, however, is that the C$a$B is expected to have a broad energy spectrum, limiting the effectiveness of resonant cavity experiments designed for narrowband searches. We propose a new strategy using multi-cavity arrays to distinguish signal from background noise by exploiting spatial correlations in the axion-induced electric fields. We compute the two-point correlation function for electric fields in spatially separated cavities sourced by an isotropic C$a$B. Analyzing various cavity geometries, we find that stacked, wide-base cavity arrays offer optimal sensitivity to the axion signal. We apply our formalism to prospective upgrades of the ADMX experiment, including configurations with four and eighteen coupled cavities.
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 18:00:01 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Chung",
"Soobeom",
""
],
[
"Dror",
"Jeff A.",
""
]
] |
2508.03801
|
Jennifer Burt
|
Jennifer A. Burt, Robert T. Zellem, David R. Ciardi, Shubham Kanodia, Geoffrey Bryden, Tiffany Kataria, Kyle A. Pearson, Jessie L. Christiansen, Charles Beichman, B.J. Fulton, Mark Swain
|
A New Approach to Compiling Exoatmospheric Target Lists And Quantifying the Ground-Based Resources Needed to Vet Them
|
25 pages, 13 figures, 2 tables. Accepted for publication in the Astronomical Journal
| null | null | null |
astro-ph.EP astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
Transiting exoplanet atmospheric characterization is currently in a golden age as dozens of exoplanet atmospheres are being studied by NASA's Hubble and James Webb Space Telescopes. This trend is expected to continue with NASA's Pandora Smallsat and Roman Space Telescope and ESA's Ariel mission (all expected to launch within this decade) and NASA's Habitable Worlds Observatory (expected to launch in the early 2040s) all of which are centered around studying the atmospheres of exoplanets. Here we explore a new approach to constructing large scale exoatmospheric survey lists, which combines the use of traditional transmission/emission spectroscopy figures of merit with a focus on more-evenly sampling planets across a range of radii and equilibrium temperatures. After assembling a sample target list comprised of 750 transmission spectroscopy targets and 150 emission spectroscopy targets, we quantify the potential time lost to stale transit and eclipse ephemerides and find that hundreds of hours of space-based observing could be wasted given current uncertainties in orbital periods, transit epochs, and orbital eccentricities. We further estimate the amount of ground-based telescope time necessary to obtain sufficiently precise exoplanet masses and find that it exceeds 100 nights of 10m telescope time. Based upon these findings, we provide a list of recommendations that would make community efforts for preparation and interpretation of atmospheric characterization endeavors more effective and efficient. The strategies we recommend here can be used to support both current (e.g., HST and JWST) and future exoplanet atmosphere characterization missions (e.g., Pandora, Ariel, Roman, and the Habitable Worlds Observatory).
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 18:00:03 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 17:07:46 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Burt",
"Jennifer A.",
""
],
[
"Zellem",
"Robert T.",
""
],
[
"Ciardi",
"David R.",
""
],
[
"Kanodia",
"Shubham",
""
],
[
"Bryden",
"Geoffrey",
""
],
[
"Kataria",
"Tiffany",
""
],
[
"Pearson",
"Kyle A.",
""
],
[
"Christiansen",
"Jessie L.",
""
],
[
"Beichman",
"Charles",
""
],
[
"Fulton",
"B. J.",
""
],
[
"Swain",
"Mark",
""
]
] |
2508.04416
|
Haoji Zhang
|
Haoji Zhang, Xin Gu, Jiawen Li, Chixiang Ma, Sule Bai, Chubin Zhang, Bowen Zhang, Zhichao Zhou, Dongliang He, Yansong Tang
|
Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
The video reasoning ability of multimodal large language models (MLLMs) is crucial for downstream tasks like video question answering and temporal grounding. While recent approaches have explored text-based chain-of-thought (CoT) reasoning for MLLMs, these methods often suffer from limited cross-modal interaction and increased hallucination, especially with longer videos or reasoning chains. To address these challenges, we propose Video Intelligence via Tool-Augmented Learning (VITAL), a novel end-to-end agentic video reasoning framework. With a visual toolbox, the model can densely sample new video frames on demand and generate multimodal CoT for precise long video reasoning. We observe that temporal grounding and question answering are mutually beneficial for video understanding tasks. Therefore, we construct two high-quality multi-task video reasoning datasets MTVR-CoT-72k for supervised fine-tuning and MTVR-RL-110k for reinforcement learning. Moreover, we propose a Difficulty-aware Group Relative Policy Optimization algorithm (DGRPO) to mitigate difficulty imbalance in multi-task reinforcement learning. Extensive experiments on 11 challenging video understanding benchmarks demonstrate the advanced reasoning ability of VITAL, outperforming existing methods in video question answering and temporal grounding tasks, especially in long video scenarios. Code is available at https://zhang9302002.github.io/thinkingwithvideos-page/.
|
[
{
"version": "v1",
"created": "Wed, 06 Aug 2025 13:03:21 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 07:11:03 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zhang",
"Haoji",
""
],
[
"Gu",
"Xin",
""
],
[
"Li",
"Jiawen",
""
],
[
"Ma",
"Chixiang",
""
],
[
"Bai",
"Sule",
""
],
[
"Zhang",
"Chubin",
""
],
[
"Zhang",
"Bowen",
""
],
[
"Zhou",
"Zhichao",
""
],
[
"He",
"Dongliang",
""
],
[
"Tang",
"Yansong",
""
]
] |
2508.05093
|
Xiaoxiao Xu
|
Tiantian He, Minzhi Xie, Runtong Li, Xiaoxiao Xu, Jiaqi Yu, Zixiu Wang, Lantao Hu, Han Li, Kun Gai
|
An End-to-End Multi-objective Ensemble Ranking Framework for Video Recommendation
| null | null | null | null |
cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
We propose a novel End-to-end Multi-objective Ensemble Ranking framework (EMER) for the multi-objective ensemble ranking module, which is the most critical component of the short video recommendation system. EMER enhances personalization by replacing manually-designed heuristic formulas with an end-to-end modeling paradigm. EMER introduces a meticulously designed loss function to address the fundamental challenge of defining effective supervision for ensemble ranking, where no single ground-truth signal can fully capture user satisfaction. Moreover, EMER introduces novel sample organization method and transformer-based network architecture to capture the comparative relationships among candidates, which are critical for effective ranking. Additionally, we have proposed an offline-online consistent evaluation system to enhance the efficiency of offline model optimization, which is an established yet persistent challenge within the multi-objective ranking domain in industry. Abundant empirical tests are conducted on a real industrial dataset, and the results well demonstrate the effectiveness of our proposed framework. In addition, our framework has been deployed in the primary scenarios of Kuaishou, a short video recommendation platform with hundreds of millions of daily active users, achieving a 1.39% increase in overall App Stay Time and a 0.196% increase in 7-day user Lifetime(LT7), which are substantial improvements.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 07:21:46 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 09:44:46 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"He",
"Tiantian",
""
],
[
"Xie",
"Minzhi",
""
],
[
"Li",
"Runtong",
""
],
[
"Xu",
"Xiaoxiao",
""
],
[
"Yu",
"Jiaqi",
""
],
[
"Wang",
"Zixiu",
""
],
[
"Hu",
"Lantao",
""
],
[
"Li",
"Han",
""
],
[
"Gai",
"Kun",
""
]
] |
2508.05552
|
Ailing Wang
|
Tao An, Ailing Wang, Yu-Chen Huang, Jia-Pei Feng, Yuanqi Liu, Zhongli Zhang, Zi-Gao Dai
|
A Deep VLA Search for a Persistent Radio Counterpart to the One-off FRB 20250316A
|
10 pages, 3 figures, two tables. Accepted by ApJL
| null | null | null |
astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Fast Radio Burst (FRB) 20250316A, detected by CHIME on 2025 March 16 with a fluence of $1.7\pm0.1~\mathrm{Jy\,ms}$ and a dispersion measure of $161.3\pm0.4~\mathrm{pc\,cm^{-3}}$, ranks among the brightest extragalactic FRBs at $\sim 40$ Mpc. We obtained deep Karl G. Jansky Very Large Array follow-up at 15~GHz on 2025 April 5 and 9 and find no persistent radio source (PRS). Our best image reaches an rms of $2.8~\mu\mathrm{Jy\,beam^{-1}}$, yielding a $3\sigma$ upper limit of $<8.4~\mu\mathrm{Jy}$ at the FRB position, corresponding to $\nu L_\nu < 2.4\times10^{35}~\mathrm{erg\,s^{-1}}$. These results represent among the most stringent constraints for a non-repeating FRB, lying $\gtrsim 3$ orders of magnitude below the $\nu L_\nu$ of compact persistent radio sources around well-studied repeaters, thereby disfavoring bright magnetar-nebula scenarios and pointing to low-density, weakly magnetized environments. Interpreting our limit through pulsar-/magnetar-wind synchrotron frameworks places joint constraints on ambient density and engine power. If the empirical PRS--rotation-measure trend reported for repeaters extends to one-off sources, our limit implies $\vert \mathrm{RM} \vert \lesssim 30~\mathrm{rad\,m^{-2}}$, consistent with a clean magneto-ionic sight line and progenitor channels such as neutron-star mergers or giant flares from older magnetars.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 16:32:59 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 12:23:15 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"An",
"Tao",
""
],
[
"Wang",
"Ailing",
""
],
[
"Huang",
"Yu-Chen",
""
],
[
"Feng",
"Jia-Pei",
""
],
[
"Liu",
"Yuanqi",
""
],
[
"Zhang",
"Zhongli",
""
],
[
"Dai",
"Zi-Gao",
""
]
] |
2508.05735
|
Daniel Grumiller
|
Ankit Aggarwal, Arjun Bagchi, Stephane Detournay, Daniel Grumiller, Max Riegler and Joan Simon
|
Universal sectors of two-dimensional Carrollian CFTs
|
45pp+acknowledgments
| null | null |
TUW-25-02
|
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We revisit modular invariance in two-dimensional Carrollian conformal field theories from a geometric perspective. Focusing on the characters of the induced and highest-weight representations of the theory, we show that there are regions of parameter space where the vacuum character dominates in the dual channel. We use this property to zoom into different subsectors of the Carrollian theory. One of them is reminiscent of the Schwarzian sector of a relativistic CFT2 and has a flat space holographic interpretation as O-plane orbifold. It exists only for the highest-weight representation. We prove that for all sectors with vacuum dominance in the dual channel, the specific heat is negative, concurrent with the holographic interpretation of the negative specific heat of asymptotically flat spacetimes with horizons.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 18:00:00 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Aggarwal",
"Ankit",
""
],
[
"Bagchi",
"Arjun",
""
],
[
"Detournay",
"Stephane",
""
],
[
"Grumiller",
"Daniel",
""
],
[
"Riegler",
"Max",
""
],
[
"Simon",
"Joan",
""
]
] |
2508.05834
|
David Jekel
|
David Jekel
|
The unitary group of a $\mathrm{II}_1$ factor is SOT-contractible
|
6 pages; v2, v3 minor corrections
| null | null | null |
math.OA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We show that the unitary group of any SOT-separable $\mathrm{II}_1$ factor $M$, with the strong operator topology, is contractible. Combined with several old results, this implies that the same is true for any SOT-separable von Neumann algebra with no type $\mathrm{I}_n$ direct summands ($n < \infty$). The proof for the $\mathrm{II}_1$-factor case uses regularization via free convolution and Popa's theorem on the existence of approximately free Haar unitaries in $\mathrm{II}_1$ factors.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 20:20:16 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Aug 2025 21:40:33 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 13:01:18 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Jekel",
"David",
""
]
] |
2508.06373
|
Jean-Samuel Roux
|
Jean-Samuel Roux, James M. Cline
|
Dark Sector Electroweak Baryogenesis In Light Of The Galactic Center Excess
|
19 pages, 13 figures; references and Fig. 8 updated in v2
| null | null | null |
hep-ph astro-ph.CO astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
We revisit a model of electroweak baryogenesis that includes a dark matter candidate, and sequesters the new CP violation required to produce the baryon asymmetry in a dark sector. The model can explain the baryon asymmetry, dark matter relic density, and the long-standing excess of gamma rays from the galactic center. The first order electroweak phase transition induced by the new physics can give rise to gravitational waves that may be observed in future experiments. The model predicts dark matter signals in direct detectors, and a significant contribution to the Higgs boson invisible decay width.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 15:04:43 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 15:07:03 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Roux",
"Jean-Samuel",
""
],
[
"Cline",
"James M.",
""
]
] |
2508.07154
|
Shijie Dong
|
Shijie Dong, Zihua Guo, and Kuijie Li
|
Scattering for the Klein-Gordon-Zakharov system in two dimensions
|
V2. Added new Proposition 1.7 on energy cascade of wave component. All comments welcome
| null | null | null |
math.AP math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the Klein-Gordon-Zakharov system in two spatial dimensions, an important model in plasma physics. For small, smooth, and spatially localized initial data, we establish the global existence of solutions and characterize their sharp long-time behavior, including sharp time decay and scattering properties. A particularly interesting phenomenon is that the Klein-Gordon component exhibits modified scattering for certain initial data, while for others it undergoes linear scattering-a dichotomy highlighting delicate long-range interaction effects.
The major obstacles are lack of symmetry and weak decay of the solution in two dimensions. To overcome these, we introduce a novel nonlinear transformation of the wave component and reinterpret the nonlinear coupling as a perturbation of the mass term in the Klein-Gordon equation. The proof employs a combination of physical space and frequency space methods.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 03:05:45 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 11:35:47 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Dong",
"Shijie",
""
],
[
"Guo",
"Zihua",
""
],
[
"Li",
"Kuijie",
""
]
] |
2508.07275
|
Arthur Straube V.
|
Arthur V. Straube, Guillermo Olic\'on M\'endez, Stefanie Winkelmann, Felix H\"ofling, Maximilian Engel
|
Unfolding the geometric structure and multiple timescales of the urea-urease pH oscillator
|
44 pages, 8 figures, submitted
| null | null | null |
math.DS nlin.CD physics.chem-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study a two-variable dynamical system modeling pH oscillations in the urea-urease reaction within giant lipid vesicles -- a problem that intrinsically contains multiple, well-separated timescales. Building on an existing, deterministic formulation via ordinary differential equations, we resolve different orders of magnitude within a small parameter and analyze the system's limit cycle behavior using geometric singular perturbation theory (GSPT). By introducing two different coordinate scalings -- each valid in a distinct region of the phase space -- we resolve the local dynamics near critical fold points, using the extension of GSPT through such singular points due to Krupa and Szmolyan. This framework enables a geometric decomposition of the periodic orbits into slow and fast segments and yields closed-form estimates for the period of oscillation. In particular, we link the existence of such oscillations to an underlying biochemical asymmetry, namely, the differential transport across the vesicle membrane.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 10:11:53 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Straube",
"Arthur V.",
""
],
[
"Méndez",
"Guillermo Olicón",
""
],
[
"Winkelmann",
"Stefanie",
""
],
[
"Höfling",
"Felix",
""
],
[
"Engel",
"Maximilian",
""
]
] |
2508.07335
|
Adan Cabello
|
Ad\'an Cabello
|
The simplest Kochen-Specker set
|
5 pages, 2 figures. New in V2: Complete discussion of the perfect strategy. Reasons why the set is fundamental in quantum theory. Majorana representation
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Kochen-Specker (KS) sets are fundamental in physics. Every time nature produces bipartite correlations attaining the nonsignaling limit, or two parties always win a nonlocal game impossible to always win classically, is because the parties are measuring a KS set. The simplest quantum system in which all these phenomena occur is a pair of three-level systems. However, the simplest KS sets in dimension three known are asymmetrical and require a large number of bases (the current minimum is 16, set by Peres and Penrose). Here we present a KS set that is much more symmetrical and easier to prove than any previous example. It sets a new record for minimum number of bases, 14, and enables us to refute Conjecture 2 in Phys. Rev. Lett. 134, 010201 (2025), setting a new record for qutrit-qutrit perfect strategies with a minimum number of inputs: 5-9. We establish the fundamental nature of this set in quantum theory.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 13:27:02 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 20:47:40 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Cabello",
"Adán",
""
]
] |
2508.07772
|
Ritesh Kumar Pradhan
|
Ritesh Kumar Pradhan, Ralte Lalnuntluanga, and Anjan Giri
|
Monte Carlo study of KDAR $\nu_{\mu}$ charged-current scattering on carbon
| null |
Phys. Rev. D 112, 053002 (2025)
|
10.1103/ygcm-4sw8
| null |
hep-ph hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
The neutrino energy reconstruction is crucial for reducing systematic uncertainties in neutrino oscillation experiments and improving cross-section measurements. Kaon-Decays-At-Rest (KDAR) neutrinos provide a unique opportunity to probe neutrino-nucleus interactions at low energies with a precisely known energy of 235.5 MeV. This work presents Monte Carlo predictions of missing energy due to nuclear effects in the KDAR $\nu_\mu$ charged-current scattering on the carbon nucleus for JSNS$^2$. The predictions from GENIE, NuWro, and GiBUU show a significant impact of nuclear and final state interaction (FSI) models, as well as nucleon removal energy, on missing energy reconstruction. Tuning the hA FSI model improves the predictions from GENIE. It is also observed that the nucleon elastic component of FSI leads to better predictions in NuWro compared to GENIE. NuWro shows the best agreement with data among the generators considered; however, the observed discrepancies with data across all generators highlight the limitations of current theoretical descriptions in the low-energy regime and the need for improvements in Monte Carlo modeling.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 08:57:02 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 05:35:09 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Pradhan",
"Ritesh Kumar",
""
],
[
"Lalnuntluanga",
"Ralte",
""
],
[
"Giri",
"Anjan",
""
]
] |
2508.07834
|
Lisa Bender
|
Mubaris Nadeem, Johannes Zenkert, Lisa Bender, Christian Weber, Madjid Fathi
|
KIRETT: Knowledge-Graph-Based Smart Treatment Assistant for Intelligent Rescue Operations
|
LWDA'23, KIRETT project, University of Siegen, Germany
|
In LWDA (pp. 259-270) 2023
| null | null |
cs.AI cs.ET
|
http://creativecommons.org/licenses/by/4.0/
|
Over the years, the need for rescue operations throughout the world has increased rapidly. Demographic changes and the resulting risk of injury or health disorders form the basis for emergency calls. In such scenarios, first responders are in a rush to reach the patient in need, provide first aid, and save lives. In these situations, they must be able to provide personalized and optimized healthcare in the shortest possible time and estimate the patients condition with the help of freshly recorded vital data in an emergency situation. However, in such a timedependent situation, first responders and medical experts cannot fully grasp their knowledge and need assistance and recommendation for further medical treatments. To achieve this, on the spot calculated, evaluated, and processed knowledge must be made available to improve treatments by first responders. The Knowledge Graph presented in this article as a central knowledge representation provides first responders with an innovative knowledge management that enables intelligent treatment recommendations with an artificial intelligence-based pre-recognition of the situation.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 10:39:15 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 10:56:09 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 09:30:51 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Nadeem",
"Mubaris",
""
],
[
"Zenkert",
"Johannes",
""
],
[
"Bender",
"Lisa",
""
],
[
"Weber",
"Christian",
""
],
[
"Fathi",
"Madjid",
""
]
] |
2508.08005
|
Xiang Li
|
Xiang Li, Shanshan Wang, Chenglong Xiao
|
Learning to Select MCP Algorithms: From Traditional ML to Dual-Channel GAT-MLP
|
10 pages, 6 figures
| null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Extensive experiments and prior studies show that no single maximum clique algorithm consistently performs best across all instances, highlighting the importance of selecting suitable algorithms based on instance features. Through an extensive analysis of relevant studies, it is found that there is a lack of research work concerning algorithm selection oriented toward the Maximum Clique Problem (MCP). In this work, we propose a learning-based framework that integrates both traditional machine learning and graph neural networks to address this gap. We construct a labeled dataset by running four exact MCP algorithms on a diverse collection of graph instances, accompanied by structural and global statistical features extracted from each graph. We first evaluate four conventional classifiers: Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), and K-Nearest Neighbors (KNN), across multiple dataset variants. Experimental results show that RF consistently shows strong performance across metrics and dataset variants, making it a reliable baseline. In addition, feature importance analysis indicates that connectivity and topological structure are strong predictors of algorithm performance. Building on these findings, we develop a dual-channel model named GAT-MLP, which combines a Graph Attention Network (GAT) for local structural encoding with a Multilayer Perceptron (MLP) for global feature modeling. The GAT-MLP model shows strong and consistent performance across all metrics. Our results highlight the effectiveness of dual-channel architectures and the promise of graph neural networks in combinatorial algorithm selection.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 14:09:58 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 14:44:40 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Li",
"Xiang",
""
],
[
"Wang",
"Shanshan",
""
],
[
"Xiao",
"Chenglong",
""
]
] |
2508.08022
|
Roopkatha Banerjee
|
Roopkatha Banerjee, Sampath Koti, Gyanendra Singh, Anirban Chakraborty, Gurunath Gurrala, Bhushan Jagyasi and Yogesh Simmhan
|
Optimizing Federated Learning for Scalable Power-demand Forecasting in Microgrids
| null | null | null | null |
cs.DC cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Real-time monitoring of power consumption in cities and micro-grids through the Internet of Things (IoT) can help forecast future demand and optimize grid operations. But moving all consumer-level usage data to the cloud for predictions and analysis at fine time scales can expose activity patterns. Federated Learning~(FL) is a privacy-sensitive collaborative DNN training approach that retains data on edge devices, trains the models on private data locally, and aggregates the local models in the cloud. But key challenges exist: (i) clients can have non-independently identically distributed~(non-IID) data, and (ii) the learning should be computationally cheap while scaling to 1000s of (unseen) clients. In this paper, we develop and evaluate several optimizations to FL training across edge and cloud for time-series demand forecasting in micro-grids and city-scale utilities using DNNs to achieve a high prediction accuracy while minimizing the training cost. We showcase the benefit of using exponentially weighted loss while training and show that it further improves the prediction of the final model. Finally, we evaluate these strategies by validating over 1000s of clients for three states in the US from the OpenEIA corpus, and performing FL both in a pseudo-distributed setting and a Pi edge cluster. The results highlight the benefits of the proposed methods over baselines like ARIMA and DNNs trained for individual consumers, which are not scalable.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 14:27:26 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 08:46:26 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Banerjee",
"Roopkatha",
""
],
[
"Koti",
"Sampath",
""
],
[
"Singh",
"Gyanendra",
""
],
[
"Chakraborty",
"Anirban",
""
],
[
"Gurrala",
"Gurunath",
""
],
[
"Jagyasi",
"Bhushan",
""
],
[
"Simmhan",
"Yogesh",
""
]
] |
2508.08040
|
Maozhen Zhang
|
Maozhen Zhang, Mengnan Zhao, Bo Wang
|
BadPromptFL: A Novel Backdoor Threat to Prompt-based Federated Learning in Multimodal Models
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Prompt-based tuning has emerged as a lightweight alternative to full fine-tuning in large vision-language models, enabling efficient adaptation via learned contextual prompts. This paradigm has recently been extended to federated learning settings (e.g., PromptFL), where clients collaboratively train prompts under data privacy constraints. However, the security implications of prompt-based aggregation in federated multimodal learning remain largely unexplored, leaving a critical attack surface unaddressed. In this paper, we introduce \textbf{BadPromptFL}, the first backdoor attack targeting prompt-based federated learning in multimodal contrastive models. In BadPromptFL, compromised clients jointly optimize local backdoor triggers and prompt embeddings, injecting poisoned prompts into the global aggregation process. These prompts are then propagated to benign clients, enabling universal backdoor activation at inference without modifying model parameters. Leveraging the contextual learning behavior of CLIP-style architectures, BadPromptFL achieves high attack success rates (e.g., \(>90\%\)) with minimal visibility and limited client participation. Extensive experiments across multiple datasets and aggregation protocols validate the effectiveness, stealth, and generalizability of our attack, raising critical concerns about the robustness of prompt-based federated learning in real-world deployments.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 14:42:44 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 09:32:04 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zhang",
"Maozhen",
""
],
[
"Zhao",
"Mengnan",
""
],
[
"Wang",
"Bo",
""
]
] |
2508.08105
|
Guy Hetzroni
|
Yakir Aharonov and Guy Hetzroni
|
Theoretical Discovery, Experiment, and Controversy in the Aharonov-Bohm Effect: An Oral History Interview
|
26 pages, 3 figures. Forthcoming in EPJH
|
Volume 50, article number 16, (2025)
|
10.1140/epjh/s13129-025-00107-9
| null |
physics.hist-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This oral history interview provides Yakir Aharonov's perspective on the theoretical discovery of the Aharonov-Bohm effect in 1959, during his PhD studies in Bristol with David Bohm, the reception of the effect, the efforts to test it empirically (up to Tonomura's experiment), and some of the debates regarding the existence of the effect and its interpretation. The interview also discusses related later developments until the 1980s, including modular momentum and Berry's phase. It includes recollections from meetings with Werner Heisenberg, Richard Feynman, and Chen-Ning Yang, also mentioning John Bell, Robert Chambers, Werner Ehrenberg, Sir Charles Frank, Wendell Furry, Gunnar K\"all\'en, Maurice Pryce, Nathan Rosen, John Wheeler, and Eugene Wigner.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 15:46:55 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Aharonov",
"Yakir",
""
],
[
"Hetzroni",
"Guy",
""
]
] |
2508.08624
|
Shuai Wang
|
Chenxuan Liu, He Li, Zongze Li, Shuai Wang, Wei Xu, Kejiang Ye, Derrick Wing Kwan Ng, Chengzhong Xu
|
Communication Efficient Robotic Mixed Reality with Gaussian Splatting Cross-Layer Optimization
|
14 pages, 18 figures, to appear in IEEE Transactions on Cognitive Communications and Networking
| null | null | null |
cs.RO cs.IT math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Realizing low-cost communication in robotic mixed reality (RoboMR) systems presents a challenge, due to the necessity of uploading high-resolution images through wireless channels. This paper proposes Gaussian splatting (GS) RoboMR (GSMR), which enables the simulator to opportunistically render a photo-realistic view from the robot's pose by calling ``memory'' from a GS model, thus reducing the need for excessive image uploads. However, the GS model may involve discrepancies compared to the actual environments. To this end, a GS cross-layer optimization (GSCLO) framework is further proposed, which jointly optimizes content switching (i.e., deciding whether to upload image or not) and power allocation (i.e., adjusting to content profiles) across different frames by minimizing a newly derived GSMR loss function. The GSCLO problem is addressed by an accelerated penalty optimization (APO) algorithm that reduces computational complexity by over $10$x compared to traditional branch-and-bound and search algorithms. Moreover, variants of GSCLO are presented to achieve robust, low-power, and multi-robot GSMR. Extensive experiments demonstrate that the proposed GSMR paradigm and GSCLO method achieve significant improvements over existing benchmarks on both wheeled and legged robots in terms of diverse metrics in various scenarios. For the first time, it is found that RoboMR can be achieved with ultra-low communication costs, and mixture of data is useful for enhancing GS performance in dynamic scenarios.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 04:27:12 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 04:32:06 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Liu",
"Chenxuan",
""
],
[
"Li",
"He",
""
],
[
"Li",
"Zongze",
""
],
[
"Wang",
"Shuai",
""
],
[
"Xu",
"Wei",
""
],
[
"Ye",
"Kejiang",
""
],
[
"Ng",
"Derrick Wing Kwan",
""
],
[
"Xu",
"Chengzhong",
""
]
] |
2508.08662
|
N. Rieger
|
N. E. Rieger
|
Embedding Signature-Changing Manifolds: A Braneworld and Kaluza-Klein Perspective
|
In the revised version, some LaTeX compilation errors have been fixed, and the proof of Proposition 1.13 has been corrected. 29 pages, 3 figures
| null | null | null |
math.DG gr-qc math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate a class of semi-Riemannian manifolds characterized by smooth metric signature changes with a transverse radical. This class includes spacetimes relevant to cosmological models such as the Hartle-Hawking "no boundary" proposal, where a Riemannian manifold transitions smoothly into a Lorentzian spacetime without boundaries or singularities. For this class, we prove the existence of global isometric embeddings into higher-dimensional pseudo-Euclidean spaces. We then strengthen this result by demonstrating that a specific type of global isometric embedding, which we term an $\mathcal{H}$-global embedding, also exists into both Minkowski space and Misner space. For the canonical $n$-dimensional signature-changing model, we explicitly construct a full global isometric embedding into $(n+1)$-dimensional Minkowski and Misner spaces, a significantly stronger result than an $\mathcal{H}$-global embedding for this specific case.
This embedding framework provides new geometric tools for studying signature change and braneworlds through the geometry of submanifolds embedded in a bulk, thus presenting a mathematically well-defined approach to these phenomena.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 06:01:04 GMT"
},
{
"version": "v2",
"created": "Wed, 13 Aug 2025 01:10:43 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 23:24:25 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Rieger",
"N. E.",
""
]
] |
2508.08753
|
Abolfazl Tarizadeh
|
Abolfazl Tarizadeh
|
Splittings in a complete local ring and decomposition its group of units
|
14 pages
| null | null | null |
math.AC math.AG math.GR math.RA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Let $(R,M,k)$ be a complete local ring (not necessarily Noetherian). Then we prove that the natural surjective ring map $R\rightarrow k$ admits a splitting if and only if $\Char(R)=\Char(k)$. In addition, if $\Char(R)\neq\Char(k)$, then we prove that the natural surjective map between the groups of units $R^{\ast}\rightarrow k^{\ast}$ admits a splitting. \\ We also show with an example that the above theorem is in its full generality. As an application of this theorem, we show that the following short exact sequence of Abelian groups: $$\xymatrix{1\ar[r]&1+M\ar[r]& R^{\ast}\ar[r]&k^{\ast} \ar[r]&1}$$ is always split. Next, we show with an example that the above exact sequence does not split for many incomplete local rings.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 08:57:49 GMT"
},
{
"version": "v2",
"created": "Sun, 24 Aug 2025 08:07:57 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 09:27:31 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Tarizadeh",
"Abolfazl",
""
]
] |
2508.09284
|
Eduardo Bautista
|
Eddie Bautista, Eric I. Corwin
|
Numerically Discovered Inherent States are Always Protocol Dependent in Jammed Packings
|
5 pages, 3 figures
| null | null | null |
cond-mat.soft
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The energy landscape for soft sphere packings exists in a high-dimensional space and plays host to an astronomical number of local minima in a hierarchical and ultrametric arrangement. Each point in the landscape is a configuration that can be unambiguously mapped to its inherent state, defined as the local minimum that the configuration will flow to under perfectly overdamped continuous dynamics. Typically, discrete in time dynamics are used to computationally find local minima, but it is not known whether these algorithms are capable of reliably finding inherent states. Here, we use steepest descent dynamics to find the distribution of the largest time step, $\delta_\textrm{best}$, which finds the inherent state. We find that for systems of $N$ particles, $\delta_\textrm{best}$ is approximately proportional to $N^{-3}$, and weakly dependent on d and $\varphi$. We argue that the proportionality is due to saddle points in the energy landscape. Our results suggest that it is impossible, in practice, to reliably find inherent states for systems of about 64 particles or more.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 18:43:52 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 06:53:52 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bautista",
"Eddie",
""
],
[
"Corwin",
"Eric I.",
""
]
] |
2508.09311
|
Zongyu Li
|
Zongyu Li, Mark Steel, Zhiyong Zhang
|
Modelling Skewed and Heavy-Tailed Errors in Bayesian Mediation Analysis
| null | null | null | null |
stat.ME
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Traditional mediation models in both the frequentist and Bayesian frameworks typically assume normality of the error terms. Violations of this assumption can impair the estimation and hypothesis testing of the mediation effect in conventional approaches. This study addresses the non-normality issue by explicitly modelling skewed and heavy-tailed error terms within the Bayesian mediation framework. Building on the work of Fern\'andez and Steel (1998), this study introduces a novel family of distributions, termed the Centred Two-Piece Student $t$ Distribution (CTPT). The new distribution incorporates a skewness parameter into the Student t distribution and centres it to have a mean of zero, enabling flexible modelling of error terms in Bayesian regression and mediation analysis. A class of standard improper priors is employed, and conditions for the existence of the posterior distribution and posterior moments are established, while enabling inference on both skewness and tail parameters. Simulation studies are conducted to examine parameter recovery accuracy and statistical power in testing mediation effects. Compared to traditional Bayesian and frequentist methods, particularly bootstrap-based approaches, our method gives greater statistical power when correctly specified, while maintaining robustness against model misspecification. The application of the proposed approach is illustrated through real data analysis. Additionally, we have developed an R package FlexBayesMed to implement our methods in linear regression and mediation analysis, available at https://github.com/Zongyu-Li/FlexBayesMed.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 19:53:33 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 04:40:22 GMT"
},
{
"version": "v3",
"created": "Wed, 03 Sep 2025 04:08:59 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Li",
"Zongyu",
""
],
[
"Steel",
"Mark",
""
],
[
"Zhang",
"Zhiyong",
""
]
] |
2508.09600
|
Xuelong Geng
|
Xuelong Geng, Qijie Shao, Hongfei Xue, Shuiyuan Wang, Hanke Xie, Zhao Guo, Yi Zhao, Guojian Li, Wenjie Tian, Chengyou Wang, Zhixian Zhao, Kangxiang Xia, Ziyu Zhang, Zhennan Lin, Tianlun Zuo, Mingchen Shao, Yuang Cao, Guobin Ma, Longhao Li, Yuhang Dai, Dehui Gao, Dake Guo, Lei Xie
|
OSUM-EChat: Enhancing End-to-End Empathetic Spoken Chatbot via Understanding-Driven Spoken Dialogue
| null | null | null | null |
cs.SD
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Empathy is crucial in enabling natural interactions within spoken dialogue systems, allowing machines to recognize and respond appropriately to paralinguistic cues such as age, gender, and emotion. Recent advancements in end-to-end speech language models, which unify speech understanding and generation, provide promising solutions. However, several challenges persist, including an over-reliance on large-scale dialogue datasets, insufficient extraction of paralinguistic cues vital for conveying empathy, and the lack of empathy-specific datasets and evaluation frameworks. To address these issues, we introduce OSUM-EChat, an open-source, end-to-end spoken dialogue system designed to enhance empathetic interactions, particularly in resource-limited settings. OSUM-EChat introduces two key innovations: (1) a three-stage understanding-driven spoken dialogue training strategy that extends the capabilities of a large speech understanding model to spoken dialogue tasks, and (2) a linguistic-paralinguistic dual thinking mechanism that integrates paralinguistic understanding through a chain of thought with dialogue generation, enabling the system to produce more empathetic responses. This approach reduces reliance on large-scale dialogue datasets while maintaining high-quality empathetic interactions. Additionally, we introduce the EChat-200K dataset, a rich corpus of empathetic speech-to-speech dialogues, and the EChat-eval benchmark, a comprehensive framework for evaluating the empathetic capabilities of dialogue systems. Experimental results demonstrate that OSUM-EChat outperforms end-to-end spoken dialogue models regarding empathetic responsiveness, validating its effectiveness.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 08:30:14 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 13:33:34 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Geng",
"Xuelong",
""
],
[
"Shao",
"Qijie",
""
],
[
"Xue",
"Hongfei",
""
],
[
"Wang",
"Shuiyuan",
""
],
[
"Xie",
"Hanke",
""
],
[
"Guo",
"Zhao",
""
],
[
"Zhao",
"Yi",
""
],
[
"Li",
"Guojian",
""
],
[
"Tian",
"Wenjie",
""
],
[
"Wang",
"Chengyou",
""
],
[
"Zhao",
"Zhixian",
""
],
[
"Xia",
"Kangxiang",
""
],
[
"Zhang",
"Ziyu",
""
],
[
"Lin",
"Zhennan",
""
],
[
"Zuo",
"Tianlun",
""
],
[
"Shao",
"Mingchen",
""
],
[
"Cao",
"Yuang",
""
],
[
"Ma",
"Guobin",
""
],
[
"Li",
"Longhao",
""
],
[
"Dai",
"Yuhang",
""
],
[
"Gao",
"Dehui",
""
],
[
"Guo",
"Dake",
""
],
[
"Xie",
"Lei",
""
]
] |
2508.09777
|
Shima Mohammadi
|
Shima Mohammadi, Mohsen Jenadeleh, Michela Testolina, Jon Sneyers, Touradj Ebrahimi, Dietmar Saupe, Jo\~ao Ascenso
|
In-place Double Stimulus Methodology for Subjective Assessment of High Quality Images
|
6 pages, 5 figures, Accepted at European Workshop on Visual Information Processing
| null | null | null |
cs.MM
|
http://creativecommons.org/publicdomain/zero/1.0/
|
This paper introduces a novel double stimulus subjective assessment methodology for the evaluation of high quality images to address the limitations of existing protocols in detecting subtle perceptual differences. The In-place Double Stimulus Quality Scale (IDSQS) allows subjects to alternately view a reference and a distorted image at the same spatial location, facilitating a more intuitive detection of differences in quality, especially at high to visually lossless quality levels. A large-scale crowdsourcing study employing this methodology was conducted, generating a comprehensive public dataset to evaluate perceived image quality across several compression algorithms and distortion levels. An additional contribution is the modeling of quality scores using a Beta distribution, allowing for the assessment of variability and subject consistency. Our findings demonstrate the effectiveness of the IDSQS methodology in achieving high correlation with more precise subjective evaluation benchmarks. The dataset, subjective data, and graphical user interface developed for this study are publicly available at https://github.com/shimamohammadi/IDSQS
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 12:59:33 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 15:22:46 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Mohammadi",
"Shima",
""
],
[
"Jenadeleh",
"Mohsen",
""
],
[
"Testolina",
"Michela",
""
],
[
"Sneyers",
"Jon",
""
],
[
"Ebrahimi",
"Touradj",
""
],
[
"Saupe",
"Dietmar",
""
],
[
"Ascenso",
"João",
""
]
] |
2508.09853
|
Jide Alaga
|
Tegan McCaslin, Jide Alaga, Samira Nedungadi, Seth Donoughe, Tom Reed, Rishi Bommasani, Chris Painter, and Luca Righetti
|
STREAM (ChemBio): A Standard for Transparently Reporting Evaluations in AI Model Reports
|
47 pages, 1 figure. Includes appendices and reporting template
| null | null | null |
cs.CY cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Evaluations of dangerous AI capabilities are important for managing catastrophic risks. Public transparency into these evaluations - including what they test, how they are conducted, and how their results inform decisions - is crucial for building trust in AI development. We propose STREAM (A Standard for Transparently Reporting Evaluations in AI Model Reports), a standard to improve how model reports disclose evaluation results, initially focusing on chemical and biological (ChemBio) benchmarks. Developed in consultation with 23 experts across government, civil society, academia, and frontier AI companies, this standard is designed to (1) be a practical resource to help AI developers present evaluation results more clearly, and (2) help third parties identify whether model reports provide sufficient detail to assess the rigor of the ChemBio evaluations. We concretely demonstrate our proposed best practices with "gold standard" examples, and also provide a three-page reporting template to enable AI developers to implement our recommendations more easily.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 14:36:36 GMT"
},
{
"version": "v2",
"created": "Wed, 03 Sep 2025 12:52:33 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"McCaslin",
"Tegan",
""
],
[
"Alaga",
"Jide",
""
],
[
"Nedungadi",
"Samira",
""
],
[
"Donoughe",
"Seth",
""
],
[
"Reed",
"Tom",
""
],
[
"Bommasani",
"Rishi",
""
],
[
"Painter",
"Chris",
""
],
[
"Righetti",
"Luca",
""
]
] |
2508.10059
|
Yueke Zhang
|
Yueke Zhang, Yifan Zhang, Kevin Leach, Yu Huang
|
CodeGrad: Integrating Multi-Step Verification with Gradient-Based LLM Refinement
|
6 Pages
| null | null | null |
cs.SE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
While Large Language Models (LLMs) have demonstrated remarkable capabilities in code generation, they often produce solutions that lack guarantees of correctness, robustness, and efficiency. This limitation is particularly acute in domains requiring strict constraints. CodeGrad introduces a principled framework that integrates rigorous verification techniques directly into an iterative LLM-based generation loop. It uniquely treats code as a differentiable variable, converting structured feedback and mathematical constraints into a textual pseudo-gradient. This gradient guides the model to iteratively refine solutions, ensuring they are not only functional but also robust and mathematically justified.
We evaluate CodeGrad on the HumanEval, HumanEval+, and LiveCodeBench benchmarks. Our implementation outperforms strong baselines, achieving an absolute improvement of up to 27% on HumanEval and a 41% relative improvement on the challenging LiveCodeBench V6. StructuredGrad generates mathematically justified code that is robust and efficient, paving the way for reliable AI-assisted software development in high-stakes applications.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 22:03:54 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 20:11:20 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zhang",
"Yueke",
""
],
[
"Zhang",
"Yifan",
""
],
[
"Leach",
"Kevin",
""
],
[
"Huang",
"Yu",
""
]
] |
2508.10198
|
Lucas Draper
|
Lucas Draper
|
Digital Contact Tracing: Examining the Effects of Understanding and Release Organization on Public Trust
| null | null | null | null |
cs.CY
|
http://creativecommons.org/licenses/by/4.0/
|
Contact tracing has existed in various forms for a very long time. With the rise of COVID-19, the concept has become increasingly important to help slow the spread of the virus. One approach to modernizing contact tracing is to introduce applications that detect all close contacts without individuals having to interact knowingly. 101 United States adults were surveyed in June of 2022 regarding their perceptions and trust of COVID-19 contact tracing applications. We see no definitive correlation between an individual's understanding of privacy protection procedures for contact tracing applications and their willingness to trust such an application. We also see that the release of the application by a private entity like Google-Apple or by a public entity like the United States Federal Government has no significant correlation with a person's trust in the application.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 21:13:40 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 18:12:51 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Draper",
"Lucas",
""
]
] |
2508.10847
|
Hongyi Yu
|
Zijian Ma, Hongyi Yu
|
Interaction enhanced inter-site hoppings for holons and interlayer exciton insulators in moir\'e correlated insulators
| null |
Chinese Phys. B 34, 097303 (2025)
|
10.1088/1674-1056/adcb9b
| null |
cond-mat.mes-hall cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
In moir\'e-patterned van der Waals structures of transition metal dichalcogenides, correlated insulators can form under integer and fractional fillings, whose transport properties are governed by various quasiparticle excitations including holons, doublons and interlayer exciton insulators. Here we theoretically investigate the nearest-neighbor inter-site hoppings of holons and interlayer exciton insulators. Our analysis indicates that these hopping strengths are significantly enhanced compared to that of a single carrier. The underlying mechanism can be attributed to the strong Coulomb interaction between carriers at different sites. For the interlayer exciton insulator consisting of a holon and a carrier in different layers, we have also obtained its effective Bohr radius and energy splitting between the ground and first-excited states.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 17:18:24 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ma",
"Zijian",
""
],
[
"Yu",
"Hongyi",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.