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
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2407.17247 | Alessio Giarnetti | Alessio Giarnetti, Simone Marciano, Davide Meloni | Exploring new physics with DUNE high energy flux: the case of Lorentz Invariance Violation, Large Extra Dimensions and Long Range Forces | 32 pages, 7 figures | Universe 2024, 10(9), 357 | 10.3390/universe10090357 | null | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | DUNE is a next-generation long-baseline neutrino oscillation experiment. It is expected to measure with an unprecedent precision the atmospheric oscillation parameters, including the CP-violating phase $\delta_{CP}$. Moreover, several studies have suggested that its unique features should allow DUNE to probe several new physics scenarios. In this work, we explore the performances of the DUNE far detector in constraining new physics if a high-energy neutrino flux is employed (HE-DUNE). We take into account three different scenarios: Lorentz Invariance Violation (LIV), Long Range Forces (LRF) and Large Extra Dimensions (LED). Our results show that HE-DUNE should be able to set bounds competitive to the current ones and, in particular, it can outperform the standard DUNE capabilities in constraining CPT-even LIV parameters and the compactification radius $R_{ED}$ of the LED model. | [
{
"version": "v1",
"created": "Wed, 24 Jul 2024 13:05:34 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Giarnetti",
"Alessio",
""
],
[
"Marciano",
"Simone",
""
],
[
"Meloni",
"Davide",
""
]
] |
2407.17621 | Juan Manuel Romero | Juan M. Romero, Emiliano Montoya-Gonzalez and Oscar Velazquez-Alvarado | Quantum Entanglement, Quantum Teleportation, Multilinear Polynomials and Geometry | 16 pages, 2 figures. Minor corrections. Comments welcome | International Journal of Geometric Methods in Modern Physics 2025 | 10.1142/S0219887825502081 | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | We show that quantum entanglement states are associated with multilinear polynomials that cannot be factored. By using these multilinear polynomials, we propose a geometric representation for entanglement states. In particular, we show that the Bell's states are associated with non-factorable real multilinear polynomial, which can be represented geometrically by three-dimensional surfaces. Furthermore, in this framework, we show that a quantum circuit can be seen as a geometric transformations of plane geometry. This phenomenon is analogous to gravity, where matter curves space-time. In addition, we show an analogy between quantum teleportation and operations involving multilinear polynomials. | [
{
"version": "v1",
"created": "Wed, 24 Jul 2024 20:25:48 GMT"
},
{
"version": "v2",
"created": "Thu, 22 Aug 2024 22:19:03 GMT"
},
{
"version": "v3",
"created": "Fri, 18 Oct 2024 20:47:44 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Romero",
"Juan M.",
""
],
[
"Montoya-Gonzalez",
"Emiliano",
""
],
[
"Velazquez-Alvarado",
"Oscar",
""
]
] |
2407.18403 | Donato Vasquez-Varas | Karl Kunisch, Donato V\'asquez-Varas | Convergence of machine learning methods for feedback control laws: averaged feedback learning scheme and data driven methods | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | This work addresses the synthesis of optimal feedback control laws via machine learning. In particular, the Averaged Feedback Learning Scheme (AFLS) and a data driven method are considered. Hypotheses for each method ensuring the convergence of the evaluation of the objective function of the underlying control problem at the obtained feedback-laws towards the optimal value function are provided. These hypotheses are connected to the regularity of the value function and the stability of the dynamics. In the case of AFLS these hypotheses only require H\"older continuity of the value function, whereas for the data driven method the value function must be at least $C^2$. It is demonstrated that these methods are connected via their optimality conditions. Additionally, numerical experiments are provided by applying both methods to a family control problems, parameterized by a positive real number which controls the regularity of the value function. For small parameters the value function is smooth and in contrast for large parameters it is non-differentiable, but semi-concave. The results of the experiments indicate that both methods have a similar performance for the case that the value function is smooth. On the other hand, if the value function is not differentiable, AFLS has a better performance which is consistent with the obtained convergence results. | [
{
"version": "v1",
"created": "Thu, 25 Jul 2024 21:35:12 GMT"
},
{
"version": "v2",
"created": "Fri, 09 May 2025 12:54:37 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 14:50:54 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Kunisch",
"Karl",
""
],
[
"Vásquez-Varas",
"Donato",
""
]
] |
2407.19093 | Amit Vashisht | Amit Vashisht, Ivan Amelio, Laurens Vanderstraeten, Georg M. Bruun, Oriana K. Diessel, Nathan Goldman | Chiral polaron formation on the edge of topological quantum matter | 28 pages, 7 Sections and 5 Appendices | Nat Commun 16, 4918 (2025) | 10.1038/s41467-025-60166-w | null | cond-mat.quant-gas cond-mat.mes-hall cond-mat.str-el quant-ph | http://creativecommons.org/licenses/by/4.0/ | Immersing a mobile impurity in a quantum many-body environment can reveal fundamental properties of the background medium, hence providing a powerful probe of quantum matter. This approach is particularly intriguing when considering media with exotic properties, such as strongly-correlated phases and topological states of matter. In this work, we study the dressing of a mobile impurity interacting with a chiral mode, as provided by the edge of topological quantum matter. The resulting ''chiral polaron'' is characterized by an asymmetric spectral function, which reflects the chirality and group velocity of the topological edge mode and the drag experienced by the mobile impurity. We first build our theoretical understanding from an effective one-dimensional chiral model, which captures the hallmark signatures of the chiral polaron. We then demonstrate how this simple picture extends to realistic models of integer and fractional Chern insulator states, by adapting tensor-network methods to polaron spectroscopy. Injecting mobile impurities on the edge of topological quantum matter is shown to be a powerful tool to probe exotic edge properties, particularly suitable for cold-atom experiments. | [
{
"version": "v1",
"created": "Fri, 26 Jul 2024 21:18:39 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Vashisht",
"Amit",
""
],
[
"Amelio",
"Ivan",
""
],
[
"Vanderstraeten",
"Laurens",
""
],
[
"Bruun",
"Georg M.",
""
],
[
"Diessel",
"Oriana K.",
""
],
[
"Goldman",
"Nathan",
""
]
] |
2407.20114 | Mikel Williams-Lekuona | Mikel Williams-Lekuona and Georgina Cosma | FiCo-ITR: bridging fine-grained and coarse-grained image-text retrieval for comparative performance analysis | 19 pages, submitted to International Journal of Multimedia Information Retrieval | null | 10.1007/s13735-025-00368-6 | null | cs.IR cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the field of Image-Text Retrieval (ITR), recent advancements have leveraged large-scale Vision-Language Pretraining (VLP) for Fine-Grained (FG) instance-level retrieval, achieving high accuracy at the cost of increased computational complexity. For Coarse-Grained (CG) category-level retrieval, prominent approaches employ Cross-Modal Hashing (CMH) to prioritise efficiency, albeit at the cost of retrieval performance. Due to differences in methodologies, FG and CG models are rarely compared directly within evaluations in the literature, resulting in a lack of empirical data quantifying the retrieval performance-efficiency tradeoffs between the two. This paper addresses this gap by introducing the \texttt{FiCo-ITR} library, which standardises evaluation methodologies for both FG and CG models, facilitating direct comparisons. We conduct empirical evaluations of representative models from both subfields, analysing precision, recall, and computational complexity across varying data scales. Our findings offer new insights into the performance-efficiency trade-offs between recent representative FG and CG models, highlighting their respective strengths and limitations. These findings provide the foundation necessary to make more informed decisions regarding model selection for specific retrieval tasks and highlight avenues for future research into hybrid systems that leverage the strengths of both FG and CG approaches. | [
{
"version": "v1",
"created": "Mon, 29 Jul 2024 15:44:22 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Williams-Lekuona",
"Mikel",
""
],
[
"Cosma",
"Georgina",
""
]
] |
2407.20624 | Gregor Skok | Gregor Skok and Lloren\c{c} Lled\'o | Spatial verification of global precipitation forecasts | Final peer-reviewed version of the manuscript is published in Quarterly Journal of the Royal Meteorological Society | null | 10.1002/qj.5006 | null | physics.ao-ph | http://creativecommons.org/licenses/by/4.0/ | Verification of global high-resolution precipitation forecasts is challenging. Spatial verification techniques address some shortcomings of traditional verification. However most existing methods do not account for the non-planar geometry of a global domain, or their computational complexity is too large for global assessments. We present an adaptation of the recently developed Precipitation Attribution Distance (PAD) metric, designed for verifying precipitation, enabling its use on the Earth's spherical geometry. PAD estimates the magnitude of location errors in the forecasts employing the mathematical theory of Optimal Transport, as it provides a close upper bound for the Wasserstein distance. The method is fast and flexible with time complexity $O(n \log(n))$. Its behavior is analyzed using idealized cases and 7 years of operational global deterministic 6-hourly precipitation forecasts from the Integrated Forecasting System (IFS) of the European Centre for Medium-Range Weather Forecasts. The summary results for the whole period show how location errors in the IFS model grow steadily with increasing lead time for all analyzed regions. Moreover, by examining the time-series of the results, we can determine the trends in the score's value and identify the regions where there is a statistically significant improvement (or worsening) of the forecast performance. The results can also be analyzed separately for different intensities of precipitation. Overall, the PAD provides meaningful results for estimating location errors in global high-resolution precipitation forecasts at an affordable computational cost. | [
{
"version": "v1",
"created": "Tue, 30 Jul 2024 07:59:39 GMT"
},
{
"version": "v2",
"created": "Thu, 06 Feb 2025 12:35:33 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 08:19:10 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Skok",
"Gregor",
""
],
[
"Lledó",
"Llorenç",
""
]
] |
2407.21054 | Khai Le-Duc | Khai-Nguyen Nguyen, Khai Le-Duc, Bach Phan Tat, Duy Le, Long Vo-Dang, Truong-Son Hy | Sentiment Reasoning for Healthcare | ACL 2025 (Oral) | null | null | null | cs.CL cs.AI cs.LG cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Transparency in AI healthcare decision-making is crucial. By incorporating rationales to explain reason for each predicted label, users could understand Large Language Models (LLMs)'s reasoning to make better decision. In this work, we introduce a new task - Sentiment Reasoning - for both speech and text modalities, and our proposed multimodal multitask framework and the world's largest multimodal sentiment analysis dataset. Sentiment Reasoning is an auxiliary task in sentiment analysis where the model predicts both the sentiment label and generates the rationale behind it based on the input transcript. Our study conducted on both human transcripts and Automatic Speech Recognition (ASR) transcripts shows that Sentiment Reasoning helps improve model transparency by providing rationale for model prediction with quality semantically comparable to humans while also improving model's classification performance (+2% increase in both accuracy and macro-F1) via rationale-augmented fine-tuning. Also, no significant difference in the semantic quality of generated rationales between human and ASR transcripts. All code, data (five languages - Vietnamese, English, Chinese, German, and French) and models are published online: https://github.com/leduckhai/Sentiment-Reasoning | [
{
"version": "v1",
"created": "Wed, 24 Jul 2024 12:07:54 GMT"
},
{
"version": "v2",
"created": "Mon, 30 Sep 2024 19:51:09 GMT"
},
{
"version": "v3",
"created": "Fri, 11 Oct 2024 05:43:19 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 07:08:34 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Nguyen",
"Khai-Nguyen",
""
],
[
"Le-Duc",
"Khai",
""
],
[
"Tat",
"Bach Phan",
""
],
[
"Le",
"Duy",
""
],
[
"Vo-Dang",
"Long",
""
],
[
"Hy",
"Truong-Son",
""
]
] |
2408.00190 | Sahabub Jahedi | Sahabub Jahedi, Abhik Sarkar | Exploring Optimal Sensitivity of Lepton Flavor Violating Effective Couplings at the $e^+e^-$ Colliders | 26 pages, 13 figures, 10 tables; typos fixed; Accepted for publication in PRD | Phys.Rev.D 110 (2024) 9, 095021 | 10.1103/PhysRevD.110.095021 | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | We analyze lepton flavor violation (LFV) using the Standard Model Effective Field Theory (SMEFT) framework at the future lepton colliders. Our focus is on the associated production of tau lepton with electron/muon at the electron-positron ($e^+e^-$) colliders, related to four-Fermi SMEFT effective operators. In accordance with the upper limits on effective couplings from lepton flavor violating tau decays, we conduct a cut-based analysis to achieve sufficient signal significance. We utilize the optimal observable technique (OOT) to estimate the optimal sensitivity of the effective couplings. The impact of electron beam polarization and the interplay of signal and background in enhancing the optimal sensitivity of the effective couplings are discussed in detail. We find that the sensitivity of flavor-violating effective couplings is enhanced by order of one for 3 TeV center-of-mass (CM) energy and 1000 $\rm{fb}^{-1}$ integrated luminosity at the $e^+e^-$ colliders. | [
{
"version": "v1",
"created": "Wed, 31 Jul 2024 23:10:54 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Oct 2024 11:19:36 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Jahedi",
"Sahabub",
""
],
[
"Sarkar",
"Abhik",
""
]
] |
2408.02578 | Oleksandr Stashko | R. A. Konoplya, O. S. Stashko | Probing the Effective Quantum Gravity via Quasinormal Modes and Shadows of Black Holes | new section added | Phys. Rev. D 111, 104055, 2025 | 10.1103/PhysRevD.111.104055 | null | gr-qc | http://creativecommons.org/licenses/by/4.0/ | Two quantum-corrected black hole models have recently been proposed within the Hamiltonian constraints approach to quantum gravity, maintaining general covariance \cite{Zhang:2024khj}. We have studied the quasinormal spectra of these black holes using four methods: the higher-order WKB approach with Pad\'e approximants, time-domain integration, Frobenius, and pseudospectral methods. The Frobenius method, in particular, allows us to determine precise values of the frequencies, including the overtones. The two models differ in their choice of quantum parameter $\xi$, and we can distinguish them by their quasinormal spectra. In the first model, increasing the quantum parameter results in higher real oscillation frequencies and damping rates of the fundamental mode. In contrast, the second model shows a decrease in the oscillation frequency of the least-damped mode when the quantum parameter is introduced. We have shown that, while the fundamental mode changes relatively gradually with the quantum parameter, the first few overtones deviate from their Schwarzschild limits at an increasing rate. This results in a qualitatively new behavior: the real parts of the frequencies of the first and higher overtones tend to zero as the quantum parameter increases. In addition to the branch of modes that are perturbative in the quantum parameter, we observe some non-perturbative modes at moderate values of the quantum parameter.
Additionally, we have calculated the radii of the shadows cast by these black holes and discussed possible constraints based on observations of Sgt $A^*$. As a byproduct, we tested the method of calculating quasinormal modes of this kind based on a recent agnostic parametrization and showed that while the parametrized formalism could be used for estimating the fundamental mode at small values of the coupling, it is insufficient even for the lowest overtones. | [
{
"version": "v1",
"created": "Mon, 05 Aug 2024 15:55:47 GMT"
},
{
"version": "v2",
"created": "Sun, 25 Aug 2024 21:06:50 GMT"
},
{
"version": "v3",
"created": "Wed, 27 Nov 2024 00:11:38 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Konoplya",
"R. A.",
""
],
[
"Stashko",
"O. S.",
""
]
] |
2408.04017 | Wojciech Jan Jankowski | Aneesh Agarwal, Wojciech J. Jankowski, Daniel Bennett, Robert-Jan Slager | Shift photocurrent vortices from topological polarization textures | 10+20 pages, 4+11 figures | npj 2D Mater. Appl. 9, 43 (2025) | 10.1038/s41699-025-00552-6 | null | cond-mat.mes-hall cond-mat.mtrl-sci physics.optics quant-ph | http://creativecommons.org/licenses/by/4.0/ | Following the recent interest in van der Waals (vdW) ferroelectrics, topologically nontrivial polar structures have been predicted to form in twisted bilayers. However, these structures have proven difficult to observe experimentally. We propose that these textures may be probed optically by showing that topological polarization textures result in exotic nonlinear optical responses. We derive this relationship analytically using non-Abelian Berry connections and a quantum-geometric framework, supported by tight-binding and first-principles calculations. For the case of moir\'e materials without centrosymmetry, which form networks of polar merons and antimerons, the shift photoconductivity forms a vortex-like structure in real space. For a range of frequencies where transitions between topologically trivial bands occur at the Brillouin zone edge, the shift photocurrents are antiparallel to the in-plane electronic polarization field. Our findings highlight the interplay between complex polarization textures and nonlinear optical responses in vdW materials and provide a sought-after strategy for their experimental detection. | [
{
"version": "v1",
"created": "Wed, 07 Aug 2024 18:04:12 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 21:30:07 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Agarwal",
"Aneesh",
""
],
[
"Jankowski",
"Wojciech J.",
""
],
[
"Bennett",
"Daniel",
""
],
[
"Slager",
"Robert-Jan",
""
]
] |
2408.04461 | Tobias Bernecker | Tobias Bernecker, Ghalia Rehawi, Francesco Paolo Casale, Janine Knauer-Arloth, Annalisa Marsico | Random Walk Diffusion for Efficient Large-Scale Graph Generation | Published as a paper at Transactions on Machine Learning Research (TMLR) in March 2025 | null | null | null | cs.LG cs.SI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Graph generation addresses the problem of generating new graphs that have a data distribution similar to real-world graphs. While previous diffusion-based graph generation methods have shown promising results, they often struggle to scale to large graphs. In this work, we propose ARROW-Diff (AutoRegressive RandOm Walk Diffusion), a novel random walk-based diffusion approach for efficient large-scale graph generation. Our method encompasses two components in an iterative process of random walk sampling and graph pruning. We demonstrate that ARROW-Diff can scale to large graphs efficiently, surpassing other baseline methods in terms of both generation time and multiple graph statistics, reflecting the high quality of the generated graphs. | [
{
"version": "v1",
"created": "Thu, 08 Aug 2024 13:42:18 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 12:25:24 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Bernecker",
"Tobias",
""
],
[
"Rehawi",
"Ghalia",
""
],
[
"Casale",
"Francesco Paolo",
""
],
[
"Knauer-Arloth",
"Janine",
""
],
[
"Marsico",
"Annalisa",
""
]
] |
2408.04520 | Daniil Boiko | Daniil A. Boiko, Thiago Resch\"utzegger, Benjamin Sanchez-Lengeling, Samuel M. Blau, Gabe Gomes | Advancing Molecular Machine Learning Representations with Stereoelectronics-Infused Molecular Graphs | 23 pages, 6 figures | Nat Mach Intell 7 (2025) 771-781 | 10.1038/s42256-025-01031-9 | null | cs.LG physics.chem-ph | http://creativecommons.org/licenses/by/4.0/ | Molecular representation is a critical element in our understanding of the physical world and the foundation for modern molecular machine learning. Previous molecular machine learning models have employed strings, fingerprints, global features, and simple molecular graphs that are inherently information-sparse representations. However, as the complexity of prediction tasks increases, the molecular representation needs to encode higher fidelity information. This work introduces a novel approach to infusing quantum-chemical-rich information into molecular graphs via stereoelectronic effects, enhancing expressivity and interpretability. Learning to predict the stereoelectronics-infused representation with a tailored double graph neural network workflow enables its application to any downstream molecular machine learning task without expensive quantum chemical calculations. We show that the explicit addition of stereoelectronic information significantly improves the performance of message-passing 2D machine learning models for molecular property prediction. We show that the learned representations trained on small molecules can accurately extrapolate to much larger molecular structures, yielding chemical insight into orbital interactions for previously intractable systems, such as entire proteins, opening new avenues of molecular design. Finally, we have developed a web application (simg.cheme.cmu.edu) where users can rapidly explore stereoelectronic information for their own molecular systems. | [
{
"version": "v1",
"created": "Thu, 08 Aug 2024 15:21:07 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 22:48:43 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Boiko",
"Daniil A.",
""
],
[
"Reschützegger",
"Thiago",
""
],
[
"Sanchez-Lengeling",
"Benjamin",
""
],
[
"Blau",
"Samuel M.",
""
],
[
"Gomes",
"Gabe",
""
]
] |
2408.05034 | Martin Ulaga | Martin Ulaga and Jure Kokalj and Takami Tohyama and Peter Prelov\v{s}ek | Easy-axis Heisenberg model on the triangular lattice: from supersolid to gapped solid | null | Phys. Rev. B 111, 174442, 2025 | 10.1103/PhysRevB.111.174442 | null | cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the easy-axis Heisenberg model on the triangular lattice by numerically studying excitations and the dynamical spin structure factor $S^{\mu\mu}({\bf q},\omega)$. Results are analyzed within the supersolid scenario, characterized by the translation-symmetry-breaking parameter $m_z$ and the supersolid offdiagonal order parameter $m_\perp$. We find very robust $m_z > 0$ in the whole easy-axis anisotropy regime $\alpha = J_\perp/J_z > 0$, even enhanced by the magnetic field $h>0$, as well as $m_\perp >0$ for intermediate $\alpha <1$ and $h>0$. Still, at small $\alpha \lesssim 0.2$, relevant for recent experiments on the magnetic material K$_2$Co(SeO$_3$)$_2$, we find at $h=0$ rather vanishing $m_\perp \sim 0$, which appears compatible with the numerically established finite magnon excitation gap $\Delta_1 \sim 0.25 \alpha J$. | [
{
"version": "v1",
"created": "Fri, 09 Aug 2024 12:43:50 GMT"
},
{
"version": "v2",
"created": "Fri, 13 Dec 2024 15:28:38 GMT"
},
{
"version": "v3",
"created": "Wed, 23 Apr 2025 16:32:38 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Ulaga",
"Martin",
""
],
[
"Kokalj",
"Jure",
""
],
[
"Tohyama",
"Takami",
""
],
[
"Prelovšek",
"Peter",
""
]
] |
2408.05481 | Koichiro Yasuda | Daiju Nakayama, Kin-ya Oda, and Koichiro Yasuda | Covariant Electromagnetism in Past-Light-Cone Formalism | 18 pages, 1 figure; second half of Sec. 2 and entirety of Sec. 5 from previous version (v2) separated into an independent paper following referee's suggestion | null | null | null | physics.class-ph hep-ph physics.comp-ph physics.ed-ph | http://creativecommons.org/licenses/by/4.0/ | We present a manifestly covariant formulation of relativistic electromagnetism, focusing on the computation of electromagnetic fields from moving charges in a manifestly Lorentz-covariant manner. The electromagnetic field at a given spacetime point is determined by the motion of point charges at the intersection of their worldlines with the past light cone of the spacetime point, ensuring causal consistency. This formalism provides a manifestly covariant generalization of the Li\'enard-Wiechert potentials and allows direct implementation on computers. We compare our formulation with standard textbook approaches and analyze its behavior in various physical limits. | [
{
"version": "v1",
"created": "Sat, 10 Aug 2024 08:18:12 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Sep 2024 04:03:56 GMT"
},
{
"version": "v3",
"created": "Mon, 26 May 2025 23:45:18 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Nakayama",
"Daiju",
""
],
[
"Oda",
"Kin-ya",
""
],
[
"Yasuda",
"Koichiro",
""
]
] |
2408.05846 | Jialin Liu | Jialin Liu, Diansheng Liao | A Universal Flexible Neuromorphic Tactile System with Multithreshold Strategy | null | null | null | null | cs.NE cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Extremely increased unstructured data brought by the large-scale intelligent sensing devices application have big challenges not only in data storing and processing but also power consumption surging. Therefore, to improve energy efficiency and processing speed, a new generation system structure and construction strategy is necessary. Most biological nervous systems, especially the tactile system, have a good flexibility and data processing performance with low power usage. Inspired from this mechanism, to optimize the intelligent system, we report a universal fully flexible neuromorphic perception system with a strong compatibility and multi-threshold signal processing strategy by mimicking tactile nervous system. Peak signal accumulated from spike encoded sensor signal in front-end processing unit can be used for recognition task directly since the bionic synaptic plasticity. Compared with conventional systems, power consumption of our system significantly decreases about 1 order of magnitude in a same recognition task. What is more, the design of voltage-based matching circuit and multithreshold processing circuit provide an excellent compatibility and multi-signal processing capability in our system. In feasibility verification, our system can output trend of different input signals (continuous signal and frequency signal etc.) accurately and have a high recognition accuracy of 90% in the symbol pattern and 90% in Morse code. These properties of our neuromorphic system show a great application potential in intelligent devices and bionic robots. | [
{
"version": "v1",
"created": "Sun, 11 Aug 2024 19:05:31 GMT"
},
{
"version": "v2",
"created": "Tue, 13 Aug 2024 14:33:36 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 15:45:48 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Liu",
"Jialin",
""
],
[
"Liao",
"Diansheng",
""
]
] |
2408.06831 | Shibo Liu | Shibo Liu and Ligang Liu and Xiao-Ming Fu | Polynomial 2D Green Coordinates for High-order Cages | null | null | null | null | cs.CG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose conformal polynomial coordinates for 2D closed high-order cages, which consist of polynomial curves of any order. The coordinates enable the transformation of the input polynomial curves into polynomial curves of any order. We extend the classical 2D Green coordinates to define our coordinates, thereby leading to cage-aware conformal harmonic deformations. We extensively test our method on various 2D deformations, allowing users to manipulate the \Bezier control points to easily generate the desired deformation. | [
{
"version": "v1",
"created": "Tue, 13 Aug 2024 11:45:21 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 12:34:12 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Liu",
"Shibo",
""
],
[
"Liu",
"Ligang",
""
],
[
"Fu",
"Xiao-Ming",
""
]
] |
2408.07401 | Zhuoyue Wan | Zhuoyue Wan, Yuanfeng Song, Shuaimin Li, Chen Jason Zhang, Raymond Chi-Wing Wong | DataVisT5: A Pre-trained Language Model for Jointly Understanding Text and Data Visualization | null | null | 10.1109/ICDE65448.2025.00131 | null | cs.CL cs.AI cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data visualization (DV) is the fundamental and premise tool to improve the efficiency in conveying the insights behind the big data, which has been widely accepted in existing data-driven world. Task automation in DV, such as converting natural language queries to visualizations (i.e., text-to-vis), generating explanations from visualizations (i.e., vis-to-text), answering DV-related questions in free form (i.e. FeVisQA), and explicating tabular data (i.e., table-to-text), is vital for advancing the field. Despite their potential, the application of pre-trained language models (PLMs) like T5 and BERT in DV has been limited by high costs and challenges in handling cross-modal information, leading to few studies on PLMs for DV. We introduce DataVisT5, a novel PLM tailored for DV that enhances the T5 architecture through a hybrid objective pre-training and multi-task fine-tuning strategy, integrating text and DV datasets to effectively interpret cross-modal semantics. Extensive evaluations on public datasets show that DataVisT5 consistently outperforms current state-of-the-art models on various DV-related tasks. We anticipate that DataVisT5 will not only inspire further research on vertical PLMs but also expand the range of applications for PLMs. | [
{
"version": "v1",
"created": "Wed, 14 Aug 2024 09:20:17 GMT"
},
{
"version": "v2",
"created": "Wed, 27 Nov 2024 17:42:57 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Wan",
"Zhuoyue",
""
],
[
"Song",
"Yuanfeng",
""
],
[
"Li",
"Shuaimin",
""
],
[
"Zhang",
"Chen Jason",
""
],
[
"Wong",
"Raymond Chi-Wing",
""
]
] |
2408.08105 | Zhiyuan Li | Zhiyuan Li, Heng Wang, Dongnan Liu, Chaoyi Zhang, Ao Ma, Jieting Long, Weidong Cai | Multimodal Causal Reasoning Benchmark: Challenging Vision Large Language Models to Discern Causal Links Across Modalities | ACL2025 Findings | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Multimodal Large Language Models (MLLMs) have showcased exceptional Chain-of-Thought (CoT) reasoning ability in complex textual inference tasks including causal reasoning. However, will these causalities remain straightforward when crucial hints hide in visual details? If not, what factors might influence cross-modal generalization? Whether we can effectively enhance their capacity for robust causal inference across both text and vision? Motivated by these, we introduce MuCR - a novel Multimodal Causal Reasoning benchmark that leverages synthetic siamese images and text pairs to challenge MLLMs. Additionally, we develop tailored metrics from multiple perspectives, including image-level match, phrase-level understanding, and sentence-level explanation, to comprehensively assess MLLMs' comprehension abilities. Our experiments reveal that current MLLMs fall short in multimodal causal reasoning compared to their performance in purely textual settings. Additionally, we find that identifying visual cues across images is key to effective cross-modal generalization. Finally, we propose a VcCoT strategy that better highlights visual cues, and our results confirm its efficacy in enhancing multimodal causal reasoning. The project is available at: https://github.com/Zhiyuan-Li-John/MuCR | [
{
"version": "v1",
"created": "Thu, 15 Aug 2024 12:04:32 GMT"
},
{
"version": "v2",
"created": "Fri, 30 Aug 2024 05:25:22 GMT"
},
{
"version": "v3",
"created": "Sun, 16 Feb 2025 04:12:08 GMT"
},
{
"version": "v4",
"created": "Mon, 26 May 2025 03:04:03 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Li",
"Zhiyuan",
""
],
[
"Wang",
"Heng",
""
],
[
"Liu",
"Dongnan",
""
],
[
"Zhang",
"Chaoyi",
""
],
[
"Ma",
"Ao",
""
],
[
"Long",
"Jieting",
""
],
[
"Cai",
"Weidong",
""
]
] |
2408.08313 | Weiyang Liu | Zeju Qiu, Weiyang Liu, Haiwen Feng, Zhen Liu, Tim Z. Xiao, Katherine M. Collins, Joshua B. Tenenbaum, Adrian Weller, Michael J. Black, Bernhard Sch\"olkopf | Can Large Language Models Understand Symbolic Graphics Programs? | ICLR 2025 Spotlight (v4: 47 pages, 26 figures, project page: https://sgp-bench.github.io/) | null | null | null | cs.LG cs.AI cs.CL cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Against the backdrop of enthusiasm for large language models (LLMs), there is a growing need to scientifically assess their capabilities and shortcomings. This is nontrivial in part because it is difficult to find tasks which the models have not encountered during training. Utilizing symbolic graphics programs, we propose a domain well-suited to test multiple spatial-semantic reasoning skills of LLMs. Popular in computer graphics, these programs procedurally generate visual data. While LLMs exhibit impressive skills in general program synthesis and analysis, symbolic graphics programs offer a new layer of evaluation: they allow us to test an LLM's ability to answer semantic questions about the images or 3D geometries without a vision encoder. To semantically understand the symbolic programs, LLMs would need to possess the ability to "imagine" and reason how the corresponding graphics content would look with only the symbolic description of the local curvatures and strokes. We use this task to evaluate LLMs by creating a large benchmark for the semantic visual understanding of symbolic graphics programs, built procedurally with minimal human effort. Particular emphasis is placed on transformations of images that leave the image level semantics invariant while introducing significant changes to the underlying program. We evaluate commercial and open-source LLMs on our benchmark to assess their ability to reason about visual output of programs, finding that LLMs considered stronger at reasoning generally perform better. Lastly, we introduce a novel method to improve this ability -- Symbolic Instruction Tuning (SIT), in which the LLM is finetuned with pre-collected instruction data on symbolic graphics programs. Interestingly, we find that SIT not only improves LLM's understanding on symbolic programs, but it also improves general reasoning ability on various other benchmarks. | [
{
"version": "v1",
"created": "Thu, 15 Aug 2024 17:59:57 GMT"
},
{
"version": "v2",
"created": "Mon, 07 Oct 2024 08:44:35 GMT"
},
{
"version": "v3",
"created": "Wed, 11 Dec 2024 21:42:14 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 16:54:13 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Qiu",
"Zeju",
""
],
[
"Liu",
"Weiyang",
""
],
[
"Feng",
"Haiwen",
""
],
[
"Liu",
"Zhen",
""
],
[
"Xiao",
"Tim Z.",
""
],
[
"Collins",
"Katherine M.",
""
],
[
"Tenenbaum",
"Joshua B.",
""
],
[
"Weller",
"Adrian",
""
],
[
"Black",
"Michael J.",
""
],
[
"Schölkopf",
"Bernhard",
""
]
] |
2408.08385 | Miguel Quartin | Pedro Riba Mello, Miguel Quartin, Bjoern Malte Schaefer, Benedikt Schosser | On the full non-Gaussian Surprise statistic and the cosmological concordance between DESI, SDSS and Pantheon+ | v2: revised results, alternative prior choices included, added an analysis without SH0ES data, two new appendices | The Open Journal of Astrophysics 8, 2025 | 10.33232/001c.138626 | null | astro-ph.CO astro-ph.IM | http://creativecommons.org/licenses/by/4.0/ | With the increasing precision of recent cosmological surveys and the discovery of important tensions within the $\Lambda$CDM paradigm, it is becoming more and more important to develop tools to quantify accurately the discordance between different probes. One such tool is the Surprise statistic, a measure based on the Kullback-Leibler divergence. The Surprise, however, has been up to now applied only under its Gaussian approximation, which can fail to properly capture discordance in cases that deviate significantly from Gaussianity. In this paper we developed the \texttt{klsurprise} code which computes the full numerical non-Gaussian Surprise, and analyse the Surprise for BAO + BBN and supernova data. We test different cosmological models, some of which the parameters deviate significantly from Gaussianity. We find that the non-Gaussianities, mainly present in the Supernova dataset, change the Surprise values significantly from its Gaussian approximation, and reveal a borderline $2.0\sigma$ tension in the curved $w$CDM model (o$w$CDM) between the combined Pantheon+ and SH0ES (Pantheon+ & SH0ES) data and the dataset which combines SDSS, BOSS and eBOSS BAO. This modest tension is hidden in the Gaussian Surprise approximation. For DESI, the discrepancy with Pantheon+ & SH0ES is at the $1.5\sigma$ level for o$w$CDM, but a large $3.4\sigma$ for $\Lambda$CDM. Removing SH0ES data drops the $\Lambda$CDM significance to $2.6\sigma$. | [
{
"version": "v1",
"created": "Thu, 15 Aug 2024 19:06:57 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 17:18:16 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Mello",
"Pedro Riba",
""
],
[
"Quartin",
"Miguel",
""
],
[
"Schaefer",
"Bjoern Malte",
""
],
[
"Schosser",
"Benedikt",
""
]
] |
2408.09719 | Yiyao Zhang | Hongyang Liu, Yitong Yin, Yiyao Zhang | Work-Efficient Parallel Counting via Sampling | null | null | null | null | cs.DS cs.CC cs.DC | http://creativecommons.org/licenses/by/4.0/ | A canonical approach to approximating the partition function of a Gibbs distribution via sampling is simulated annealing. This method has led to efficient reductions from counting to sampling, including:
$\bullet$ classic non-adaptive (parallel) algorithms with sub-optimal cost (Dyer-Frieze-Kannan '89; Bez\'akov\'a-\v{S}tefankovi\v{c}-Vazirani-Vigoda '08);
$\bullet$ adaptive (sequential) algorithms with near-optimal cost (\v{S}tefankovi\v{c}-Vempala-Vigoda '09; Huber '15; Kolmogorov '18; Harris-Kolmogorov '24).
We present an algorithm that achieves both near-optimal total work and efficient parallelism, providing a reduction from counting to sampling with logarithmic depth and near-optimal work. As consequences, we obtain work-efficient parallel counting algorithms for several important models, including the hardcore and Ising models within the uniqueness regime. | [
{
"version": "v1",
"created": "Mon, 19 Aug 2024 06:18:59 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 05:49:43 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Liu",
"Hongyang",
""
],
[
"Yin",
"Yitong",
""
],
[
"Zhang",
"Yiyao",
""
]
] |
2408.09763 | Anupam Gupta | Soni D. Prajapati, Akshay Bhatnagar, Anupam Gupta | Effect of the background flow on the motility induced phase separation | 10 pages, 9 figures | null | 10.1039/D5SM00362H | null | cond-mat.soft | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We simulate active Brownian particles (ABPs) with soft-repulsive interactions subjected to a four-roll-mill flow. In the absence of flow, this system exhibits motility-induced phase separation (MIPS). To investigate the interplay between MIPS and flow-induced mixing, we introduce dimensionless parameters: a scaled time, $\tau$, and a scaled velocity, ${\rm v}$, characterizing the ratio of ABP to fluid time and velocity scales, respectively. The parameter space defined by $\tau$ and ${\rm v}$ reveals three distinct ABP distribution regimes. At low velocities ${\rm v} \ll 1$, flow dominates, leading to a homogeneous mixture. Conversely, at high velocities ${\rm v} \gg 1$, motility prevails, resulting in MIPS. In the intermediate regime (${\rm v} \sim 1$), the system's behavior depends on $\tau$. For $\tau <1$, a moderately mixed homogeneous phase emerges, while for $\tau >1$, a novel phase, termed flow-induced phase separation (FIPS), arises due to the combined effects of flow topology and ABP motility and size. To characterize these phases, we analyze drift velocity, diffusivity, mean-squared displacement, giant number fluctuations, radial distribution function, and cluster-size distribution. | [
{
"version": "v1",
"created": "Mon, 19 Aug 2024 07:44:51 GMT"
},
{
"version": "v2",
"created": "Tue, 20 Aug 2024 18:37:26 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 10:29:11 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Prajapati",
"Soni D.",
""
],
[
"Bhatnagar",
"Akshay",
""
],
[
"Gupta",
"Anupam",
""
]
] |
2408.10455 | Kaiyu He | Kaiyu He, Mian Zhang, Shuo Yan, Peilin Wu, Zhiyu Zoey Chen | IDEA: Enhancing the Rule Learning Ability of Large Language Model Agent through Induction, Deduction, and Abduction | Accepted to ACL 2025 Findings | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | While large language models (LLMs) have been thoroughly evaluated for deductive and inductive reasoning, their proficiency in holistic rule learning in interactive environments remains less explored. We introduce RULEARN, a novel benchmark to assess the rule-learning abilities of LLM agents in interactive settings. In RULEARN, agents strategically interact with simulated environments to gather observations, discern patterns, and solve complex problems. To enhance the rule-learning capabilities for LLM agents, we propose IDEA, a novel reasoning framework that integrates the process of Induction, Deduction, and Abduction. The IDEA agent generates initial hypotheses from limited observations through abduction, devises plans to validate these hypotheses or leverages them to solve problems via deduction, and refines previous hypotheses through induction, dynamically establishing and applying rules that mimic human rule-learning behaviors. Our evaluation of the IDEA framework, which involves five representative LLMs, demonstrates significant improvements over the baseline. Furthermore, our study with human participants reveals notable discrepancies in rule-learning behaviors between humans and LLMs. We believe our benchmark will serve as a valuable and challenging resource, and IDEA will provide crucial insights for the development of LLM agents capable of human-like rule learning in real-world scenarios. Our code and data is publicly available. | [
{
"version": "v1",
"created": "Mon, 19 Aug 2024 23:37:07 GMT"
},
{
"version": "v2",
"created": "Sat, 24 Aug 2024 03:23:09 GMT"
},
{
"version": "v3",
"created": "Wed, 02 Oct 2024 07:30:27 GMT"
},
{
"version": "v4",
"created": "Thu, 03 Oct 2024 01:47:53 GMT"
},
{
"version": "v5",
"created": "Thu, 19 Dec 2024 05:45:35 GMT"
},
{
"version": "v6",
"created": "Tue, 27 May 2025 05:26:57 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"He",
"Kaiyu",
""
],
[
"Zhang",
"Mian",
""
],
[
"Yan",
"Shuo",
""
],
[
"Wu",
"Peilin",
""
],
[
"Chen",
"Zhiyu Zoey",
""
]
] |
2408.11186 | Surya Murthy | Surya Murthy, Mustafa O. Karabag, Ufuk Topcu | Sequential Resource Trading Using Comparison-Based Gradient Estimation | null | null | null | null | cs.MA cs.AI math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autonomous agents interact with other autonomous agents and humans of unknown preferences to share resources in their environment. We explore sequential trading for resource allocation in a setting where two greedily rational agents sequentially trade resources from a finite set of categories. Each agent has a utility function that depends on the amount of resources it possesses in each category. The offering agent makes trade offers to improve its utility without knowing the responding agent's utility function, and the responding agent only accepts offers that improve its utility. To facilitate cooperation between an autonomous agent and another autonomous agent or a human, we present an algorithm for the offering agent to estimate the responding agent's gradient (preferences) and make offers based on previous acceptance or rejection responses. The algorithm's goal is to reach a Pareto-optimal resource allocation state while ensuring that the utilities of both agents improve after every accepted trade. The algorithm estimates the responding agent's gradient by leveraging the rejected offers and the greedy rationality assumption, to prune the space of potential gradients. We show that, after the algorithm makes a finite number of rejected offers, the algorithm either finds a mutually beneficial trade or certifies that the current state is epsilon-weakly Pareto optimal. We compare the proposed algorithm against various baselines in continuous and discrete trading scenarios and show that it improves the societal benefit with fewer offers. Additionally, we validate these findings in a user study with human participants, where the algorithm achieves high performance in scenarios with high resource conflict due to aligned agent goals. | [
{
"version": "v1",
"created": "Tue, 20 Aug 2024 20:42:41 GMT"
},
{
"version": "v2",
"created": "Sun, 03 Nov 2024 23:38:11 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 16:36:01 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Murthy",
"Surya",
""
],
[
"Karabag",
"Mustafa O.",
""
],
[
"Topcu",
"Ufuk",
""
]
] |
2408.13150 | Joao V. Cavalcanti | Joao V. Cavalcanti and Laurent Lessard and Ashia C. Wilson | Adaptive Backtracking Line Search | null | null | null | null | math.OC cs.LG | http://creativecommons.org/licenses/by/4.0/ | Backtracking line search is foundational in numerical optimization. The basic idea is to adjust the step-size of an algorithm by a constant factor until some chosen criterion (e.g. Armijo, Descent Lemma) is satisfied. We propose a novel way to adjust step-sizes, replacing the constant factor used in regular backtracking with one that takes into account the degree to which the chosen criterion is violated, with no additional computational burden. This light-weight adjustment leads to significantly faster optimization, which we confirm by performing a variety of experiments on over fifteen real world datasets. For convex problems, we prove adaptive backtracking requires no more adjustments to produce a feasible step-size than regular backtracking does. For nonconvex smooth problems, we prove adaptive backtracking enjoys the same guarantees of regular backtracking. Furthermore, we prove adaptive backtracking preserves the convergence rates of gradient descent and its accelerated variant. | [
{
"version": "v1",
"created": "Fri, 23 Aug 2024 15:16:57 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 02:19:54 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Cavalcanti",
"Joao V.",
""
],
[
"Lessard",
"Laurent",
""
],
[
"Wilson",
"Ashia C.",
""
]
] |
2408.14775 | Junliang Shen | Davesh Maulik, Junliang Shen, Qizheng Yin, Ruxuan Zhang | The D-equivalence conjecture for hyper-K\"ahler varieties via hyperholomorphic bundles | Accepted version; to appear at Inventiones Mathematicae | null | null | null | math.AG | http://creativecommons.org/licenses/by/4.0/ | We show that birational hyper-K\"ahler varieties of $K3^{[n]}$-type are derived equivalent, establishing the D-equivalence conjecture in these cases. The Fourier-Mukai kernels of our derived equivalences are constructed from projectively hyperholomorphic bundles, following ideas of Markman. Our method also proves a stronger version of the D-equivalence conjecture for hyper-K\"ahler varieties of $K3^{[n]}$-type with Brauer classes. | [
{
"version": "v1",
"created": "Tue, 27 Aug 2024 04:44:26 GMT"
},
{
"version": "v2",
"created": "Sat, 31 Aug 2024 18:01:06 GMT"
},
{
"version": "v3",
"created": "Fri, 01 Nov 2024 03:39:41 GMT"
},
{
"version": "v4",
"created": "Fri, 15 Nov 2024 04:24:18 GMT"
},
{
"version": "v5",
"created": "Tue, 27 May 2025 03:56:43 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Maulik",
"Davesh",
""
],
[
"Shen",
"Junliang",
""
],
[
"Yin",
"Qizheng",
""
],
[
"Zhang",
"Ruxuan",
""
]
] |
2408.15501 | Yifu Yuan | Yifu Yuan, Zhenrui Zheng, Zibin Dong, Jianye Hao | MODULI: Unlocking Preference Generalization via Diffusion Models for Offline Multi-Objective Reinforcement Learning | Accepted by ICML2025, code link: https://github.com/pickxiguapi/MODULI | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-objective Reinforcement Learning (MORL) seeks to develop policies that simultaneously optimize multiple conflicting objectives, but it requires extensive online interactions. Offline MORL provides a promising solution by training on pre-collected datasets to generalize to any preference upon deployment. However, real-world offline datasets are often conservatively and narrowly distributed, failing to comprehensively cover preferences, leading to the emergence of out-of-distribution (OOD) preference areas. Existing offline MORL algorithms exhibit poor generalization to OOD preferences, resulting in policies that do not align with preferences. Leveraging the excellent expressive and generalization capabilities of diffusion models, we propose MODULI (Multi-objective Diffusion Planner with Sliding Guidance), which employs a preference-conditioned diffusion model as a planner to generate trajectories that align with various preferences and derive action for decision-making. To achieve accurate generation, MODULI introduces two return normalization methods under diverse preferences for refining guidance. To further enhance generalization to OOD preferences, MODULI proposes a novel sliding guidance mechanism, which involves training an additional slider adapter to capture the direction of preference changes. Incorporating the slider, it transitions from in-distribution (ID) preferences to generating OOD preferences, patching, and extending the incomplete Pareto front. Extensive experiments on the D4MORL benchmark demonstrate that our algorithm outperforms state-of-the-art Offline MORL baselines, exhibiting excellent generalization to OOD preferences. | [
{
"version": "v1",
"created": "Wed, 28 Aug 2024 03:10:45 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 03:51:30 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Yuan",
"Yifu",
""
],
[
"Zheng",
"Zhenrui",
""
],
[
"Dong",
"Zibin",
""
],
[
"Hao",
"Jianye",
""
]
] |
2408.15605 | Guillermo Gallego | Suman Ghosh, Valentina Cavinato, Guillermo Gallego | ES-PTAM: Event-based Stereo Parallel Tracking and Mapping | 17 pages, 7 figures, 4 tables, https://github.com/tub-rip/ES-PTAM | European Conference on Computer Vision (ECCV) Workshops, Milan, Italy 2024 | 10.1007/978-3-031-92460-6_5 | null | cs.RO cs.CV eess.SP | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Visual Odometry (VO) and SLAM are fundamental components for spatial perception in mobile robots. Despite enormous progress in the field, current VO/SLAM systems are limited by their sensors' capability. Event cameras are novel visual sensors that offer advantages to overcome the limitations of standard cameras, enabling robots to expand their operating range to challenging scenarios, such as high-speed motion and high dynamic range illumination. We propose a novel event-based stereo VO system by combining two ideas: a correspondence-free mapping module that estimates depth by maximizing ray density fusion and a tracking module that estimates camera poses by maximizing edge-map alignment. We evaluate the system comprehensively on five real-world datasets, spanning a variety of camera types (manufacturers and spatial resolutions) and scenarios (driving, flying drone, hand-held, egocentric, etc). The quantitative and qualitative results demonstrate that our method outperforms the state of the art in majority of the test sequences by a margin, e.g., trajectory error reduction of 45% on RPG dataset, 61% on DSEC dataset, and 21% on TUM-VIE dataset. To benefit the community and foster research on event-based perception systems, we release the source code and results: https://github.com/tub-rip/ES-PTAM | [
{
"version": "v1",
"created": "Wed, 28 Aug 2024 07:56:28 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Ghosh",
"Suman",
""
],
[
"Cavinato",
"Valentina",
""
],
[
"Gallego",
"Guillermo",
""
]
] |
2408.15949 | Alexander S. Sakharov | Mauricio Bustamante, John Ellis, Rostislav Konoplich, Alexander S. Sakharov | Probing Lorentz invariance with a high-energy neutrino flare | 19 pages, 7 figures, plus an appendix, version accepted for publication in Physical Review D | null | null | KCL-PH-TH/2024-22, CERN-TH-2024-056 | astro-ph.HE gr-qc hep-ex hep-ph hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Time-of-flight measurements of high-energy astrophysical neutrinos can be used to probe Lorentz invariance, a pillar of modern physics. If Lorentz-invariance violation (LIV) occurs, it could cause neutrinos to slow down, with the delay scaling linearly or quadratically with their energy. We introduce non-parametric statistical methods designed to detect LIV-induced distortions in the temporal structure of a high-energy neutrino flare as it travels to Earth from a distant astrophysical source, independently of the intrinsic timing properties of the source. Our approach, illustrated using the 2014/2015 TeV-PeV neutrino flare from the blazar TXS 0506+056 detected by IceCube, finds that the LIV energy scale must exceed 10^{14} GeV (linear) or 10^9 GeV (quadratic). Our methods provide a robust means to investigate LIV by focusing solely on a neutrino flare, without relying on electromagnetic counterparts, and account for realistic energy and directional uncertainties. For completeness, we compare our limits inferred from TXS 0506+056 to the sensitivity inferred from multi-messenger detection of tentative coincidences between neutrinos and electromagnetic emission from active galactic nuclei and tidal disruption events. | [
{
"version": "v1",
"created": "Wed, 28 Aug 2024 17:08:18 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 15:32:34 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Bustamante",
"Mauricio",
""
],
[
"Ellis",
"John",
""
],
[
"Konoplich",
"Rostislav",
""
],
[
"Sakharov",
"Alexander S.",
""
]
] |
2408.16128 | Brennan de Neeve | Brennan de Neeve, Thanh-Long Nguyen, Alexander Ferk, Tanja Behrle, Francesco Lancellotti, Matteo Simoni, Stephan Welte, Jonathan Home | Modular variable laser cooling for efficient entropy extraction | null | null | 10.1103/PhysRevLett.134.203603 | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | We propose and experimentally demonstrate a method for laser cooling an oscillator based on sequences of spin-state-dependent displacements followed by spin repumping. For a thermal state with mean occupation $\bar{n}\gg 1$ the method attains a reduction to 0.632 of the initial thermal oscillator occupation for two repumps of the two-level spin state. This is within a factor of 2.53 of the optimum that might be expected due to the reduction of the oscillator entropy by $2 \ln(2)$. We show that the method, which is based on encoding the value of the modular-variable of the oscillator into the spin, has a simple semi-classical description in terms of a Bayesian update. We demonstrate the method experimentally using the internal and motional states of a single trapped ion. | [
{
"version": "v1",
"created": "Wed, 28 Aug 2024 20:28:21 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"de Neeve",
"Brennan",
""
],
[
"Nguyen",
"Thanh-Long",
""
],
[
"Ferk",
"Alexander",
""
],
[
"Behrle",
"Tanja",
""
],
[
"Lancellotti",
"Francesco",
""
],
[
"Simoni",
"Matteo",
""
],
[
"Welte",
"Stephan",
""
],
[
"Home",
"Jonathan",
""
]
] |
2408.16639 | Michalis (Mike) Chatzittofi | Michalis Chatzittofi, Jaime Agudo-Canalejo, Ramin Golestanian | Mechanistic rules for de novo design of enzymes | null | Chem Catalysis 5, 101394 (2025) | 10.1016/j.checat.2025.101394 | null | cond-mat.soft cond-mat.stat-mech physics.bio-ph physics.chem-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Enzymes are nano-scale machines that have evolved to drive chemical reactions out of equilibrium in the right place at the right time. Given the complexity and specificity of enzymatic function, bottom-up design of enzymes presents a daunting task that is far more challenging than making passive molecules with specific binding affinities or building nano-scale mechanically active devices. We present a thermodynamically-consistent model for the operation of such a fuelled enzyme, which uses the energy from a favourable reaction to undergo non-equilibrium conformational changes that in turn catalyze a chemical reaction on an attached substrate molecule. We show that enzymatic function can emerge through a bifurcation upon appropriate implementation of momentum conservation on the effective reaction coordinates of the low dimensional description of the enzyme, and thanks to a generically present dissipative coupling. Our results can complement the recently developed strategies for de novo enzyme design based on machine learning approaches. | [
{
"version": "v1",
"created": "Thu, 29 Aug 2024 15:44:41 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 14:53:19 GMT"
},
{
"version": "v3",
"created": "Wed, 21 May 2025 08:14:48 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Chatzittofi",
"Michalis",
""
],
[
"Agudo-Canalejo",
"Jaime",
""
],
[
"Golestanian",
"Ramin",
""
]
] |
2408.16711 | Smita Rajan | Smita Rajan, Svala Sverrisd\'ottir, Bernd Sturmfels | Kinematic Varieties for Massless Particles | 19 pages | Special volume on Positive Geometry, Le Matematiche 80 (1) (2025), 453-471 | null | null | math.AG hep-th | http://creativecommons.org/licenses/by/4.0/ | We study algebraic varieties that encode the kinematic data for $n$ massless particles in $d$-dimensional spacetime subject to momentum conservation. Their coordinates are spinor brackets, which we derive from the Clifford algebra associated to the Lorentz group. This was proposed for $d=5$ in the recent physics literature. Our kinematic varieties are given by polynomial constraints on tensors with both symmetric and skew symmetric slices. | [
{
"version": "v1",
"created": "Thu, 29 Aug 2024 17:05:17 GMT"
},
{
"version": "v2",
"created": "Mon, 09 Dec 2024 06:30:08 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Rajan",
"Smita",
""
],
[
"Sverrisdóttir",
"Svala",
""
],
[
"Sturmfels",
"Bernd",
""
]
] |
2409.01249 | Giorgio Piras | Giorgio Piras, Maura Pintor, Ambra Demontis, Battista Biggio, Giorgio Giacinto, Fabio Roli | Adversarial Pruning: A Survey and Benchmark of Pruning Methods for Adversarial Robustness | null | null | 10.1016/j.patcog.2025.111788 | null | cs.LG cs.CR cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recent work has proposed neural network pruning techniques to reduce the size of a network while preserving robustness against adversarial examples, i.e., well-crafted inputs inducing a misclassification. These methods, which we refer to as adversarial pruning methods, involve complex and articulated designs, making it difficult to analyze the differences and establish a fair and accurate comparison. In this work, we overcome these issues by surveying current adversarial pruning methods and proposing a novel taxonomy to categorize them based on two main dimensions: the pipeline, defining when to prune; and the specifics, defining how to prune. We then highlight the limitations of current empirical analyses and propose a novel, fair evaluation benchmark to address them. We finally conduct an empirical re-evaluation of current adversarial pruning methods and discuss the results, highlighting the shared traits of top-performing adversarial pruning methods, as well as common issues. We welcome contributions in our publicly-available benchmark at https://github.com/pralab/AdversarialPruningBenchmark | [
{
"version": "v1",
"created": "Mon, 02 Sep 2024 13:34:01 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Piras",
"Giorgio",
""
],
[
"Pintor",
"Maura",
""
],
[
"Demontis",
"Ambra",
""
],
[
"Biggio",
"Battista",
""
],
[
"Giacinto",
"Giorgio",
""
],
[
"Roli",
"Fabio",
""
]
] |
2409.03140 | Soumik Dey | Ashirbad Mishra, Soumik Dey, Marshall Wu, Jinyu Zhao, He Yu, Kaichen Ni, Binbin Li, Kamesh Madduri | GraphEx: A Graph-based Extraction Method for Advertiser Keyphrase Recommendation | null | 2025 IEEE 41st International Conference on Data Engineering (ICDE) | 10.1109/ICDE65448.2025.00330 | null | cs.IR cs.CL cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Online sellers and advertisers are recommended keyphrases for their listed products, which they bid on to enhance their sales. One popular paradigm that generates such recommendations is Extreme Multi-Label Classification (XMC), which involves tagging/mapping keyphrases to items. We outline the limitations of using traditional item-query based tagging or mapping techniques for keyphrase recommendations on E-Commerce platforms. We introduce GraphEx, an innovative graph-based approach that recommends keyphrases to sellers using extraction of token permutations from item titles. Additionally, we demonstrate that relying on traditional metrics such as precision/recall can be misleading in practical applications, thereby necessitating a combination of metrics to evaluate performance in real-world scenarios. These metrics are designed to assess the relevance of keyphrases to items and the potential for buyer outreach. GraphEx outperforms production models at eBay, achieving the objectives mentioned above. It supports near real-time inferencing in resource-constrained production environments and scales effectively for billions of items. | [
{
"version": "v1",
"created": "Thu, 05 Sep 2024 00:25:37 GMT"
},
{
"version": "v2",
"created": "Fri, 06 Sep 2024 18:41:50 GMT"
},
{
"version": "v3",
"created": "Sat, 25 Jan 2025 01:47:28 GMT"
},
{
"version": "v4",
"created": "Mon, 28 Apr 2025 14:38:36 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Mishra",
"Ashirbad",
""
],
[
"Dey",
"Soumik",
""
],
[
"Wu",
"Marshall",
""
],
[
"Zhao",
"Jinyu",
""
],
[
"Yu",
"He",
""
],
[
"Ni",
"Kaichen",
""
],
[
"Li",
"Binbin",
""
],
[
"Madduri",
"Kamesh",
""
]
] |
2409.03358 | Friedhelm Hamann | Friedhelm Hamann, Hanxiong Li, Paul Mieske, Lars Lewejohann, Guillermo Gallego | MouseSIS: A Frames-and-Events Dataset for Space-Time Instance Segmentation of Mice | 18 pages, 5 figures, ECCV Workshops | European Conference on Computer Vision (ECCV) Workshops, Milan, Italy 2024 | 10.1007/978-3-031-92460-6_10 | null | cs.CV cs.LG cs.RO | http://creativecommons.org/licenses/by-sa/4.0/ | Enabled by large annotated datasets, tracking and segmentation of objects in videos has made remarkable progress in recent years. Despite these advancements, algorithms still struggle under degraded conditions and during fast movements. Event cameras are novel sensors with high temporal resolution and high dynamic range that offer promising advantages to address these challenges. However, annotated data for developing learning-based mask-level tracking algorithms with events is not available. To this end, we introduce: ($i$) a new task termed \emph{space-time instance segmentation}, similar to video instance segmentation, whose goal is to segment instances throughout the entire duration of the sensor input (here, the input are quasi-continuous events and optionally aligned frames); and ($ii$) \emph{\dname}, a dataset for the new task, containing aligned grayscale frames and events. It includes annotated ground-truth labels (pixel-level instance segmentation masks) of a group of up to seven freely moving and interacting mice. We also provide two reference methods, which show that leveraging event data can consistently improve tracking performance, especially when used in combination with conventional cameras. The results highlight the potential of event-aided tracking in difficult scenarios. We hope our dataset opens the field of event-based video instance segmentation and enables the development of robust tracking algorithms for challenging conditions.\url{https://github.com/tub-rip/MouseSIS} | [
{
"version": "v1",
"created": "Thu, 05 Sep 2024 09:01:11 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Hamann",
"Friedhelm",
""
],
[
"Li",
"Hanxiong",
""
],
[
"Mieske",
"Paul",
""
],
[
"Lewejohann",
"Lars",
""
],
[
"Gallego",
"Guillermo",
""
]
] |
2409.03732 | Keenan Down | Keenan J. A. Down and Pedro A. M. Mediano | A Logarithmic Decomposition and a Signed Measure Space for Entropy | 35 pages, 11 figures, extended version of a conference paper. Updated 2025-05-27 to remove erroneous references to the Wyner common information | null | null | null | cs.IT math.IT | http://creativecommons.org/licenses/by/4.0/ | The Shannon entropy of a random variable has much behaviour analogous to a signed measure. Previous work has explored this connection by defining a signed measure on abstract sets, which are taken to represent the information that different random variables contain. This construction is sufficient to derive many measure-theoretical counterparts to information quantities such as the mutual information (the intersection of sets), the joint entropy (the union of sets), and the conditional entropy (the difference of sets). Here we provide concrete characterisations of these abstract sets and a corresponding signed measure by extending the approach used by Yeung to all possible outcomes in an outcome space $\Omega$, and in doing so we demonstrate that there exists a much finer decomposition with intuitive properties which we call the logarithmic decomposition (LD). We show that this signed measure space has the useful property that its logarithmic atoms are easily characterised with negative or positive entropy, depending only on their structure, while also being consistent with Yeung's I-measure. We present the usability of our approach by re-examining the G\'acs-K\"orner common information and minimally sufficient statistics from this new geometric perspective and characterising it in terms of our logarithmic atoms -- a property we call logarithmic decomposability. We present possible extensions of this construction to continuous probability distributions before discussing implications for quality-led information theory. As a motivating example, we apply our new decomposition to the Dyadic and Triadic systems of James and Crutchfield and show that, in contrast to the I-measure alone, our decomposition is able to qualitatively distinguish between them. | [
{
"version": "v1",
"created": "Thu, 05 Sep 2024 17:43:17 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 12:15:29 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Down",
"Keenan J. A.",
""
],
[
"Mediano",
"Pedro A. M.",
""
]
] |
2409.04183 | Ziyin Zhang | Ziyin Zhang, Hang Yu, Shijie Li, Peng Di, Jianguo Li, Rui Wang | GALLa: Graph Aligned Large Language Models for Improved Source Code Understanding | ACL 2025 camera-ready | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Programming languages possess rich semantic information - such as data flow - that is represented by graphs and not available from the surface form of source code. Recent code language models have scaled to billions of parameters, but model source code solely as text tokens while ignoring any other structural information. Conversely, models that do encode structural information of code make modifications to the Transformer architecture, limiting their scale and compatibility with pretrained LLMs. In this work, we take the best of both worlds with GALLa - Graph Aligned Large Language Models. GALLa utilizes graph neural networks and cross-modal alignment technologies to inject the structural information of code into LLMs as an auxiliary task during finetuning. This framework is both model-agnostic and task-agnostic, as it can be applied to any code LLM for any code downstream task, and requires the structural graph data only at training time from a corpus unrelated to the finetuning data, while incurring no cost at inference time over the baseline LLM. Experiments on five code tasks with seven different baseline LLMs ranging in size from 350M to 14B validate the effectiveness of GALLa, demonstrating consistent improvement over the baseline, even for powerful models such as LLaMA3 and Qwen2.5-Coder. | [
{
"version": "v1",
"created": "Fri, 06 Sep 2024 10:57:34 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 14:15:14 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Zhang",
"Ziyin",
""
],
[
"Yu",
"Hang",
""
],
[
"Li",
"Shijie",
""
],
[
"Di",
"Peng",
""
],
[
"Li",
"Jianguo",
""
],
[
"Wang",
"Rui",
""
]
] |
2409.04332 | Chengkun Li | Chengkun Li, Aki Vehtari, Paul-Christian B\"urkner, Stefan T. Radev, Luigi Acerbi, Marvin Schmitt | Amortized Bayesian Workflow | 26 pages, 11 figures | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by-sa/4.0/ | Bayesian inference often faces a trade-off between computational speed and sampling accuracy. We propose an adaptive workflow that integrates rapid amortized inference with gold-standard MCMC techniques to achieve a favorable combination of both speed and accuracy when performing inference on many observed datasets. Our approach uses principled diagnostics to guide the choice of inference method for each dataset, moving along the Pareto front from fast amortized sampling via generative neural networks to slower but guaranteed-accurate MCMC when needed. By reusing computations across steps, our workflow synergizes amortized and MCMC-based inference. We demonstrate the effectiveness of this integrated approach on several synthetic and real-world problems with tens of thousands of datasets, showing efficiency gains while maintaining high posterior quality. | [
{
"version": "v1",
"created": "Fri, 06 Sep 2024 15:09:04 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 06:54:38 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Li",
"Chengkun",
""
],
[
"Vehtari",
"Aki",
""
],
[
"Bürkner",
"Paul-Christian",
""
],
[
"Radev",
"Stefan T.",
""
],
[
"Acerbi",
"Luigi",
""
],
[
"Schmitt",
"Marvin",
""
]
] |
2409.05947 | Ambreesh Khurana | Ambreesh Khurana and Sourav Chatterjee | The Role of High-mass Stellar Binaries in the Formation of High-mass Black Holes in Dense Star Clusters | 19 pages, 15 figures, and 2 tables; in press for publication in the Astrophysical Journal | null | 10.3847/1538-4357/add337 | null | astro-ph.SR astro-ph.GA astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | Recent detections of gravitational waves from mergers of binary black holes (BBHs) with pre-merger source-frame individual masses in the so-called upper mass-gap, expected due to (pulsational) pair instability supernova ((P)PISN), have created immense interest in the astrophysical production of high-mass black holes (BHs). Previous studies show that high-mass BHs may be produced via repeated BBH mergers inside dense star clusters. Alternatively, inside dense star clusters, stars with unusually low core-to-envelope mass ratios can form via mergers of high-mass stars, which then can avoid (P)PISN, but produce high-mass BHs via mass fallback. We simulate detailed star-by-star multi-physics models of dense star clusters using the Monte Carlo cluster evolution code, CMC, to investigate the role of primordial binary fraction among high-mass stars (>=15 Msun) on the formation of high-mass BHs. We vary the high-mass stellar binary fraction (fb_15_prime) while keeping all other initial properties, including the population of high-mass stars, unchanged. We find that the number of high-mass BHs, as well as the mass of the most massive BH formed via stellar core-collapse are proportional to fb_15_prime. In contrast, there is no correlation between fb_15_prime and the number of high-mass BHs formed via BH-BH mergers. Since the total production of high-mass BHs is dominated by BH-BH mergers in old clusters, the overall number of high-mass BHs produced over the typical lifetime of globular clusters is insensitive to fb_15_prime. We study the differences in the demographics of BH-BH mergers and their implications for the LIGO-Virgo-Kagra detections as a function of fb_15_prime. | [
{
"version": "v1",
"created": "Mon, 09 Sep 2024 18:00:01 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 18:00:01 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Khurana",
"Ambreesh",
""
],
[
"Chatterjee",
"Sourav",
""
]
] |
2409.05974 | Iman Marvian | Shiv Akshar Yadavalli and Iman Marvian | Optimal Distillation of Coherent States with Phase-Insensitive Operations | Comments welcome! v2: Typos are corrected; A reference is added. 14 pages + 49 pages of Appendices; 10 Figures | null | null | null | quant-ph cond-mat.stat-mech math-ph math.MP physics.atom-ph physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | By combining multiple copies of noisy coherent states of light (or other bosonic systems), it is possible to obtain a single mode in a state with lesser noise, a process known as distillation or purification of coherent states. We investigate the distillation of coherent states from coherent thermal states under general phase-insensitive operations, and find a distillation protocol that is optimal in the asymptotic regime, i.e., when the number of input copies is much greater than 1. Remarkably, we find that in this regime, the error -- as quantified by infidelity (one minus the fidelity) of the output state with the desired coherent state -- is proportional to the inverse of the purity of coherence of the input state, a quantity obtained from the Right-Logarithmic-Derivative (RLD) Fisher information metric, hence revealing an operational interpretation of this quantity. The heart of this protocol is a phase-insensitive channel that optimally converts an input coherent thermal state with high amplitude, into an output with significantly lower amplitude and temperature. Under this channel, the purity of coherence remains asymptotically conserved. While both the input and desired output are Gaussian states, we find that the optimal protocol cannot be a Gaussian channel. Among Gaussian phase-insensitive channels, the optimal distillation protocol is a simple linear optical scheme that can be implemented with beam splitters. | [
{
"version": "v1",
"created": "Mon, 09 Sep 2024 18:18:42 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 15:17:25 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Yadavalli",
"Shiv Akshar",
""
],
[
"Marvian",
"Iman",
""
]
] |
2409.06084 | James Amarel | James Amarel, Christopher Rudolf, Athanasios Iliopoulos, John Michopoulos, Leslie N. Smith | Symmetry constrained neural networks for detection and localization of damage in metal plates | null | APL Mach. Learn. 1 June 2025; 3 (2): 026106 | 10.1063/5.0242345 | null | cs.LG eess.SP | http://creativecommons.org/licenses/by/4.0/ | The present paper is concerned with deep learning techniques applied to detection and localization of damage in a thin aluminum plate. We used data collected on a tabletop apparatus by mounting to the plate four piezoelectric transducers, each of which took turn to generate a Lamb wave that then traversed the region of interest before being received by the remaining three sensors. On training a neural network to analyze time-series data of the material response, which displayed damage-reflective features whenever the plate guided waves interacted with a contact load, we achieved a model that detected with greater than $99\%$ accuracy in addition to a model that localized with $2.58 \pm 0.12$ mm mean distance error. For each task, the best-performing model was designed according to the inductive bias that our transducers were both similar and arranged in a square pattern on a nearly uniform plate. | [
{
"version": "v1",
"created": "Mon, 09 Sep 2024 21:36:08 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Sep 2024 18:05:17 GMT"
},
{
"version": "v3",
"created": "Mon, 26 May 2025 18:39:27 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Amarel",
"James",
""
],
[
"Rudolf",
"Christopher",
""
],
[
"Iliopoulos",
"Athanasios",
""
],
[
"Michopoulos",
"John",
""
],
[
"Smith",
"Leslie N.",
""
]
] |
2409.06396 | Duc Ninh Le | Thi Nhung Dao, Duc Ninh Le | Polarized $W^+W^-$ pairs at the LHC: Effects from bottom-quark induced processes at NLO QCD+EW | Added scale uncertainties for kinematic distributions. Matches published version | null | null | null | hep-ph hep-ex | http://creativecommons.org/licenses/by/4.0/ | We investigate the effects of the bottom-quark induced processes on the doubly polarized cross sections of $W^+W^-$ pair production at the LHC. The method to extract the on-shell single-top contribution is provided. Results for phenomenological and experimental analyses are given at next-to-leading order (NLO) QCD+EW accuracy, with the leading contribution from the gluon-gluon and photon-photon fusion included. We found that the contribution of the bottom-quark induced processes, after the subtraction of the on-shell $tW$ channel, is largest for the doubly longitudinal polarization. At the integrated cross section level, using a fiducial ATLAS cut with a jet veto, the effect is $9\%$ compared to the NLO value of the light-quark contribution. It increases to $13\%$ after removing the jet veto. A bound of the $tW$ interference is calculated for various kinematic distributions, showing that this interference effect is, in general, smaller for the no jet veto case. Relevant scale uncertainties are calculated to help us decide on the importance of this interference. | [
{
"version": "v1",
"created": "Tue, 10 Sep 2024 10:23:21 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 06:57:56 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Dao",
"Thi Nhung",
""
],
[
"Le",
"Duc Ninh",
""
]
] |
2409.06527 | Vivienne Langen | Vivienne Langen, Nicola Tamanini, Sylvain Marsat, Elisa Bortolas | Hierarchical Bayesian inference on an analytical model of the LISA massive black hole binary population | 19 pages, 11 figures | null | null | null | astro-ph.CO gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Massive black hole binary (MBHB) mergers will be detectable in large numbers by the Lisa Interferometer Space Antenna (LISA), which will thus provide new insights on how they form via repeated dark matter (DM) halo and galaxy mergers. Here we present a simple analytical model to generate a population of MBHB mergers based on a theoretical prescription that connects them to DM halo mergers. The high flexibility of our approach allows us to explore the broad and uncertain range of MBH seeding and growth mechanisms, as well as the different effects behind the interplay between MBH and galactic astrophysics. Such a flexibility is fundamental for the successful implementation and optimisation of the hierarchical Bayesian parameter estimation approach that here we apply to the MBHB population of LISA for the first time. Our inferred population hyper-parameters are chosen as proxies to characterise the MBH--DM halo mass scaling relation, the occupation fraction of MBHs in DM halos and the delay between halo and MBHB mergers. We find that LISA will provide tight constraints at the lower-end of the MBH-halo scaling relation, well complementing EM observations which are biased towards large masses. Furthermore, our results suggest that LISA will constrain some features of the MBH occupation fraction at high redshift, as well as merger time delays of the order of a few hundreds of Myr, opening the possibility to constrain dynamical evolution time scales such as the dynamical friction. The analysis presented here constitutes a first attempt at developing a hierarchical Bayesian inference approach to the LISA MBHB population, opening the way for several further improvements and investigations. | [
{
"version": "v1",
"created": "Tue, 10 Sep 2024 14:03:05 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 12:49:44 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Langen",
"Vivienne",
""
],
[
"Tamanini",
"Nicola",
""
],
[
"Marsat",
"Sylvain",
""
],
[
"Bortolas",
"Elisa",
""
]
] |
2409.06885 | Juan Manuel Romero | Juan M. Romero, Emiliano Montoya-Gonzalez and Oscar Velazquez-Alvarado | Alternative Bell's states and teleportation | 39 pages, 15 figures, minor corrections. Comments welcome | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | Bell's states are among the most useful in quantum computing. These state are an orthonormal base of entagled states with two qubits. We propose alternative bases of entangled states. Some of these states depend on a continuous parameter. We present the quantum circuit and code of these alternative bases. In addition, we study quantum teleportation with these entangled states and present their quantum circuits and codes associated. | [
{
"version": "v1",
"created": "Tue, 10 Sep 2024 21:56:00 GMT"
},
{
"version": "v2",
"created": "Thu, 10 Oct 2024 20:37:51 GMT"
},
{
"version": "v3",
"created": "Wed, 23 Oct 2024 17:53:14 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Romero",
"Juan M.",
""
],
[
"Montoya-Gonzalez",
"Emiliano",
""
],
[
"Velazquez-Alvarado",
"Oscar",
""
]
] |
2409.07101 | Alex Glyn-Davies | Alex Glyn-Davies, Connor Duffin, Ieva Kazlauskaite, Mark Girolami, \"O. Deniz Akyildiz | Statistical Finite Elements via Interacting Particle Langevin Dynamics | 26 pages, 6 figures | null | null | null | stat.CO physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we develop a class of interacting particle Langevin algorithms to solve inverse problems for partial differential equations (PDEs). In particular, we leverage the statistical finite elements (statFEM) formulation to obtain a finite-dimensional latent variable statistical model where the parameter is that of the (discretised) forward map and the latent variable is the statFEM solution of the PDE which is assumed to be partially observed. We then adapt a recently proposed expectation-maximisation like scheme, interacting particle Langevin algorithm (IPLA), for this problem and obtain a joint estimation procedure for the parameters and the latent variables. We consider three main examples: (i) estimating the forcing for linear Poisson PDE, (ii) estimating diffusivity for linear Poisson PDE, and (iii) estimating the forcing for nonlinear Poisson PDE. We provide computational complexity estimates for forcing estimation in the linear case. We also provide comprehensive numerical experiments and preconditioning strategies that significantly improve the performance, showing that the proposed class of methods can be the choice for parameter inference in PDE models. | [
{
"version": "v1",
"created": "Wed, 11 Sep 2024 08:45:12 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 15:24:36 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Glyn-Davies",
"Alex",
""
],
[
"Duffin",
"Connor",
""
],
[
"Kazlauskaite",
"Ieva",
""
],
[
"Girolami",
"Mark",
""
],
[
"Akyildiz",
"Ö. Deniz",
""
]
] |
2409.07305 | Si-Jiang Yang | Wei-Jie Miao and Si-Jiang Yang | Overcharging a nonsingular black hole in general relativity: the nonlinear electrodynamic field effects | 18 pages, published version | JCAP 05, 022 (2025) | 10.1088/1475-7516/2025/05/022 | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Ay\'on-Beato Garc\'ia (ABG) solution describes a nonlinear electrodynamic nonsingular black hole in general relativity and can be regarded as a strong field correction to the Reissner-Nordstr\"om solution. We exam the possibility of destroying the ABG nonsingular black hole with a test charged particle and a complex scalar field. By comparing with the results of gadenken experiment to destroy the Reissner-Nordstr\"om black hole, we get the nonlinear electrodynamic field effects on the destruction of the event horizon. We obtain the parameter regions of the particle and scalar field, which allow us to destroy the event horizon of an extremal and near-extremal ABG black hole. Our findings show that both can be destroyed due to nonlinear electrodynamic effects. Interestingly, the parameter intervals for the charged particle and scalar field needed to destroy the event horizon of an extremal black hole are identical. Since the ABG black hole is nonsingular, our results remain consistent with the weak cosmic censorship conjecture and may offer a means to explore the interior of a black hole. | [
{
"version": "v1",
"created": "Wed, 11 Sep 2024 14:35:40 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Feb 2025 14:00:29 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 05:41:01 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Miao",
"Wei-Jie",
""
],
[
"Yang",
"Si-Jiang",
""
]
] |
2409.11348 | Adam Bednorz | Tomasz Rybotycki, Tomasz Bia{\l}ecki, Josep Batle, Adam Bednorz | Violation of no-signaling on a public quantum computer | 10 pages, 9 figures, the violation CANNOT be explained by a (simple) ZZ crosstalk, see detailed explanation https://www.fuw.edu.pl/~abednorz/bell/index.html | Adv. Quantum Technol. 2025, 2400661 | 10.1002/qute.202400661 | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | No-signaling is a consequence of the no-communication theorem that states that bipartite systems cannot transfer information
unless a communication channel exists. It is also a by-product of the assumptions of Bell theorem about quantum nonlocality. We have tested no-signaling in bipartite systems of qubits from IBM Quantum devices in extremely large statistics, resulting in significant violations. Although the time and space scales of IBM Quantum cannot in principle rule out subluminal communications, there is no obvious physical mechanism leading to signaling. The violation is also at similar level as observed in Bell tests. It is therefore mandatory to check possible technical imperfections that may cause the violation and to repeat the loophole-free Bell test at much larger statistics, in order to be ruled out definitively at strict spacelike conditions. | [
{
"version": "v1",
"created": "Tue, 17 Sep 2024 16:51:52 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 19:28:42 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Rybotycki",
"Tomasz",
""
],
[
"Białecki",
"Tomasz",
""
],
[
"Batle",
"Josep",
""
],
[
"Bednorz",
"Adam",
""
]
] |
2409.12256 | Elliot Preston-Krebs | Elliot Preston-Krebs, Daniil Lisus, Timothy D. Barfoot | The Finer Points: A Systematic Comparison of Point-Cloud Extractors for Radar Odometry | 8 pages, 2 figures, 1 table. Submitted to the 22nd Conference on Computer and Robot Vision (CRV 2025) | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A key element of many odometry pipelines using spinning frequency-modulated continuous-wave (FMCW) radar is the extraction of a point-cloud from the raw signal. This extraction greatly impacts the overall performance of point-cloud-based odometry. This paper provides a first-of-its-kind, comprehensive comparison of 13 common radar point-cloud extractors for the task of iterative closest point based odometry in autonomous driving environments. Each extractor's parameters are tuned and tested on two FMCW radar datasets using approximately 176km of data from public roads. We find that the simplest, and fastest extractor, K-strongest, is the best overall extractor, consistently outperforming the average by 13.59% and 24.94% on each dataset, respectively. Additionally, we highlight the significance of tuning an extractor and the substantial improvement in odometry accuracy that it yields. | [
{
"version": "v1",
"created": "Wed, 18 Sep 2024 18:36:07 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 18:47:29 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Preston-Krebs",
"Elliot",
""
],
[
"Lisus",
"Daniil",
""
],
[
"Barfoot",
"Timothy D.",
""
]
] |
2409.14052 | Mayank Deora | Mayank Deora, Pinakpani Pal | An average case efficient algorithm for solving two-variable linear Diophantine equations | null | null | null | null | cs.CR cs.DS math.NT | http://creativecommons.org/licenses/by/4.0/ | We revisit two algorithms to solve two-variable linear Diophantine equations. We write the iterative version of one of them. For another, we analyze the number of recursive calls. We find that the number of recursive calls is a periodic function. We derive the period of this function and various upper bounds on this function. We find that for a fixed value of $a$, $b$ and $a$ varying $c$, such that equation $ax+by = c$ (where $a > b$) is solvable, we can find the solution in $O\left(\frac{\log b}{gcd(a,b)}\right)$ average number of recursions. We computationally evaluate and verify this bound. We compare this bound and another bound with the average number of recursive calls in Extended Euclid's algorithm on random $512$-bit inputs. We propose an iterative version of the algorithm. Our algorithm performs better in terms of the average number of iterations against the two existing algorithms. | [
{
"version": "v1",
"created": "Sat, 21 Sep 2024 07:51:12 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Apr 2025 07:05:41 GMT"
},
{
"version": "v3",
"created": "Wed, 21 May 2025 07:04:11 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 07:31:15 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Deora",
"Mayank",
""
],
[
"Pal",
"Pinakpani",
""
]
] |
2409.14469 | Kaikai An | Kaikai An, Shuzheng Si, Helan Hu, Haozhe Zhao, Yuchi Wang, Qingyan Guo, Baobao Chang | Rethinking Semantic Parsing for Large Language Models: Enhancing LLM Performance with Semantic Hints | Accepted by ACL 2025 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Semantic Parsing aims to capture the meaning of a sentence and convert it into a logical, structured form. Previous studies show that semantic parsing enhances the performance of smaller models (e.g., BERT) on downstream tasks. However, it remains unclear whether the improvements extend similarly to LLMs. In this paper, our empirical findings reveal that, unlike smaller models, directly adding semantic parsing results into LLMs reduces their performance. To overcome this, we propose SENSE, a novel prompting approach that embeds semantic hints within the prompt. Experiments show that SENSE consistently improves LLMs' performance across various tasks, highlighting the potential of integrating semantic information to improve LLM capabilities. | [
{
"version": "v1",
"created": "Sun, 22 Sep 2024 14:35:09 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 07:25:49 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"An",
"Kaikai",
""
],
[
"Si",
"Shuzheng",
""
],
[
"Hu",
"Helan",
""
],
[
"Zhao",
"Haozhe",
""
],
[
"Wang",
"Yuchi",
""
],
[
"Guo",
"Qingyan",
""
],
[
"Chang",
"Baobao",
""
]
] |
2409.14599 | Mohammad R. Rezaei | Mohammad R. Rezaei, Milos R. Popovic, Milad Lankarany, Rahul G. Krishnan | Implicit Dynamical Flow Fusion (IDFF) for Generative Modeling | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Conditional Flow Matching (CFM) models can generate high-quality samples from a non-informative prior, but they can be slow, often needing hundreds of network evaluations (NFE). To address this, we propose Implicit Dynamical Flow Fusion (IDFF); IDFF learns a new vector field with an additional momentum term that enables taking longer steps during sample generation while maintaining the fidelity of the generated distribution. Consequently, IDFFs reduce the NFEs by a factor of ten (relative to CFMs) without sacrificing sample quality, enabling rapid sampling and efficient handling of image and time-series data generation tasks. We evaluate IDFF on standard benchmarks such as CIFAR-10 and CelebA for image generation, where we achieve likelihood and quality performance comparable to CFMs and diffusion-based models with fewer NFEs. IDFF also shows superior performance on time-series datasets modeling, including molecular simulation and sea surface temperature (SST) datasets, highlighting its versatility and effectiveness across different domains.\href{https://github.com/MrRezaeiUofT/IDFF}{Github Repository} | [
{
"version": "v1",
"created": "Sun, 22 Sep 2024 21:22:35 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Oct 2024 19:00:33 GMT"
},
{
"version": "v3",
"created": "Mon, 09 Dec 2024 15:06:46 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 13:49:05 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Rezaei",
"Mohammad R.",
""
],
[
"Popovic",
"Milos R.",
""
],
[
"Lankarany",
"Milad",
""
],
[
"Krishnan",
"Rahul G.",
""
]
] |
2409.14608 | Xili Yi | Xili Yi, Jayjun Lee, Nima Fazeli | Visual-auditory Extrinsic Contact Estimation | 16 pages, 9 figures | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Robust manipulation often hinges on a robot's ability to perceive extrinsic contacts-contacts between a grasped object and its surrounding environment. However, these contacts are difficult to observe through vision alone due to occlusions, limited resolution, and ambiguous near-contact states. In this paper, we propose a visual-auditory method for extrinsic contact estimation that integrates global scene information from vision with local contact cues obtained through active audio sensing. Our approach equips a robotic gripper with contact microphones and conduction speakers, enabling the system to emit and receive acoustic signals through the grasped object to detect external contacts. We train our perception pipeline entirely in simulation and zero-shot transfer to the real world. To bridge the sim-to-real gap, we introduce a real-to-sim audio hallucination technique, injecting real-world audio samples into simulated scenes with ground-truth contact labels. The resulting multimodal model accurately estimates both the location and size of extrinsic contacts across a range of cluttered and occluded scenarios. We further demonstrate that explicit contact prediction significantly improves policy learning for downstream contact-rich manipulation tasks. | [
{
"version": "v1",
"created": "Sun, 22 Sep 2024 22:07:13 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 20:38:42 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Yi",
"Xili",
""
],
[
"Lee",
"Jayjun",
""
],
[
"Fazeli",
"Nima",
""
]
] |
2409.15982 | Sergey Kirgizov S. | Jean-Luc Baril and Mireille Bousquet-M\'elou and Sergey Kirgizov and Mehdi Naima | The ascent lattice on Dyck paths | 38 pages | Elec. J. Combinatorics, 32(2), 2025, P2.36 | null | null | math.CO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In the Stanley lattice defined on Dyck paths of size $n$, cover relations are obtained by replacing a valley $DU$ by a peak $UD$. We investigate a greedy version of this lattice, first introduced by Chenevi\`ere, where cover relations replace a factor $DU^k D$ by $U^kD^2$. By relating this poset to another poset recently defined by Nadeau and Tewari, we prove that this still yields a lattice, which we call the ascent lattice, $L_n$. We then count intervals in $L_n$. Their generating function is found to be algebraic of degree $3$. The proof is based on a recursive decomposition of intervals involving two catalytic parameters. The solution of the corresponding functional equation is inspired by recent work on the enumeration of walks confined to a quadrant. We also consider the order induced in $L_{mn}$ on $m$-Dyck paths, that is, paths in which all ascent lengths are multiples of $m$, and on mirrored $m$-Dyck paths, in which all descent lengths are multiples of $m$. The first poset $L_{m,n}$ is still a lattice for any $m$, while the second poset $L'_{m,n}$ is only a join semilattice when $m>1$. In both cases, the enumeration of intervals is still described by an equation in two catalytic variables. Interesting connections arise with the sylvester congruence of Hivert, Novelli and Thibon, and again with walks confined to a quadrant. We combine the latter connection with probabilistic results to give asymptotic estimates of the number of intervals in both $L_{m,n}$ and $L'_{m,n}$. Their form implies that the generating functions of intervals are no longer algebraic, nor even D-finite, when $m>1$. | [
{
"version": "v1",
"created": "Tue, 24 Sep 2024 11:27:42 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 14:36:15 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Baril",
"Jean-Luc",
""
],
[
"Bousquet-Mélou",
"Mireille",
""
],
[
"Kirgizov",
"Sergey",
""
],
[
"Naima",
"Mehdi",
""
]
] |
2409.17716 | Tiago De Souza Farias | Tiago de Souza Farias, Lucas Friedrich, Jonas Maziero | QuForge: A Library for Qudits Simulation | null | Farias, T.S., Friedrich, L., Maziero, J. QuForge: A Library for Qudits Simulation. Computer Physics Communications, vol. 314, Sept. 2025, p. 109687. ScienceDirect | 10.1016/j.cpc.2025.109687 | null | quant-ph cs.LG | http://creativecommons.org/licenses/by/4.0/ | Quantum computing with qudits, an extension of qubits to multiple levels, is a research field less mature than qubit-based quantum computing. However, qudits can offer some advantages over qubits, by representing information with fewer separated components. In this article, we present QuForge, a Python-based library designed to simulate quantum circuits with qudits. This library provides the necessary quantum gates for implementing quantum algorithms, tailored to any chosen qudit dimension. Built on top of differentiable frameworks, QuForge supports execution on accelerating devices such as GPUs and TPUs, significantly speeding up simulations. It also supports sparse operations, leading to a reduction in memory consumption compared to other libraries. Additionally, by constructing quantum circuits as differentiable graphs, QuForge facilitates the implementation of quantum machine learning algorithms, enhancing the capabilities and flexibility of quantum computing research. | [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 10:38:35 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 17:02:26 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Farias",
"Tiago de Souza",
""
],
[
"Friedrich",
"Lucas",
""
],
[
"Maziero",
"Jonas",
""
]
] |
2409.18594 | Ricardo Knauer | Ricardo Knauer, Mario Koddenbrock, Raphael Wallsberger, Nicholas M. Brisson, Georg N. Duda, Deborah Falla, David W. Evans, Erik Rodner | "Oh LLM, I'm Asking Thee, Please Give Me a Decision Tree": Zero-Shot Decision Tree Induction and Embedding with Large Language Models | KDD 2025 Research Track | null | 10.1145/3711896.3736818 | null | cs.AI cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) provide powerful means to leverage prior knowledge for predictive modeling when data is limited. In this work, we demonstrate how LLMs can use their compressed world knowledge to generate intrinsically interpretable machine learning models, i.e., decision trees, without any training data. We find that these zero-shot decision trees can even surpass data-driven trees on some small-sized tabular datasets and that embeddings derived from these trees perform better than data-driven tree-based embeddings on average. Our decision tree induction and embedding approaches can therefore serve as new knowledge-driven baselines for data-driven machine learning methods in the low-data regime. Furthermore, they offer ways to harness the rich world knowledge within LLMs for tabular machine learning tasks. Our code and results are available at https://github.com/ml-lab-htw/llm-trees. | [
{
"version": "v1",
"created": "Fri, 27 Sep 2024 09:53:48 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 09:01:46 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Knauer",
"Ricardo",
""
],
[
"Koddenbrock",
"Mario",
""
],
[
"Wallsberger",
"Raphael",
""
],
[
"Brisson",
"Nicholas M.",
""
],
[
"Duda",
"Georg N.",
""
],
[
"Falla",
"Deborah",
""
],
[
"Evans",
"David W.",
""
],
[
"Rodner",
"Erik",
""
]
] |
2409.19885 | Ederson Moreira dos Santos | Eduardo de Souza B\"oer and Ederson Moreira dos Santos | Standing waves for nonlinear Hartree type equations: existence and qualitative properties | 35 pages, 7 figures, few typos were fixed | Calculus of Variations and Partial Differential Equations - 2025 | 10.1007/s00526-025-03025-2 | null | math.AP | http://creativecommons.org/licenses/by/4.0/ | We consider systems of the form \[ \left\{ \begin{array}{l} -\Delta u + u = \frac{2p}{p+q}(I_\alpha \ast |v|^{q})|u|^{p-2}u \ \ \textrm{ in } \mathbb{R}^N, \\ -\Delta v + v = \frac{2q}{p+q}(I_\alpha \ast |u|^{p})|v|^{q-2}v \ \ \textrm{ in } \mathbb{R}^N, \end{array} \right. \] for $\alpha\in (0, N)$, $\max\left\{\frac{2\alpha}{N}, 1\right\} < p, q < 2^*$ and $\frac{2(N+\alpha)}{N} < p+ q < 2^{*}_{\alpha}$, where $I_\alpha$ denotes the Riesz potential, \[ 2^* = \left\{ \begin{array}{l}\frac{2N}{N-2} \ \ \text{for} \ \ N\geq 3,\\ +\infty \ \ \text{for} \ \ N =1,2, \end{array}\right. \quad \text{and} \quad 2^*_{\alpha} = \left\{ \begin{array}{l}\frac{2(N+\alpha)}{N-2} \ \ \text{for} \ \ N\geq 3,\\ +\infty \ \ \text{for} \ \ N =1,2. \end{array} \right. \] This type of systems arises in the study of standing wave solutions for a certain approximation of the Hartree theory for a two-component attractive interaction. We prove existence and some qualitative properties for ground state solutions, such as definite sign for each component, radial symmetry and sharp asymptotic decay at infinity, and a regularity/integrability result for the (weak) solutions. Moreover, we show that the straight lines $p+q=\frac{2(N+\alpha)}{N}$ and $ p+ q = 2^{*}_{\alpha}$ are critical for the existence of solutions. | [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 02:30:21 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Oct 2024 16:55:03 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Böer",
"Eduardo de Souza",
""
],
[
"Santos",
"Ederson Moreira dos",
""
]
] |
2409.20319 | Lei Pan | Xuanpu Yang, Xiang-Ping Jiang, Zijun Wei, Yucheng Wang, and Lei Pan | Dissipation induced transition between delocalization and localization in the three-dimensional Anderson model | 10 pages, 5 figures, comments are welcome | Phys. Rev. B 111, 134203 (2025) | 10.1103/PhysRevB.111.134203 | null | cond-mat.dis-nn | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the probable delocalization-localization transition in open quantum systems with disorder. The disorder can induce localization in isolated quantum systems and it is generally recognized that localization is fragile under the action of dissipations from the external environment due to its interfering nature. Recent work [Y. Liu, et al, Phys. Rev. Lett. 132, 216301 (2024).] found that a one-dimensional quasiperiodic system can be driven into the localization phase by a tailored local dissipation where a dissipation-induced delocalized-localized transition is proposed. Based on this, we consider a more realistic system and show that a dissipation-induced transition between delocalization and localization appears in the three-dimensional (3D) Anderson model. By tuning local dissipative operators acting on nearest neighboring sites, we find that the system can relax to localized states dominated steady state instead of the choice of initial conditions and dissipation strengths. Moreover, we can also realize a delocalized states predominated steady-state from a localized initial state by using a kind of dissipation operators acting on next nearest neighboring sites. Our results enrich the applicability of dissipation-induced localization and identify the transition between delocalized and localized phases in 3D disordered systems. | [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 14:21:13 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Apr 2025 04:42:52 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Yang",
"Xuanpu",
""
],
[
"Jiang",
"Xiang-Ping",
""
],
[
"Wei",
"Zijun",
""
],
[
"Wang",
"Yucheng",
""
],
[
"Pan",
"Lei",
""
]
] |
2409.20547 | Dongze Wu | Dongze Wu, Yao Xie | Annealing Flow Generative Models Towards Sampling High-Dimensional and Multi-Modal Distributions | This paper has been accepted to ICML 2025 and will appear in the Proceedings of Machine Learning Research (PMLR) | null | null | null | stat.ML cs.LG stat.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sampling from high-dimensional, multi-modal distributions remains a fundamental challenge across domains such as statistical Bayesian inference and physics-based machine learning. In this paper, we propose Annealing Flow (AF), a method built on Continuous Normalizing Flow (CNF) for sampling from high-dimensional and multi-modal distributions. AF is trained with a dynamic Optimal Transport (OT) objective incorporating Wasserstein regularization, and guided by annealing procedures, facilitating effective exploration of modes in high-dimensional spaces. Compared to recent NF methods, AF greatly improves training efficiency and stability, with minimal reliance on MC assistance. We demonstrate the superior performance of AF compared to state-of-the-art methods through experiments on various challenging distributions and real-world datasets, particularly in high-dimensional and multi-modal settings. We also highlight AF potential for sampling the least favorable distributions. | [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 17:48:22 GMT"
},
{
"version": "v2",
"created": "Mon, 25 Nov 2024 23:05:27 GMT"
},
{
"version": "v3",
"created": "Thu, 12 Dec 2024 01:11:07 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 17:47:15 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Wu",
"Dongze",
""
],
[
"Xie",
"Yao",
""
]
] |
2410.00677 | Gregor Pasemann | Gregor Pasemann and Markus Rei{\ss} | Nonparametric Diffusivity Estimation for the Stochastic Heat Equation from Noisy Observations | null | null | null | null | math.ST stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We estimate nonparametrically the spatially varying diffusivity of a stochastic heat equation from observations perturbed by additional noise. To that end, we employ a two-step localization procedure, more precisely, we combine local state estimates into a locally linear regression approach. Our analysis relies on quantitative Trotter--Kato type approximation results for the heat semigroup that are of independent interest. The presence of observational noise leads to non-standard scaling behaviour of the model. Numerical simulations illustrate the results. | [
{
"version": "v1",
"created": "Tue, 01 Oct 2024 13:35:07 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 11:04:35 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Pasemann",
"Gregor",
""
],
[
"Reiß",
"Markus",
""
]
] |
2410.01082 | Juan Daniel Sebastia-Saez | Daniel Sebastia-Saez and Tao Chen and Cynthia Yu-Wai-Man | A Computational Fluid Dynamics study of drug-releasing ocular implants for glaucoma treatment: Comparison of implant size and locations | null | null | null | null | q-bio.TO | http://creativecommons.org/licenses/by/4.0/ | Drug-releasing implants are gaining momentum in the treatment of glaucoma. Implants present however several limitations. Among these limitations, there is the inability to provide an even distribution of the drug in the trabecular meshwork. CFD simulations were used in this work to study the interplay between ocular fluid dynamics and drug diffusion to explore the options to achieve a homogeneous drug distribution. The analysis finds that the balance between convection and diffusion flux hinders mixing within the eye's anterior chamber. This results in highly localized drug delivery in the trabecular meshwork when using a gravity-driven location implant. The results also show that varying the size of the implant can help to solve the issue. Given the natural variability of the trabecular meshwork size amongst patients, implants with personalized size may become a potential solution. The location of the implant within the eye is also key to effective drug delivery. Natural laminar flow of aqueous humor within the eye's anterior chamber prevents achieving an even drug distribution at the target tissue for gravity-driven location implants. However, the iris-lens gap can be utilized as a natural mixer when placing the implant in the eye's posterior chamber, thus helping with effective delivery. | [
{
"version": "v1",
"created": "Tue, 01 Oct 2024 21:22:27 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 17:17:52 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Sebastia-Saez",
"Daniel",
""
],
[
"Chen",
"Tao",
""
],
[
"Yu-Wai-Man",
"Cynthia",
""
]
] |
2410.01635 | Xiangguo Sun | Qunzhong Wang, Xiangguo Sun, Hong Cheng | Does Graph Prompt Work? A Data Operation Perspective with Theoretical Analysis | accepted by ICML 2025 | null | null | null | cs.LG cs.AI cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, graph prompting has emerged as a promising research direction, enabling the learning of additional tokens or subgraphs appended to the original graphs without requiring retraining of pre-trained graph models across various applications. This novel paradigm, shifting from the traditional pretraining and finetuning to pretraining and prompting has shown significant empirical success in simulating graph data operations, with applications ranging from recommendation systems to biological networks and graph transferring. However, despite its potential, the theoretical underpinnings of graph prompting remain underexplored, raising critical questions about its fundamental effectiveness. The lack of rigorous theoretical proof of why and how much it works is more like a dark cloud over the graph prompt area to go further. To fill this gap, this paper introduces a theoretical framework that rigorously analyzes graph prompting from a data operation perspective. Our contributions are threefold: First, we provide a formal guarantee theorem, demonstrating graph prompts capacity to approximate graph transformation operators, effectively linking upstream and downstream tasks. Second, we derive upper bounds on the error of these data operations by graph prompts for a single graph and extend this discussion to batches of graphs, which are common in graph model training. Third, we analyze the distribution of data operation errors, extending our theoretical findings from linear graph models (e.g., GCN) to non-linear graph models (e.g., GAT). Extensive experiments support our theoretical results and confirm the practical implications of these guarantees. | [
{
"version": "v1",
"created": "Wed, 02 Oct 2024 15:07:13 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 04:43:17 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Wang",
"Qunzhong",
""
],
[
"Sun",
"Xiangguo",
""
],
[
"Cheng",
"Hong",
""
]
] |
2410.01651 | Xiang Hu | Xiang Hu, Zhihao Teng, Jun Zhao, Wei Wu, Kewei Tu | Efficient Length-Generalizable Attention via Causal Retrieval for Long-Context Language Modeling | accepted to ICML 2025 | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite the success of Transformers, handling long contexts remains challenging due to the limited length generalization and quadratic complexity of self-attention. Thus Transformers often require post-training with a larger attention window, significantly increasing computational and memory costs. In this paper, we propose a novel attention mechanism based on dynamic context, Grouped Cross Attention (GCA), which can generalize to 1000 times the pre-training context length while maintaining the ability to access distant information with a constant attention window size. For a given input sequence, we split it into chunks and use each chunk to retrieve top-k relevant past chunks for subsequent text generation. Specifically, unlike most previous works that use an off-the-shelf retriever, our key innovation allows the retriever to learn how to retrieve past chunks that better minimize the auto-regressive loss of subsequent tokens in an end-to-end manner. Such a mechanism accommodates retrieved chunks with a fixed-size attention window to achieve long-range information access, significantly reducing computational and memory costs during training and inference. Experiments show that GCA-based models achieve near-perfect accuracy in passkey retrieval for 16M context lengths, which is 1000 times the training length. | [
{
"version": "v1",
"created": "Wed, 02 Oct 2024 15:18:34 GMT"
},
{
"version": "v2",
"created": "Mon, 27 Jan 2025 07:08:49 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 12:49:47 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Hu",
"Xiang",
""
],
[
"Teng",
"Zhihao",
""
],
[
"Zhao",
"Jun",
""
],
[
"Wu",
"Wei",
""
],
[
"Tu",
"Kewei",
""
]
] |
2410.01894 | Joshua Mundinger | Joshua Mundinger | On the differentials of the Hochschild-Kostant-Rosenberg spectral sequence | 40 pages. v2: rewrote section 4.2 on the shifted tangent bundle on a prestack, along with other minor edits. Comments welcome! | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Hochschild-Kostant-Rosenberg theorem implies the existence of a spectral sequence computing the Hochschild homology of a variety in terms of the cohomology of differential forms. When the base field $k$ has characteristic $p>0$, we show that the differentials in this spectral sequence are zero before page $p$; when the variety admits a lift to $W_2(k)$, we give a formula for the differential on page $p$. The formula involves the Bockstein associated to the lift and a $p$th power operation for the Atiyah class. | [
{
"version": "v1",
"created": "Wed, 02 Oct 2024 18:00:10 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 22:00:34 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Mundinger",
"Joshua",
""
]
] |
2410.02790 | Robin Burchard | Hrishikesh Balkrishna Karande, Ravikiran Arasur Thippeswamy Shivalingappa, Abdelhafid Nassim Yaici, Iman Haghbin, Niravkumar Bavadiya, Robin Burchard and Kristof Van Laerhoven | Raising the Bar(ometer): Identifying a User's Stair and Lift Usage Through Wearable Sensor Data Analysis | submitted to iWOAR 2024 | null | 10.1007/978-3-031-80856-2_14 | null | eess.SP cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Many users are confronted multiple times daily with the choice of whether to take the stairs or the elevator. Whereas taking the stairs could be beneficial for cardiovascular health and wellness, taking the elevator might be more convenient but it also consumes energy. By precisely tracking and boosting users' stairs and elevator usage through their wearable, users might gain health insights and motivation, encouraging a healthy lifestyle and lowering the risk of sedentary-related health problems. This research describes a new exploratory dataset, to examine the patterns and behaviors related to using stairs and lifts. We collected data from 20 participants while climbing and descending stairs and taking a lift in a variety of scenarios. The aim is to provide insights and demonstrate the practicality of using wearable sensor data for such a scenario. Our collected dataset was used to train and test a Random Forest machine learning model, and the results show that our method is highly accurate at classifying stair and lift operations with an accuracy of 87.61% and a multi-class weighted F1-score of 87.56% over 8-second time windows. Furthermore, we investigate the effect of various types of sensors and data attributes on the model's performance. Our findings show that combining inertial and pressure sensors yields a viable solution for real-time activity detection. | [
{
"version": "v1",
"created": "Wed, 18 Sep 2024 06:26:50 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Karande",
"Hrishikesh Balkrishna",
""
],
[
"Shivalingappa",
"Ravikiran Arasur Thippeswamy",
""
],
[
"Yaici",
"Abdelhafid Nassim",
""
],
[
"Haghbin",
"Iman",
""
],
[
"Bavadiya",
"Niravkumar",
""
],
[
"Burchard",
"Robin",
""
],
[
"Van Laerhoven",
"Kristof",
""
]
] |
2410.03549 | Robin Burchard | Robin Burchard and Kristof Van Laerhoven | Multi-modal Atmospheric Sensing to Augment Wearable IMU-Based Hand Washing Detection | iWOAR2024 | null | 10.1007/978-3-031-80856-2_4 | null | cs.HC cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Hand washing is a crucial part of personal hygiene. Hand washing detection is a relevant topic for wearable sensing with applications in the medical and professional fields. Hand washing detection can be used to aid workers in complying with hygiene rules. Hand washing detection using body-worn IMU-based sensor systems has been shown to be a feasible approach, although, for some reported results, the specificity of the detection was low, leading to a high rate of false positives. In this work, we present a novel, open-source prototype device that additionally includes a humidity, temperature, and barometric sensor. We contribute a benchmark dataset of 10 participants and 43 hand-washing events and perform an evaluation of the sensors' benefits. Added to that, we outline the usefulness of the additional sensor in both the annotation pipeline and the machine learning models. By visual inspection, we show that especially the humidity sensor registers a strong increase in the relative humidity during a hand-washing activity. A machine learning analysis of our data shows that distinct features benefiting from such relative humidity patterns remain to be identified. | [
{
"version": "v1",
"created": "Tue, 17 Sep 2024 14:04:56 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Burchard",
"Robin",
""
],
[
"Van Laerhoven",
"Kristof",
""
]
] |
2410.04756 | Wooseong Yang | Wooseong Yang, Chen Wang, Zihe Song, Weizhi Zhang, Philip S. Yu | Item Cluster-aware Prompt Learning for Session-based Recommendation | 10 pages | null | null | null | cs.IR cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Session-based recommendation (SBR) aims to capture dynamic user preferences by analyzing item sequences within individual sessions. However, most existing approaches focus mainly on intra-session item relationships, neglecting the connections between items across different sessions (inter-session relationships), which limits their ability to fully capture complex item interactions. While some methods incorporate inter-session information, they often suffer from high computational costs, leading to longer training times and reduced efficiency. To address these challenges, we propose the CLIP-SBR (Cluster-aware Item Prompt learning for Session-Based Recommendation) framework. CLIP-SBR is composed of two modules: 1) an item relationship mining module that builds a global graph to effectively model both intra- and inter-session relationships, and 2) an item cluster-aware prompt learning module that uses soft prompts to integrate these relationships into SBR models efficiently. We evaluate CLIP-SBR across eight SBR models and three benchmark datasets, consistently demonstrating improved recommendation performance and establishing CLIP-SBR as a robust solution for session-based recommendation tasks. | [
{
"version": "v1",
"created": "Mon, 07 Oct 2024 05:20:21 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 15:12:41 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Yang",
"Wooseong",
""
],
[
"Wang",
"Chen",
""
],
[
"Song",
"Zihe",
""
],
[
"Zhang",
"Weizhi",
""
],
[
"Yu",
"Philip S.",
""
]
] |
2410.04941 | Irene Cannistraci | Irene Cannistraci, Simone Antonelli, Emanuele Palumbo, Thomas M. Sutter, Emanuele Rodol\`a, Bastian Rieck, Julia E. Vogt | Foundation Models on a Budget: Approximating Blocks in Large Vision Models | 25 pages, 14 figures, 16 tables | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Foundation Models have shown impressive performance in various tasks and domains, yet they require massive computational resources, raising concerns about accessibility and sustainability. Previous attempts to reduce foundation model size fall short of fully addressing the problem, as they end up increasing computational load through additional training steps. Recent works reveal that deep neural networks exhibit internal representation similarities. While inter-network similarities have enabled techniques such as model stitching and merging, intra-network similarities remain underexplored for improving efficiency. In this paper, we propose Transformer Blocks Approximation (TBA), a novel method that leverages intra-network similarities to identify and approximate transformer blocks in large vision models. TBA replaces these blocks using lightweight, closed-form transformations, without retraining or fine-tuning the rest of the model. The proposed method reduces the number of parameters while having minimal impact on the downstream task. We validate the effectiveness and generalizability of TBA through extensive experiments across multiple datasets (e.g., Imagenet-1k and CIFAR100) and state-of-the-art pretrained vision models (e.g, ViT, DiNO-v2, and DEiT). | [
{
"version": "v1",
"created": "Mon, 07 Oct 2024 11:35:24 GMT"
},
{
"version": "v2",
"created": "Tue, 08 Oct 2024 06:59:45 GMT"
},
{
"version": "v3",
"created": "Thu, 10 Oct 2024 09:00:55 GMT"
},
{
"version": "v4",
"created": "Fri, 11 Oct 2024 08:43:56 GMT"
},
{
"version": "v5",
"created": "Tue, 27 May 2025 16:22:32 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Cannistraci",
"Irene",
""
],
[
"Antonelli",
"Simone",
""
],
[
"Palumbo",
"Emanuele",
""
],
[
"Sutter",
"Thomas M.",
""
],
[
"Rodolà",
"Emanuele",
""
],
[
"Rieck",
"Bastian",
""
],
[
"Vogt",
"Julia E.",
""
]
] |
2410.05754 | Daniel Barzilai | Daniel Barzilai, Ohad Shamir | Simple Relative Deviation Bounds for Covariance and Gram Matrices | Added some references to version 1 | null | null | null | math.PR cs.LG math.ST stat.ML stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We provide non-asymptotic, relative deviation bounds for the eigenvalues of empirical covariance and Gram matrices in general settings. Unlike typical uniform bounds, which may fail to capture the behavior of smaller eigenvalues, our results provide sharper control across the spectrum. Our analysis is based on a general-purpose theorem that allows one to convert existing uniform bounds into relative ones. The theorems and techniques emphasize simplicity and should be applicable across various settings. | [
{
"version": "v1",
"created": "Tue, 08 Oct 2024 07:28:56 GMT"
},
{
"version": "v2",
"created": "Wed, 27 Nov 2024 07:22:55 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 08:54:01 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Barzilai",
"Daniel",
""
],
[
"Shamir",
"Ohad",
""
]
] |
2410.05823 | Claudio Bonati | Claudio Bonati and Andrea Pelissetto and Ettore Vicari | Three-dimensional Abelian and non-Abelian gauge Higgs theories | 117 pages, 50 pdf figures. References and comments added. Accepted for publication by Physics Reports | null | null | null | cond-mat.stat-mech hep-lat hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Gauge symmetries and Higgs mechanisms are key features of theories describing high-energy particle physics and collective phenomena in statistical and condensed-matter physics. In this review we address the collective behavior of systems of multicomponent scalar fields interacting with gauge fields, which can be already present in the underlying microscopic system or emerge only at criticality. The interplay between local gauge and global symmetries determines the phase diagram, the nature of the Higgs phases, and the nature of phase transitions between the high-temperature disordered and the low-temperature Higgs phases. However, additional crucial features determine the universal properties of the critical behavior at continuous transitions. Specifically, their nature also depends on the role played by the gauge modes at criticality. Effective (Abelian or non-Abelian) gauge Higgs field theories emerge when gauge modes develop critical correlations. On the other hand, a more standard critical behavior, which admits an effective description in terms of Landau-Ginzburg-Wilson $\Phi^4$ theories, occurs when gauge-field modes are short ranged at the transition. In the latter case, gauge fields only prevent non-gauge invariant correlation functions from becoming critical. This review covers the recent progress made in the study of Higgs systems with Abelian and non-Abelian gauge fields. We discuss the equilibrium thermodynamic properties of systems with a classical partition function, focusing mainly on three-dimensional systems, and only briefly discussing two-dimensional models. However, by using the quantum-to-classical mapping, the results on the critical behavior for classical systems in $D=d+1$ dimensions can be extended to quantum transitions in $d$ dimensions. | [
{
"version": "v1",
"created": "Tue, 08 Oct 2024 08:52:10 GMT"
},
{
"version": "v2",
"created": "Thu, 07 Nov 2024 13:04:59 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 09:12:32 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Bonati",
"Claudio",
""
],
[
"Pelissetto",
"Andrea",
""
],
[
"Vicari",
"Ettore",
""
]
] |
2410.06129 | Kylie Yeung | Kylie Yeung, Christine Tobler, Rolf F Schulte, Benjamin White, Anthony McIntyre, Sebastien Serres, Peter Morris, Dorothee Auer, Fergus V Gleeson, Damian J Tyler, James T Grist, Florian Wiesinger | Exploring the Computational Feasibility of Direct Pseudoinversion of the Encoding Matrix for MR Image Reconstruction (Pinv-Recon) | 28 pages, 8 figures (+ Supplementary Material). Submitted to Scientific Reports. V3 corrects wrong labels on Figure 3 | null | null | null | physics.med-ph eess.IV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Image reconstruction in Magnetic Resonance Imaging (MRI) is fundamentally a linear inverse problem, such that the image can be recovered via explicit pseudoinversion of the encoding matrix by solving $\textbf{data} = \textbf{Encode} \times \textbf{image}$ - a method referred to here as Pinv-Recon. While the benefits of this approach were acknowledged in early studies, the field has historically favored fast Fourier transforms (FFT) and iterative techniques due to perceived computational limitations of the pseudoinversion approach. This work revisits Pinv-Recon in the context of modern hardware, software, and optimized linear algebra routines. We compare various matrix inversion strategies, assess regularization effects, and demonstrate incorporation of advanced encoding physics into a unified reconstruction framework.
While hardware advances have already significantly reduced computation time compared to earlier studies, our work further demonstrates that leveraging Cholesky decomposition and block-wise inversion leads to a two-order-of-magnitude improvement in computational efficiency over previous Singular Value Decomposition-based implementations. Moreover, we demonstrate the versatility of Pinv-Recon on diverse \textit{in vivo} datasets encompassing a range of encoding schemes, starting with low- to medium-resolution functional and metabolic imaging and extending to high-resolution cases. Our findings establish Pinv-Recon as a practical and adaptable reconstruction method that aligns with the increasing emphasis on open-source and reproducible MRI research. | [
{
"version": "v1",
"created": "Tue, 08 Oct 2024 15:31:41 GMT"
},
{
"version": "v2",
"created": "Fri, 02 May 2025 17:08:32 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 15:53:23 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Yeung",
"Kylie",
""
],
[
"Tobler",
"Christine",
""
],
[
"Schulte",
"Rolf F",
""
],
[
"White",
"Benjamin",
""
],
[
"McIntyre",
"Anthony",
""
],
[
"Serres",
"Sebastien",
""
],
[
"Morris",
"Peter",
""
],
[
"Auer",
"Dorothee",
""
],
[
"Gleeson",
"Fergus V",
""
],
[
"Tyler",
"Damian J",
""
],
[
"Grist",
"James T",
""
],
[
"Wiesinger",
"Florian",
""
]
] |
2410.06638 | Kaishuai Xu | Kaishuai Xu, Tiezheng Yu, Wenjun Hou, Yi Cheng, Chak Tou Leong, Liangyou Li, Xin Jiang, Lifeng Shang, Qun Liu, Wenjie Li | Subtle Errors in Reasoning: Preference Learning via Error-injected Self-editing | accepted as ACL 2025 main | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) have exhibited strong mathematical reasoning prowess, tackling tasks ranging from basic arithmetic to advanced competition-level problems. However, frequently occurring subtle yet critical errors, such as miscalculations or incorrect substitutions, limit the LLMs' full potential. Existing studies to improve mathematical ability typically involve applying preference learning to step-wise solution pairs. Although these methods leverage samples of varying granularity to mitigate reasoning errors, they overlook critical subtle errors. In this work, we propose a novel preference learning framework called eRror-Injected Self-Editing (RISE), which injects predefined subtle errors into pivotal tokens in reasoning or computation steps to construct hard pairs for error mitigation. In detail, RISE uses the LLM itself to edit a small number of tokens in the solution, injecting designed subtle errors. Then, pairs composed of self-edited solutions and their corresponding correct ones, along with pairs of correct and incorrect solutions obtained through sampling, are used together for subtle error-aware DPO training. Compared with other preference learning methods, RISE further refines the training objective without requiring fine-grained sampling or preference annotation. Extensive experiments validate the effectiveness of RISE, with preference learning on Qwen2-7B-Instruct yielding notable improvements of 3.0% on GSM8K and 7.9% on MATH with only 4.5K training samples. Moreover, the effect of error mitigation extends from mathematical reasoning to logical reasoning and code generation. | [
{
"version": "v1",
"created": "Wed, 09 Oct 2024 07:43:38 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 06:53:40 GMT"
},
{
"version": "v3",
"created": "Mon, 03 Mar 2025 07:09:42 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 03:32:00 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Xu",
"Kaishuai",
""
],
[
"Yu",
"Tiezheng",
""
],
[
"Hou",
"Wenjun",
""
],
[
"Cheng",
"Yi",
""
],
[
"Leong",
"Chak Tou",
""
],
[
"Li",
"Liangyou",
""
],
[
"Jiang",
"Xin",
""
],
[
"Shang",
"Lifeng",
""
],
[
"Liu",
"Qun",
""
],
[
"Li",
"Wenjie",
""
]
] |
2410.07267 | Pengcheng Ai | Pengcheng Ai, Xiangming Sun, Zhi Deng, Xinchi Ran | Scintillation pulse characterization with spectrum-inspired temporal neural networks: case studies on particle detector signals | 30 pages, 14 figures | null | 10.1140/epjp/s13360-025-06385-y | null | physics.ins-det cs.LG physics.data-an | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Particle detectors based on scintillators are widely used in high-energy physics and astroparticle physics experiments, nuclear medicine imaging, industrial and environmental detection, etc. Precisely extracting scintillation signal characteristics at the event level is important for these applications, not only in respect of understanding the scintillator itself, but also kinds and physical property of incident particles. Recent researches demonstrate data-driven neural networks surpass traditional statistical methods, especially when the analytical form of signals is hard to obtain, or noise is significant. However, most densely connected or convolution-based networks fail to fully exploit the spectral and temporal structure of scintillation signals, leaving large space for performance improvement. In this paper, we propose a network architecture specially tailored for scintillation pulse characterization based on previous works on time series analysis. The core insight is that, by directly applying Fast Fourier Transform on original signals and utilizing different frequency components, the proposed network architecture can serve as a lightweight and enhanced representation learning backbone. We prove our idea in two case studies: (a) simulation data generated with the setting of the LUX dark matter detector, and (b) experimental electrical signals with fast electronics to emulate scintillation variations for the NICA/MPD calorimeter. The proposed model achieves significantly better results than the reference model in literature and densely connected models and demonstrates higher cost-efficiency than conventional machine learning methods. | [
{
"version": "v1",
"created": "Wed, 09 Oct 2024 02:44:53 GMT"
},
{
"version": "v2",
"created": "Mon, 03 Mar 2025 08:45:31 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 02:30:26 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Ai",
"Pengcheng",
""
],
[
"Sun",
"Xiangming",
""
],
[
"Deng",
"Zhi",
""
],
[
"Ran",
"Xinchi",
""
]
] |
2410.07399 | Hunter Dinkins | Jeffrey Ayers and Hunter Dinkins | Wreath Macdonald polynomials, quiver varieties, and quasimap counts | null | null | null | null | math.AG hep-th math-ph math.CO math.MP math.RT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the $K$-theoretic enumerative geometry of cyclic Nakajima quiver varieties, with particular focus on $\text{Hilb}^{m}([\mathbb{C}^{2}/\mathbb{Z}_{l}])$, the equivariant Hilbert scheme of points on $\mathbb{C}^2$. The direct sum over $m$ of the equivariant $K$-theories of these varieties is known to be isomorphic to the ring symmetric functions in $l$ colors, with structure sheaves of torus fixed points identified with wreath Macdonald polynomials. Using properties of wreath Macdonald polynomials and the recent identification of the Maulik-Okounkov quantum affine algebra for cyclic quivers with the quantum toroidal algebras of type $A$, we derive an explicit formula for the generating function of capped vertex functions of $\text{Hilb}^{m}([\mathbb{C}^{2}/\mathbb{Z}_{l}])$ with descendants given by exterior powers of the $0$th tautological bundle. We also sharpen the large framing vanishing results of Okounkov, providing a class of descendants and cyclic quiver varieties for which the capped vertex functions are purely classical. | [
{
"version": "v1",
"created": "Wed, 09 Oct 2024 19:55:49 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 18:39:31 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Ayers",
"Jeffrey",
""
],
[
"Dinkins",
"Hunter",
""
]
] |
2410.07467 | Atli Fannar Frankl\'in | Atli Fannar Frankl\'in | Pattern Avoiding Permutations Enumerated by Inversions | Update: Address review comments after submission to DMTCS Update: Copy version for DMTCS | Discrete Mathematics & Theoretical Computer Science, vol. 27:1, Permutation Patterns 2024, Special issues (May 26, 2025) dmtcs:14437 | 10.46298/dmtcs.14437 | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Permutations are usually enumerated by size, but new results can be found by enumerating them by inversions instead, in which case one must restrict one's attention to indecomposable permutations. In the style of the seminal paper by Simion and Schmidt, we investigate all combinations of permutation patterns of length at most 3. | [
{
"version": "v1",
"created": "Wed, 09 Oct 2024 22:10:35 GMT"
},
{
"version": "v2",
"created": "Sat, 12 Oct 2024 09:23:22 GMT"
},
{
"version": "v3",
"created": "Mon, 19 May 2025 11:08:51 GMT"
},
{
"version": "v4",
"created": "Fri, 23 May 2025 16:25:17 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Franklín",
"Atli Fannar",
""
]
] |
2410.07587 | Keiju Murata Dr | Shunichiro Kinoshita, Keiju Murata, Daisuke Yamamoto, Ryosuke Yoshii | Spin systems as quantum simulators of quantum field theories in curved spacetimes | 16 pages, 3 figures, v2: Typos fixed, reference added | null | null | null | hep-th cond-mat.quant-gas cond-mat.stat-mech gr-qc quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We demonstrate that a quantum field theory (QFT) in general two-dimensional curved spacetimes can be realized by a system of quantum spins or qubits. We consider a spin-1/2 model on a one-dimensional ring with spatially and temporally varying exchange couplings and magnetic fields. This model reduces to a QFT of Majorana fermions in the continuum limit. From this correspondence, we establish a dictionary for translating between the spacetime-dependent parameters of the spin model and the general metric on which the QFT is defined. After addressing the general case, we consider the Friedmann-Lema\^{\i}tre-Robertson-Walker (FLRW) metric as a simple example. According to the dictionary, the QFT of Majorana fermions on the FLRW metric corresponds to the Ising model with a time-dependent transverse magnetic field. We demonstrate that the production of Majorana particles in the expanding universe can be simulated with the transverse-field Ising model by increasing the strength of the magnetic field. Furthermore, we examine the Unruh effect through the spin system by using our prescription and show the direct relation between the entanglement (or modular) Hamiltonian in the spin system and the Rindler Hamiltonian. This approach provides an experimentally viable system for probing various phenomena in QFT within curved spacetime, while also opening the door to uncovering nontrivial phenomena in spin systems inspired by curved spacetime physics. It offers fresh perspectives on both QFT in curved spacetimes and quantum many-body spin systems, revealing profound connections between these fields. | [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 03:38:26 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 05:44:19 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Kinoshita",
"Shunichiro",
""
],
[
"Murata",
"Keiju",
""
],
[
"Yamamoto",
"Daisuke",
""
],
[
"Yoshii",
"Ryosuke",
""
]
] |
2410.07655 | Ziming Shi | Ziming Shi | Sobolev and H\"older estimates for the $\overline \partial$ equation on pseudoconvex domains of finite type in $\mathbb C^2$ | 32 pages, published in J. Math. Anal. Appl | null | null | null | math.CV math.AP | http://creativecommons.org/licenses/by/4.0/ | We prove a homotopy formula which yields almost sharp estimates in all (positive-indexed) Sobolev and H\"older-Zygmund spaces for the $\overline \partial$ equation on pseudoconvex domains of finite type in $\mathbb C^2$, extending the earlier results of Fefferman-Kohn (1988), Range (1990), and Chang-Nagel-Stein (1992). The main novelty of our proof is the construction of holomorphic support functions that admit precise estimates when the parameter variable lies in a thin shell outside the domain. | [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 06:50:10 GMT"
},
{
"version": "v2",
"created": "Sat, 07 Dec 2024 19:57:11 GMT"
},
{
"version": "v3",
"created": "Mon, 26 May 2025 20:12:47 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Shi",
"Ziming",
""
]
] |
2410.08042 | Michel Duprez Mr | Michel Duprez, Vanessa Lleras, Alexei Lozinski, Vincent Vigon and Killian Vuillemot | {\varphi}-FD : A well-conditioned finite difference method inspired by {\varphi}-FEM for general geometries on elliptic PDEs | null | Journal of Scientific Computing, Volume 104, article number 23, (2025) | 10.1007/s10915-025-02914-0 | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a new finite difference method, called {\varphi}-FD, inspired by the {\phi}-FEM approach for solving elliptic partial differential equations (PDEs) on general geometries. The proposed method uses Cartesian grids, ensuring simplicity in implementation. Moreover, contrary to the previous finite difference scheme on non-rectangular domain, the associated matrix is well-conditioned. The use of a level-set function for the geometry description makes this approach relatively flexible. We prove the quasi-optimal convergence rates in several norms and the fact that the matrix is well-conditioned. Additionally, the paper explores the use of multigrid techniques to further accelerate the computation. Finally, numerical experiments in both 2D and 3D validate the performance of the {\varphi}-FD method compared to standard finite element methods and the Shortley-Weller approach. | [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 15:34:20 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 09:20:02 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Duprez",
"Michel",
""
],
[
"Lleras",
"Vanessa",
""
],
[
"Lozinski",
"Alexei",
""
],
[
"Vigon",
"Vincent",
""
],
[
"Vuillemot",
"Killian",
""
]
] |
2410.08459 | Chi Qiu | Chi Qiu, Qingqing Wu, Wen Chen, Meng Hua, Wanming Hao, Mengnan Jian, Fen Hou | Beamforming Design for Intelligent Reffecting Surface Aided Near-Field THz Communications | null | null | null | null | eess.SP cs.IT cs.NI math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Intelligent reflecting surface (IRS) operating in the terahertz (THz) band has recently gained considerable interest due to its high spectrum bandwidth. Due to the exploitation of large scale of IRS, there is a high probability that the transceivers will be situated within the near-field region of the IRS. Thus, the near-field beam split effect poses a major challenge for the design of wideband IRS beamforming, which causes the radiation beam to deviate from its intended location, leading to significant gain losses and limiting the efficient use of available bandwidths. While delay-based IRS has emerged as a potential solution, current beamforming schemes generally assume unbounded range time delays (TDs). In this letter, we first investigate the near-field beam split issue at the IRS. Then, we extend the piece-wise far-field model to the IRS, based on which, a double-layer delta-delay (DLDD) IRS beamforming scheme is proposed. Specifically, we employ an element-grouping strategy and the TD imposed on each sub-surface of IRS is achieved by a series of TD modules. This method significantly reduces the required range of TDs. Numerical results show that the proposed DLDD IRS beamforming scheme can effectively mitigate the near-field beam split and achieve near-optimal performance. | [
{
"version": "v1",
"created": "Fri, 11 Oct 2024 02:24:04 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 08:42:38 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Qiu",
"Chi",
""
],
[
"Wu",
"Qingqing",
""
],
[
"Chen",
"Wen",
""
],
[
"Hua",
"Meng",
""
],
[
"Hao",
"Wanming",
""
],
[
"Jian",
"Mengnan",
""
],
[
"Hou",
"Fen",
""
]
] |
2410.09102 | Tong Wu | Tong Wu, Shujian Zhang, Kaiqiang Song, Silei Xu, Sanqiang Zhao, Ravi Agrawal, Sathish Reddy Indurthi, Chong Xiang, Prateek Mittal, Wenxuan Zhou | Instructional Segment Embedding: Improving LLM Safety with Instruction Hierarchy | Preprint | ICLR 2025 | null | null | cs.LG cs.AI cs.CL cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) are susceptible to security and safety threats, such as prompt injection, prompt extraction, and harmful requests. One major cause of these vulnerabilities is the lack of an instruction hierarchy. Modern LLM architectures treat all inputs equally, failing to distinguish between and prioritize various types of instructions, such as system messages, user prompts, and data. As a result, lower-priority user prompts may override more critical system instructions, including safety protocols. Existing approaches to achieving instruction hierarchy, such as delimiters and instruction-based training, do not address this issue at the architectural level. We introduce the Instructional Segment Embedding (ISE) technique, inspired by BERT, to modern large language models, which embeds instruction priority information directly into the model. This approach enables models to explicitly differentiate and prioritize various instruction types, significantly improving safety against malicious prompts that attempt to override priority rules. Our experiments on the Structured Query and Instruction Hierarchy benchmarks demonstrate an average robust accuracy increase of up to 15.75% and 18.68%, respectively. Furthermore, we observe an improvement in instruction-following capability of up to 4.1% evaluated on AlpacaEval. Overall, our approach offers a promising direction for enhancing the safety and effectiveness of LLM architectures. | [
{
"version": "v1",
"created": "Wed, 09 Oct 2024 12:52:41 GMT"
},
{
"version": "v2",
"created": "Sat, 01 Mar 2025 19:06:35 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Wu",
"Tong",
""
],
[
"Zhang",
"Shujian",
""
],
[
"Song",
"Kaiqiang",
""
],
[
"Xu",
"Silei",
""
],
[
"Zhao",
"Sanqiang",
""
],
[
"Agrawal",
"Ravi",
""
],
[
"Indurthi",
"Sathish Reddy",
""
],
[
"Xiang",
"Chong",
""
],
[
"Mittal",
"Prateek",
""
],
[
"Zhou",
"Wenxuan",
""
]
] |
2410.09186 | Md Elias Hossain | Noorbakhsh Amiri Golilarz, Elias Hossain, Abdoljalil Addeh, Keyan Alexander Rahimi | AI Learning Algorithms: Deep Learning, Hybrid Models, and Large-Scale Model Integration | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | In this paper, we discuss learning algorithms and their importance in different types of applications which includes training to identify important patterns and features in a straightforward, easy-to-understand manner. We will review the main concepts of artificial intelligence (AI), machine learning (ML), deep learning (DL), and hybrid models. Some important subsets of Machine Learning algorithms such as supervised, unsupervised, and reinforcement learning are also discussed in this paper. These techniques can be used for some important tasks like prediction, classification, and segmentation. Convolutional Neural Networks (CNNs) are used for image and video processing and many more applications. We dive into the architecture of CNNs and how to integrate CNNs with ML algorithms to build hybrid models. This paper explores the vulnerability of learning algorithms to noise, leading to misclassification. We further discuss the integration of learning algorithms with Large Language Models (LLM) to generate coherent responses applicable to many domains such as healthcare, marketing, and finance by learning important patterns from large volumes of data. Furthermore, we discuss the next generation of learning algorithms and how we may have an unified Adaptive and Dynamic Network to perform important tasks. Overall, this article provides brief overview of learning algorithms, exploring their current state, applications and future direction. | [
{
"version": "v1",
"created": "Fri, 11 Oct 2024 18:39:25 GMT"
},
{
"version": "v2",
"created": "Thu, 08 May 2025 19:38:13 GMT"
},
{
"version": "v3",
"created": "Mon, 26 May 2025 18:32:17 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Golilarz",
"Noorbakhsh Amiri",
""
],
[
"Hossain",
"Elias",
""
],
[
"Addeh",
"Abdoljalil",
""
],
[
"Rahimi",
"Keyan Alexander",
""
]
] |
2410.09403 | Renqi Chen | Haoyang Su, Renqi Chen, Shixiang Tang, Zhenfei Yin, Xinzhe Zheng, Jinzhe Li, Biqing Qi, Qi Wu, Hui Li, Wanli Ouyang, Philip Torr, Bowen Zhou, Nanqing Dong | Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM-Based Multi-Agent System | Accepted by ACL 2025 Main Conference | null | null | null | cs.AI cs.CL cs.CV cs.LG cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid advancement of scientific progress requires innovative tools that can accelerate knowledge discovery. Although recent AI methods, particularly large language models (LLMs), have shown promise in tasks such as hypothesis generation and experimental design, they fall short of replicating the collaborative nature of real-world scientific practices, where diverse experts work together in teams to tackle complex problems. To address the limitations, we propose an LLM-based multi-agent system, i.e., Virtual Scientists (VirSci), designed to mimic the teamwork inherent in scientific research. VirSci organizes a team of agents to collaboratively generate, evaluate, and refine research ideas. Through comprehensive experiments, we demonstrate that this multi-agent approach outperforms the state-of-the-art method in producing novel scientific ideas. We further investigate the collaboration mechanisms that contribute to its tendency to produce ideas with higher novelty, offering valuable insights to guide future research and illuminating pathways toward building a robust system for autonomous scientific discovery. The code is available at https://github.com/open-sciencelab/Virtual-Scientists. | [
{
"version": "v1",
"created": "Sat, 12 Oct 2024 07:16:22 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Feb 2025 06:07:47 GMT"
},
{
"version": "v3",
"created": "Sun, 25 May 2025 10:48:01 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 03:47:05 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Su",
"Haoyang",
""
],
[
"Chen",
"Renqi",
""
],
[
"Tang",
"Shixiang",
""
],
[
"Yin",
"Zhenfei",
""
],
[
"Zheng",
"Xinzhe",
""
],
[
"Li",
"Jinzhe",
""
],
[
"Qi",
"Biqing",
""
],
[
"Wu",
"Qi",
""
],
[
"Li",
"Hui",
""
],
[
"Ouyang",
"Wanli",
""
],
[
"Torr",
"Philip",
""
],
[
"Zhou",
"Bowen",
""
],
[
"Dong",
"Nanqing",
""
]
] |
2410.09516 | David Zapata Gonzalez | David Zapata Gonzalez, Marcel Meyer and Oliver Mueller | Bridging the Gap Between Data-Driven And Theory-Driven Modelling - Leveraging Causal Machine Learning for Integrative Modelling of Dynamical Systems | 16 pages, 9 figures and 3 tables | https://aisel.aisnet.org/ecis2025/bus_analytics/bus_analytics/2/ | null | null | cs.CE | http://creativecommons.org/licenses/by/4.0/ | Classical machine learning techniques often struggle with overfitting and unreliable predictions when exposed to novel conditions. Introducing causality into the modelling process offers a promising way to mitigate these challenges by enhancing predictive robustness. However, constructing an initial causal graph manually using domain knowledge is time-consuming, particularly in complex time series with numerous variables. To address this, causal discovery algorithms can provide a preliminary causal structure that domain experts can refine. This study investigates causal feature selection with domain knowledge using a data center system as an example. We use simulated time-series data to compare different causal feature selection with traditional machine-learning feature selection methods. Our results show that predictions based on causal features are more robust compared to those derived from traditional methods. These findings underscore the potential of combining causal discovery algorithms with human expertise to improve machine learning applications. | [
{
"version": "v1",
"created": "Sat, 12 Oct 2024 12:33:18 GMT"
},
{
"version": "v2",
"created": "Mon, 02 Dec 2024 16:25:09 GMT"
},
{
"version": "v3",
"created": "Tue, 27 May 2025 11:31:49 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Gonzalez",
"David Zapata",
""
],
[
"Meyer",
"Marcel",
""
],
[
"Mueller",
"Oliver",
""
]
] |
2410.09677 | Hua Wang | Jiaming Hu, Wenbin Li, Zhichao Guo, Hua Wang, Kai Chang | Optical Manifestations of Quantum Geometry in Electron-Phonon Coupling | 7pages, 2 figures | null | null | null | cond-mat.mes-hall cond-mat.mtrl-sci cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum geometry is crucial for understanding intricate condensed matter systems, governing transport phenomena and optical responses. However, traditional studies predominantly consider a static crystal lattice, focusing exclusively on the pure-electronic quantum geometry of the Hilbert space parameterized by electronic wave vectors, thereby overlooking the dynamic effects arising from phonons and their coupling with electrons. In this work, we reveal the intrinsic quantum geometry of the electron-phonon coupling (EPC), which resides in the hybrid Hilbert space parameterized by both the electronic wave vectors and phonon displacements. The EPC quantum metric, EPC Berry curvature and EPC shift vector, as central elements, quantify the EPC-induced velocity, polarization, and anomalous charge-center shift, respectively. We further connect this geometry to phonon-mediated optical responses, particularly in-gap resonances, enabling experimental detection and characterization of EPC quantum geometry. | [
{
"version": "v1",
"created": "Sun, 13 Oct 2024 00:10:03 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 03:17:04 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Hu",
"Jiaming",
""
],
[
"Li",
"Wenbin",
""
],
[
"Guo",
"Zhichao",
""
],
[
"Wang",
"Hua",
""
],
[
"Chang",
"Kai",
""
]
] |
2410.09829 | Rimvydas Rubavicius | Rimvydas Rubavicius, Antonio Valerio Miceli-Barone, Alex Lascarides, Subramanian Ramamoorthy | Conversational Code Generation: a Case Study of Designing a Dialogue System for Generating Driving Scenarios for Testing Autonomous Vehicles | 12 pages, 5 figures, 2 tables | null | null | null | cs.CL cs.IR cs.RO | http://creativecommons.org/licenses/by/4.0/ | Cyber-physical systems like autonomous vehicles are tested in simulation before deployment, using domain-specific programs for scenario specification. To aid the testing of autonomous vehicles in simulation, we design a natural language interface, using an instruction-following large language model, to assist a non-coding domain expert in synthesising the desired scenarios and vehicle behaviours. We show that using it to convert utterances to the symbolic program is feasible, despite the very small training dataset. Human experiments show that dialogue is critical to successful simulation generation, leading to a 4.5 times higher success rate than a generation without engaging in extended conversation. | [
{
"version": "v1",
"created": "Sun, 13 Oct 2024 13:07:31 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 09:29:18 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Rubavicius",
"Rimvydas",
""
],
[
"Miceli-Barone",
"Antonio Valerio",
""
],
[
"Lascarides",
"Alex",
""
],
[
"Ramamoorthy",
"Subramanian",
""
]
] |
2410.10479 | Haochuan Wang | Haochuan Wang, Xiachong Feng, Lei Li, Yu Guo, Zhanyue Qin, Dianbo Sui, Lingpeng Kong | TMGBench: A Systematic Game Benchmark for Evaluating Strategic Reasoning Abilities of LLMs | null | null | null | null | cs.AI cs.GT | http://creativecommons.org/licenses/by/4.0/ | The rapid advancement of large language models has accelerated their application in reasoning, with strategic reasoning drawing increasing attention. To evaluate the strategic reasoning capabilities of LLMs, game theory, with its concise structure, has become the preferred approach for many researchers. However, current research typically focuses on a limited selection of games, resulting in low coverage of game types. Additionally, classic game scenarios carry risks of data leakage, and the benchmarks used often lack extensibility, rendering them inadequate for evaluating state-of-the-art models. To address these challenges, we propose TMGBench, characterized by comprehensive game type coverage, diverse scenarios and flexible game organization. Specifically, we incorporate all 144 game types summarized by the Robinson-Goforth topology of 2x2 games, constructed as classic games in our benchmark; we also synthetize diverse, higher-quality game scenarios for each classic game, which we refer to as story-based games. Lastly, to provide a sustainable evaluation framework adaptable to increasingly powerful LLMs, we treat the aforementioned games as atomic units and organize them into more complex forms through sequential, parallel, and nested structures. We conducted a comprehensive evaluation of mainstream LLMs, covering tests on rational reasoning, reasoning robustness, Theory-of-Mind capabilities, and reasoning in complex game forms. The results revealed LLMs still have flaws in the accuracy and consistency of strategic reasoning processes, and their levels of mastery over Theory-of-Mind also vary. Additionally, SOTA models like o3-mini, Qwen3 and deepseek-reasoner, were also evaluated across the sequential, parallel, and nested game structures while the results highlighted the challenges posed by TMGBench. | [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 13:15:34 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 14:29:54 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Wang",
"Haochuan",
""
],
[
"Feng",
"Xiachong",
""
],
[
"Li",
"Lei",
""
],
[
"Guo",
"Yu",
""
],
[
"Qin",
"Zhanyue",
""
],
[
"Sui",
"Dianbo",
""
],
[
"Kong",
"Lingpeng",
""
]
] |
2410.10502 | Martina Cinquini | Martina Cinquini, Isacco Beretta, Salvatore Ruggieri, Isabel Valera | A Practical Approach to Causal Inference over Time | null | Proceedings of the AAAI Conference on Artificial Intelligence 2025 | 10.1609/aaai.v39i14.33626 | null | cs.AI cs.LG cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | In this paper, we focus on estimating the causal effect of an intervention over time on a dynamical system. To that end, we formally define causal interventions and their effects over time on discrete-time stochastic processes (DSPs). Then, we show under which conditions the equilibrium states of a DSP, both before and after a causal intervention, can be captured by a structural causal model (SCM). With such an equivalence at hand, we provide an explicit mapping from vector autoregressive models (VARs), broadly applied in econometrics, to linear, but potentially cyclic and/or affected by unmeasured confounders, SCMs. The resulting causal VAR framework allows us to perform causal inference over time from observational time series data. Our experiments on synthetic and real-world datasets show that the proposed framework achieves strong performance in terms of observational forecasting while enabling accurate estimation of the causal effect of interventions on dynamical systems. We demonstrate, through a case study, the potential practical questions that can be addressed using the proposed causal VAR framework. | [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 13:45:20 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Cinquini",
"Martina",
""
],
[
"Beretta",
"Isacco",
""
],
[
"Ruggieri",
"Salvatore",
""
],
[
"Valera",
"Isabel",
""
]
] |
2410.10941 | Bartolomeo Trefoloni | Bartolomeo Trefoloni, Roberto Gilli, Elisabeta Lusso, Alessandro Marconi, Giovanni Mazzolari, Emanuele Nardini, Guido Risaliti, Matilde Signorini | The near-infrared SED of blue quasars: what drives the evolution of the dusty torus? | 22 pages, 15 figures | A&A 697, A223 (2025) | 10.1051/0004-6361/202452609 | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | A fundamental ingredient in the unified model of active galactic nuclei (AGN) is the obscuring torus, whose innermost, hottest region dominates the near infrared (NIR) emission. Characterising the change in the torus properties and its interplay with the main AGN emission is key for our understanding of AGN physics, evolution and classification. Its covering factor ($CF$) is largely responsible for the classification of AGN on the basis of the detection of broad emission lines. It is still not clear whether the torus properties evolve over time and how they relate with the accretion parameters of the nucleus. In this work, we aim at investigating the evolution of the NIR properties with the redshift ($z$) and the bolometric luminosity ($L_{\rm bol}$) of the AGN. To this end, we assembled a large dataset of $\sim$36,000 Type 1 AGN between $0.5<z<2.9$ and $45.0<\log(L_{\rm bol} / (\rm erg / s))<48.0$ with UV, optical and near-infrared photometry. We produced average spectral energy distributions (SED) in different bins of the $z-L_{\rm bol}$ parameter space to estimate how the NIR SED evolves according to these parameters. We find that the NIR luminosity decreases for increasing $L_{\rm bol}$ at any redshift. At the same, time the shape of the NIR SED in our sample is consistent with a non-evolution with $z$. As a consequence, all the explored proxies for the $CF$ exhibit significant anti-correlations with $L_{\rm bol}$, but not with $z$. Additionally, the $CF$ also shows a shallower anti-correlation with the Eddington ratio ($\lambda_{\rm Edd}$), yet current systematic uncertainties, as well as the limited dynamical range, do not allow us to precisely constrain the role of the Eddington ratio. Lastly, we derived the covering factor from the ratio between the NIR and optical luminosity and we employed it to set a lower limit for the X-ray obscuration at different redshifts. | [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 18:00:00 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Trefoloni",
"Bartolomeo",
""
],
[
"Gilli",
"Roberto",
""
],
[
"Lusso",
"Elisabeta",
""
],
[
"Marconi",
"Alessandro",
""
],
[
"Mazzolari",
"Giovanni",
""
],
[
"Nardini",
"Emanuele",
""
],
[
"Risaliti",
"Guido",
""
],
[
"Signorini",
"Matilde",
""
]
] |
2410.11668 | Aishwarya Paliwal | A. Paliwal, M. De Petris, A. Ferragamo, R. Adam, P. Ade, H. Ajeddig, P. Andr\'e, E. Artis, H. Aussel, I. Bartalucci, A. Beelen, A. Beno\^it, S. Berta, L. Bing, O. Bourrion, M. Calvo, A. Catalano, F. De Luca, F.-X. D\'esert, S. Doyle, E. F. C. Driessen, G. Ejlali, A. Gomez, J. Goupy, C. Hanser, S. Katsioli, F. K\'eruzor\'e, C. Kramer, B. Ladjelate, G. Lagache, S. Leclercq, J.-F. Lestrade, J. F. Mac\'ias-P\'erez, S. C. Madden, A. Maury, P. Mauskopf, F. Mayet, J.-B. Melin, A. Monfardini, A. Moyer-Anin, M. Mu\~noz-Echeverr\'ia, L. Perotto, G. Pisano, E. Pointecouteau, N. Ponthieu, G.W. Pratt, V. Rev\'eret, A. J. Rigby, A. Ritacco, C. Romero, H. Roussel, F. Ruppin, K. Schuster, A. Sievers, C. Tucker, R. Wicker, R. Zylka | Exploiting the high-resolution NIKA2 data to study the intracluster medium and dynamical state of ACT-CL J0240.0+0116 | 13 pages, 7 figures, submitted to A&A | A&A 698, A2 (2025) | 10.1051/0004-6361/202452597 | null | astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | Having a detailed knowledge of the intracluster medium (ICM) to infer the exact cluster physics such as the cluster dynamical state is crucial for cluster-based cosmological studies. This knowledge limits the accuracy and precision of mass estimation, a key parameter for such studies. In this paper, we conduct an in-depth analysis of cluster ACT-CL J0240.0+0116 using a multi-wavelength approach, with a primary focus on high angular resolution Sunyaev-Zeldovich (SZ) thermal component observations obtained under the NIKA2 Sunyaev-Zeldovich Large Programme (LPSZ). We create composite images using NIKA2, X-ray, and optical galaxy number density maps. The results reveal distinct signs of disturbance within the cluster with the distributions of gas and member galaxies that do not overlap. We also find suggestions of an inflow of matter onto the cluster from the southwestern direction. Ultimately, we classify the cluster as disturbed, using morphological indicators derived from its SZ, X-ray, and optical image. The cluster SZ signal is also contaminated by a strong central point source. We adopt different approaches to handling this contaminant and find the estimates of our pressure and hydrostatic mass profiles robust to the point source mitigation model. The cluster hydrostatic mass is estimated at $4.25^{+0.50}_{-0.45\, } \times 10^{14} \,\mathrm{M}_{\odot}$ for the case where the point source was masked. These values are consistent with the mass estimated using only X-ray data and with those from previous SZ studies of the Atacama cosmology telescope (ACT) survey, with improved precision on the mass estimate. Our findings strongly suggest that ACT-CL J0240.0+0116 is a disturbed cluster system, and the detailed observations and derived values serve as a compelling case study for the capabilities of the LPSZ in mapping the cluster ICM with high precision. | [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 14:54:53 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Paliwal",
"A.",
""
],
[
"De Petris",
"M.",
""
],
[
"Ferragamo",
"A.",
""
],
[
"Adam",
"R.",
""
],
[
"Ade",
"P.",
""
],
[
"Ajeddig",
"H.",
""
],
[
"André",
"P.",
""
],
[
"Artis",
"E.",
""
],
[
"Aussel",
"H.",
""
],
[
"Bartalucci",
"I.",
""
],
[
"Beelen",
"A.",
""
],
[
"Benoît",
"A.",
""
],
[
"Berta",
"S.",
""
],
[
"Bing",
"L.",
""
],
[
"Bourrion",
"O.",
""
],
[
"Calvo",
"M.",
""
],
[
"Catalano",
"A.",
""
],
[
"De Luca",
"F.",
""
],
[
"Désert",
"F. -X.",
""
],
[
"Doyle",
"S.",
""
],
[
"Driessen",
"E. F. C.",
""
],
[
"Ejlali",
"G.",
""
],
[
"Gomez",
"A.",
""
],
[
"Goupy",
"J.",
""
],
[
"Hanser",
"C.",
""
],
[
"Katsioli",
"S.",
""
],
[
"Kéruzoré",
"F.",
""
],
[
"Kramer",
"C.",
""
],
[
"Ladjelate",
"B.",
""
],
[
"Lagache",
"G.",
""
],
[
"Leclercq",
"S.",
""
],
[
"Lestrade",
"J. -F.",
""
],
[
"Macías-Pérez",
"J. F.",
""
],
[
"Madden",
"S. C.",
""
],
[
"Maury",
"A.",
""
],
[
"Mauskopf",
"P.",
""
],
[
"Mayet",
"F.",
""
],
[
"Melin",
"J. -B.",
""
],
[
"Monfardini",
"A.",
""
],
[
"Moyer-Anin",
"A.",
""
],
[
"Muñoz-Echeverría",
"M.",
""
],
[
"Perotto",
"L.",
""
],
[
"Pisano",
"G.",
""
],
[
"Pointecouteau",
"E.",
""
],
[
"Ponthieu",
"N.",
""
],
[
"Pratt",
"G. W.",
""
],
[
"Revéret",
"V.",
""
],
[
"Rigby",
"A. J.",
""
],
[
"Ritacco",
"A.",
""
],
[
"Romero",
"C.",
""
],
[
"Roussel",
"H.",
""
],
[
"Ruppin",
"F.",
""
],
[
"Schuster",
"K.",
""
],
[
"Sievers",
"A.",
""
],
[
"Tucker",
"C.",
""
],
[
"Wicker",
"R.",
""
],
[
"Zylka",
"R.",
""
]
] |
2410.12128 | Zhengyang Zhou | Zhengyang Zhou, Yunrui Li, Pengyu Hong, Hao Xu | Multimodal Fusion with Relational Learning for Molecular Property Prediction | null | null | null | null | cs.CE | http://creativecommons.org/licenses/by/4.0/ | Graph based molecular representation learning is essential for accurately predicting molecular properties in drug discovery and materials science; however, it faces significant challenges due to the intricate relationships among molecules and the limited chemical knowledge utilized during training. While contrastive learning is often employed to handle molecular relationships, its reliance on binary metrics is insufficient for capturing the complexity of these interactions. Multimodal fusion has gained attention for property reasoning, but previous work has explored only a limited range of modalities, and the optimal stages for fusing different modalities in molecular property tasks remain underexplored. In this paper, we introduce MMFRL (Multimodal Fusion with Relational Learning for Molecular Property Prediction), a novel framework designed to overcome these limitations. Our method enhances embedding initialization through multimodal pretraining using relational learning. We also conduct a systematic investigation into the impact of modality fusion at different stages such as early, intermediate, and late, highlighting their advantages and shortcomings. Extensive experiments on MoleculeNet benchmarks demonstrate that MMFRL significantly outperforms existing methods. Furthermore, MMFRL enables task-specific optimizations. Additionally, the explainability of MMFRL provides valuable chemical insights, emphasizing its potential to enhance real-world drug discovery applications. | [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 00:09:43 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 21:49:15 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Zhou",
"Zhengyang",
""
],
[
"Li",
"Yunrui",
""
],
[
"Hong",
"Pengyu",
""
],
[
"Xu",
"Hao",
""
]
] |
2410.12255 | Ziqi Ji | Ziqi Ji, Penghao Duan, Gang Du | Enhancing generalizability of machine learning general effective-viscosity turbulence model via tensor basis normalization | null | null | null | null | physics.flu-dyn | http://creativecommons.org/licenses/by/4.0/ | With the rapid advancement of machine learning techniques, the development and study of machine learning turbulence models have become increasingly prevalent. As a critical component of turbulence modeling, the constitutive relationship between the Reynolds stress tensor and the mean flow quantities, modeled using machine learning methods, faces a pressing challenge: the lack of generalizability. To address this issue, we propose a novel tensor basis normalization technique to improve machine learning turbulence models, grounded in the general effective-viscosity hypothesis. In this study, we utilize direct numerical simulation (DNS) results of periodic hill flows as training data to develop a symbolic regression-based turbulence model based on the general effective-viscosity hypothesis. Furthermore, we construct a systematic validation dataset to evaluate the generalizability of our symbolic regression-based turbulence model. This validation set includes periodic hills with different aspect ratios from the training dataset, zero pressure gradient flat plate flows, three-dimensional incompressible flows over a NACA0012 airfoil, and transonic axial compressor rotor flows. These validation cases exhibit significant flow characteristics and geometrical variations, progressively increasing their differences from the training dataset. Such a diverse validation set is a robust benchmark to assess the generalizability of the proposed turbulence model. Finally, we demonstrate that our symbolic regression-based turbulence model performs effectively across validation cases, encompassing various separation features, geometries, and Reynolds numbers. | [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 05:44:07 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Oct 2024 02:41:33 GMT"
},
{
"version": "v3",
"created": "Mon, 24 Mar 2025 12:53:57 GMT"
},
{
"version": "v4",
"created": "Tue, 25 Mar 2025 13:17:51 GMT"
},
{
"version": "v5",
"created": "Tue, 27 May 2025 05:56:31 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Ji",
"Ziqi",
""
],
[
"Duan",
"Penghao",
""
],
[
"Du",
"Gang",
""
]
] |
2410.12458 | Minghao Wu | Minghao Wu, Thuy-Trang Vu, Lizhen Qu, Gholamreza Haffari | The Best of Both Worlds: Bridging Quality and Diversity in Data Selection with Bipartite Graph | Accepted by ICML 2025 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The performance of large language models (LLMs) is strongly influenced by the quality and diversity of data used during supervised fine-tuning (SFT). However, current data selection methods often prioritize one aspect over the other, resulting in suboptimal training outcomes. To address this, we formulate data selection as a set cover problem and present GraphFilter, a novel approach that balances both quality and diversity in data selection. GraphFilter models the dataset as a bipartite graph connecting sentences to their constituent n-grams, then employs a priority function that combines quality and diversity metrics multiplicatively. GraphFilter iteratively selects sentences with the highest priority, removes covered n-grams from the bipartite graph, and recomputes priorities to reflect the changing data landscape. We validate GraphFilter using three model backbones across six widely-used benchmarks, demonstrating that it outperforms nine existing baselines in both model performance and computational efficiency. Further analysis shows that our design choices lead to more effective subset selection, underscores the value of instruction diversity, and provides insights into how quality and diversity interact with different subset sizes. | [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 11:16:34 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 19:41:16 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Wu",
"Minghao",
""
],
[
"Vu",
"Thuy-Trang",
""
],
[
"Qu",
"Lizhen",
""
],
[
"Haffari",
"Gholamreza",
""
]
] |
2410.12526 | Lechao Cheng | Mingce Guo, Jingxuan He, Shengeng Tang, Zhangye Wang, Lechao Cheng | Shaping a Stabilized Video by Mitigating Unintended Changes for Concept-Augmented Video Editing | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Text-driven video editing utilizing generative diffusion models has garnered significant attention due to their potential applications. However, existing approaches are constrained by the limited word embeddings provided in pre-training, which hinders nuanced editing targeting open concepts with specific attributes. Directly altering the keywords in target prompts often results in unintended disruptions to the attention mechanisms. To achieve more flexible editing easily, this work proposes an improved concept-augmented video editing approach that generates diverse and stable target videos flexibly by devising abstract conceptual pairs. Specifically, the framework involves concept-augmented textual inversion and a dual prior supervision mechanism. The former enables plug-and-play guidance of stable diffusion for video editing, effectively capturing target attributes for more stylized results. The dual prior supervision mechanism significantly enhances video stability and fidelity. Comprehensive evaluations demonstrate that our approach generates more stable and lifelike videos, outperforming state-of-the-art methods. | [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 13:03:15 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 04:43:05 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Guo",
"Mingce",
""
],
[
"He",
"Jingxuan",
""
],
[
"Tang",
"Shengeng",
""
],
[
"Wang",
"Zhangye",
""
],
[
"Cheng",
"Lechao",
""
]
] |
2410.12671 | Yanyun Wang | Yanyun Wang, Li Liu, Zi Liang, Yi R. (May) Fung, Qingqing Ye, Haibo Hu | New Paradigm of Adversarial Training: Releasing Accuracy-Robustness Trade-Off via Dummy Class | Preprint. Under review | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial Training (AT) is one of the most effective methods to enhance the robustness of Deep Neural Networks (DNNs). However, existing AT methods suffer from an inherent accuracy-robustness trade-off. Previous works have studied this issue under the current AT paradigm, but still face over 10% accuracy reduction without significant robustness improvement over simple baselines such as PGD-AT. This inherent trade-off raises a question: Whether the current AT paradigm, which assumes to learn corresponding benign and adversarial samples as the same class, inappropriately mixes clean and robust objectives that may be essentially inconsistent. In fact, our empirical results show that up to 40% of CIFAR-10 adversarial samples always fail to satisfy such an assumption across various AT methods and robust models, explicitly indicating the room for improvement of the current AT paradigm. To relax from this overstrict assumption and the tension between clean and robust learning, in this work, we propose a new AT paradigm by introducing an additional dummy class for each original class, aiming to accommodate hard adversarial samples with shifted distribution after perturbation. The robustness w.r.t. these adversarial samples can be achieved by runtime recovery from the predicted dummy classes to the corresponding original ones, without conflicting with the clean objective on accuracy of benign samples. Finally, based on our new paradigm, we propose a novel DUmmy Classes-based Adversarial Training (DUCAT) method that concurrently improves accuracy and robustness in a plug-and-play manner only relevant to logits, loss, and a proposed two-hot soft label-based supervised signal. Our method outperforms state-of-the-art (SOTA) benchmarks, effectively releasing the current trade-off. The code is available at https://github.com/FlaAI/DUCAT. | [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 15:36:10 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 02:55:37 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Wang",
"Yanyun",
"",
"May"
],
[
"Liu",
"Li",
"",
"May"
],
[
"Liang",
"Zi",
"",
"May"
],
[
"R.",
"Yi",
"",
"May"
],
[
"Fung",
"",
""
],
[
"Ye",
"Qingqing",
""
],
[
"Hu",
"Haibo",
""
]
] |
2410.12754 | Jacquelyn Ho | Jacquelyn Ho, Yue-Hui Lu, Tai Xiang, Cosimo C. Rusconi, Stuart J. Masson, Ana Asenjo-Garcia, Zhenjie Yan, Dan M. Stamper-Kurn | Optomechanical self-organization in a mesoscopic atom array | 22 pages, 9 figures | Nature Physics (2025) | 10.1038/s41567-025-02916-7 | null | quant-ph cond-mat.quant-gas physics.atom-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Increasing the number of particles in a system often leads to qualitative changes in its properties, such as breaking of symmetries and the appearance of phase transitions. This renders a macroscopic system fundamentally different from its individual microscopic constituents. Lying between these extremes, mesoscopic systems exhibit microscopic fluctuations that influence behavior on longer length scales, leading to critical phenomena and dynamics. Therefore, tracing the properties of well-controlled mesoscopic systems can help bridge the gap between an exact description of few-body microscopic systems and the emergent description of many-body systems. Here, we explore mesoscopic signatures of an optomechanical self-organization phase transition using arrays of cold atoms inside an optical cavity. By precisely engineering atom-cavity interactions, we reveal how critical behavior depends on atom number, identify characteristic dynamical behaviors in the self-organized regime, and observe a finite optomechanical susceptibility at the critical point. These findings advance our understanding of particle-number- and time-resolved properties of phase transitions in mesoscopic systems. | [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 17:18:15 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 18:32:30 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Ho",
"Jacquelyn",
""
],
[
"Lu",
"Yue-Hui",
""
],
[
"Xiang",
"Tai",
""
],
[
"Rusconi",
"Cosimo C.",
""
],
[
"Masson",
"Stuart J.",
""
],
[
"Asenjo-Garcia",
"Ana",
""
],
[
"Yan",
"Zhenjie",
""
],
[
"Stamper-Kurn",
"Dan M.",
""
]
] |
2410.13206 | Shintaro Ozaki | Shintaro Ozaki, Kazuki Hayashi, Miyu Oba, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe | BQA: Body Language Question Answering Dataset for Video Large Language Models | Accepted to ACL2025 (Main) | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A large part of human communication relies on nonverbal cues such as facial expressions, eye contact, and body language. Unlike language or sign language, such nonverbal communication lacks formal rules, requiring complex reasoning based on commonsense understanding. Enabling current Video Large Language Models (VideoLLMs) to accurately interpret body language is a crucial challenge, as human unconscious actions can easily cause the model to misinterpret their intent. To address this, we propose a dataset, BQA, a body language question answering dataset, to validate whether the model can correctly interpret emotions from short clips of body language comprising 26 emotion labels of videos of body language. We evaluated various VideoLLMs on BQA and revealed that understanding body language is challenging, and our analyses of the wrong answers by VideoLLMs show that certain VideoLLMs made significantly biased answers depending on the age group and ethnicity of the individuals in the video. The dataset is available. | [
{
"version": "v1",
"created": "Thu, 17 Oct 2024 04:19:26 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 05:21:28 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Ozaki",
"Shintaro",
""
],
[
"Hayashi",
"Kazuki",
""
],
[
"Oba",
"Miyu",
""
],
[
"Sakai",
"Yusuke",
""
],
[
"Kamigaito",
"Hidetaka",
""
],
[
"Watanabe",
"Taro",
""
]
] |
2410.13776 | Georgios Chochlakis | Georgios Chochlakis, Alexandros Potamianos, Kristina Lerman, Shrikanth Narayanan | Aggregation Artifacts in Subjective Tasks Collapse Large Language Models' Posteriors | 16 pages, 12 figures, 3 tables | Proceedings of the 2025 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 5513-5528, Albuquerque, New Mexico, April 2025 | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | In-context Learning (ICL) has become the primary method for performing natural language tasks with Large Language Models (LLMs). The knowledge acquired during pre-training is crucial for this few-shot capability, providing the model with task priors. However, recent studies have shown that ICL predominantly relies on retrieving task priors rather than "learning" to perform tasks. This limitation is particularly evident in complex subjective domains such as emotion and morality, where priors significantly influence posterior predictions. In this work, we examine whether this is the result of the aggregation used in corresponding datasets, where trying to combine low-agreement, disparate annotations might lead to annotation artifacts that create detrimental noise in the prompt. Moreover, we evaluate the posterior bias towards certain annotators by grounding our study in appropriate, quantitative measures of LLM priors. Our results indicate that aggregation is a confounding factor in the modeling of subjective tasks, and advocate focusing on modeling individuals instead. However, aggregation does not explain the entire gap between ICL and the state of the art, meaning other factors in such tasks also account for the observed phenomena. Finally, by rigorously studying annotator-level labels, we find that it is possible for minority annotators to both better align with LLMs and have their perspectives further amplified. | [
{
"version": "v1",
"created": "Thu, 17 Oct 2024 17:16:00 GMT"
},
{
"version": "v2",
"created": "Fri, 18 Oct 2024 18:17:41 GMT"
},
{
"version": "v3",
"created": "Tue, 04 Feb 2025 19:59:05 GMT"
},
{
"version": "v4",
"created": "Mon, 26 May 2025 20:34:30 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Chochlakis",
"Georgios",
""
],
[
"Potamianos",
"Alexandros",
""
],
[
"Lerman",
"Kristina",
""
],
[
"Narayanan",
"Shrikanth",
""
]
] |
2410.13778 | Luca Frittoli | Michelangelo Olmo Nogara Notarianni, Filippo Leveni, Diego Stucchi, Luca Frittoli, Giacomo Boracchi | Change Detection in Multivariate data streams: Online Analysis with Kernel-QuantTree | AALTD workshop at ECML 2024 (https://ecml-aaltd.github.io/aaltd2024/) | null | 10.1007/978-3-031-77066-1_7 | null | cs.LG cs.AI stat.ML | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We present Kernel-QuantTree Exponentially Weighted Moving Average (KQT-EWMA), a non-parametric change-detection algorithm that combines the Kernel-QuantTree (KQT) histogram and the EWMA statistic to monitor multivariate data streams online. The resulting monitoring scheme is very flexible, since histograms can be used to model any stationary distribution, and practical, since the distribution of test statistics does not depend on the distribution of datastream in stationary conditions (non-parametric monitoring). KQT-EWMA enables controlling false alarms by operating at a pre-determined Average Run Length ($ARL_0$), which measures the expected number of stationary samples to be monitored before triggering a false alarm. The latter peculiarity is in contrast with most non-parametric change-detection tests, which rarely can control the $ARL_0$ a priori. Our experiments on synthetic and real-world datasets demonstrate that KQT-EWMA can control $ARL_0$ while achieving detection delays comparable to or lower than state-of-the-art methods designed to work in the same conditions. | [
{
"version": "v1",
"created": "Thu, 17 Oct 2024 17:17:38 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Notarianni",
"Michelangelo Olmo Nogara",
""
],
[
"Leveni",
"Filippo",
""
],
[
"Stucchi",
"Diego",
""
],
[
"Frittoli",
"Luca",
""
],
[
"Boracchi",
"Giacomo",
""
]
] |
2410.13789 | Matthias G\"obel | Tanja Kirchner, Matthias G\"obel, Hans-Werner Hammer | Neutron-neutron distribution of the triton from pionless EFT | 18 pages, 7 figures. Further explanations, discussions and appendices with figures added. Results unchanged. Accepted for publication in Phys. Rev. C | Phys. Rev. C 111, 044002 (2025) | 10.1103/PhysRevC.111.044002 | null | nucl-th nucl-ex | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We compute the neutron-neutron relative-energy distribution of the triton following the hard knockout of the proton in pionless effective field theory. This distribution can be used to study universality as well as to obtain information on the neutron-neutron interaction. Especially, one can infer the scattering length from fitting theory predictions for the shape of the distribution to experimental data. To obtain the distribution for the triton, we first solve the ground-state three-body problem using momentum-space Faddeev equations. Next, we include the neutron-neutron final-state interaction by applying the corresponding M{\o}ller operator to the ground state. We present leading-order (LO) and next-to-leading order (NLO) pionless effective field theory results with quantified uncertainties. At NLO, we include the effective ranges semi-perturbatively. We conclude, that pionless EFT works reliably as expected and that the neutron-neutron distribution of the triton shows a significant sensitivity to the scattering length. | [
{
"version": "v1",
"created": "Thu, 17 Oct 2024 17:29:23 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 14:52:07 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Kirchner",
"Tanja",
""
],
[
"Göbel",
"Matthias",
""
],
[
"Hammer",
"Hans-Werner",
""
]
] |
2410.14116 | Yichen Zhou | Yichen Zhou, Yanglei Song, Serdar Y\"uksel | Robustness to Model Approximation, Model Learning From Data, and Sample Complexity in Wasserstein Regular MDPs | 36 pages | null | null | null | eess.SY cs.SY math.OC | http://creativecommons.org/licenses/by/4.0/ | The paper studies the robustness properties of discrete-time stochastic optimal control under Wasserstein model approximation for both discounted cost and average cost criteria. Specifically, we study the performance loss when applying an optimal policy designed for an approximate model to the true dynamics compared with the optimal cost for the true model under the sup-norm-induced metric, and relate it to the Wasserstein-1 distance between the approximate and true transition kernels. A primary motivation of this analysis is empirical model learning, as well as empirical noise distribution learning, where Wasserstein convergence holds under mild conditions but stronger convergence criteria, such as total variation, may not. We discuss applications of the results to the disturbance estimation problem, where sample complexity bounds are given, and also to a general empirical model learning approach, obtained under either Markov or i.i.d. learning settings. | [
{
"version": "v1",
"created": "Fri, 18 Oct 2024 02:01:47 GMT"
},
{
"version": "v2",
"created": "Thu, 07 Nov 2024 00:20:24 GMT"
},
{
"version": "v3",
"created": "Thu, 14 Nov 2024 05:00:46 GMT"
},
{
"version": "v4",
"created": "Tue, 27 May 2025 03:05:09 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Zhou",
"Yichen",
""
],
[
"Song",
"Yanglei",
""
],
[
"Yüksel",
"Serdar",
""
]
] |
2410.14319 | Samuel Ponc\'e | Samuel Ponc\'e and Jae-Mo Lihm and Cheol-Hwan Park | Verification and Validation of zero-point electron-phonon renormalization of the bandgap, mass enhancement, and spectral functions | 20 pages and 10 figures | npj Computational Materials 11, 117 (2025) | 10.1038/s41524-025-01587-5 | null | cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | Verification and validation of methods and first-principles software are at the core of computational solid-state physics but are too rarely addressed. We compare four first-principles codes: Abinit, Quantum ESPRESSO, EPW, ZG, and three methods: (i) the Allen-Heine-Cardona theory using density functional perturbation theory (DFPT), (ii) the Allen-Heine-Cardona theory using Wannier function perturbation theory (WFPT), and (iii) an adiabatic non-perturbative frozen-phonon method. For these cases, we compute the real and imaginary parts of the electron-phonon self-energy in diamond and BAs, including dipoles and quadrupoles when interpolating. We find excellent agreement between software that implements the same formalism as well as good agreement between the DFPT and WFPT methods. Importantly, we find that the Deybe-Waller term is momentum dependent which impacts the mass enhancement, yielding approximate results when using the Luttinger approximations. Finally, we compare the electron-phonon spectral functions between Abinit and EPW and find excellent agreement even away from the band edges. | [
{
"version": "v1",
"created": "Fri, 18 Oct 2024 09:25:36 GMT"
},
{
"version": "v2",
"created": "Mon, 31 Mar 2025 09:57:04 GMT"
}
] | 2025-05-28T00:00:00 | [
[
"Poncé",
"Samuel",
""
],
[
"Lihm",
"Jae-Mo",
""
],
[
"Park",
"Cheol-Hwan",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.