id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 6
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-05-16 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2503.15183 | Mary Carstairs | Anatolii I. Visser-Zadvornyi (Kapteyn Institute), Mary E. Carstairs
(Durham-ICC), Kyle A. Oman (Durham-ICC), Marc A. W. Verheijen (Kapteyn
Institute) | Star formation and stellar & AGN feedback in the absence of accretion,
not gas stripping, set the quenching timescale in satellite galaxies | 16 pages, 9 figures, MNRAS accepted | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Observational measurements hint at a peak in the quenching timescale of
satellite galaxies in groups and clusters as a function of their stellar masses
at $M_{\star} \approx 10^{9.5} \mathrm{M}_{\odot}$; less and more massive
satellite galaxies quench faster. We investigate the origin of these trends
using the EAGLE simulation in which they are qualitatively reproduced for
satellites with $10^{9}<M_{\star}/\mathrm{M}_\odot<10^{11}$ around hosts of
$10^{13}<M_\mathrm{200c}/\mathrm{M}_\odot<10^{14.6}$. We select gas particles
of simulated galaxies at the time that they become satellites and track their
evolution. Interpreting these data yields insights into the prevailing
mechanism that leads to the depletion of the interstellar medium (ISM) and the
cessation of star formation. We find that for satellites across our entire
range in stellar mass the quenching timescale is to leading order set by the
depletion of the ISM by star formation and stellar & AGN feedback in the
absence of sustained accretion of fresh gas. The turnover in the quenching
timescale as a function of stellar mass is a direct consequence of the maximum
in the star formation efficiency (or equivalently the minimum in the total --
stellar plus AGN -- feedback efficiency) at the same stellar mass. We can
discern the direct stripping of the ISM by ram pressure and/or tides in the
simulations; these mechanisms modulate the quenching timescale but do not drive
its overall scaling with satellite stellar mass. Our findings argue against a
scenario in which the turnover in the quenching timescale is a consequence of
the competing influences of gas stripping and 'starvation'.
| [
{
"version": "v1",
"created": "Wed, 19 Mar 2025 13:10:04 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 12:33:14 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Visser-Zadvornyi",
"Anatolii I.",
"",
"Kapteyn Institute"
],
[
"Carstairs",
"Mary E.",
"",
"Durham-ICC"
],
[
"Oman",
"Kyle A.",
"",
"Durham-ICC"
],
[
"Verheijen",
"Marc A. W.",
"",
"Kapteyn\n Institute"
]
] |
2503.16371 | Minori Narita | Minori Narita, Ryo Kuroiwa, J. Christopher Beck | Reinforcement Learning-based Heuristics to Guide Domain-Independent
Dynamic Programming | 24 pages, 4 figures, to be published in CPAIOR 2025
(https://sites.google.com/view/cpaior2025) | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Domain-Independent Dynamic Programming (DIDP) is a state-space search
paradigm based on dynamic programming for combinatorial optimization. In its
current implementation, DIDP guides the search using user-defined dual bounds.
Reinforcement learning (RL) is increasingly being applied to combinatorial
optimization problems and shares several key structures with DP, being
represented by the Bellman equation and state-based transition systems. We
propose using reinforcement learning to obtain a heuristic function to guide
the search in DIDP. We develop two RL-based guidance approaches: value-based
guidance using Deep Q-Networks and policy-based guidance using Proximal Policy
Optimization. Our experiments indicate that RL-based guidance significantly
outperforms standard DIDP and problem-specific greedy heuristics with the same
number of node expansions. Further, despite longer node evaluation times, RL
guidance achieves better run-time performance than standard DIDP on three of
four benchmark domains.
| [
{
"version": "v1",
"created": "Thu, 20 Mar 2025 17:33:08 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 19:08:33 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Narita",
"Minori",
""
],
[
"Kuroiwa",
"Ryo",
""
],
[
"Beck",
"J. Christopher",
""
]
] |
2503.16623 | Zhongtang Luo | Zhongtang Luo | ICLR Points: How Many ICLR Publications Is One Paper in Each Area? | null | null | null | null | cs.DL cs.CY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Scientific publications significantly impact academic-related decisions in
computer science, where top-tier conferences are particularly influential.
However, efforts required to produce a publication differ drastically across
various subfields. While existing citation-based studies compare venues within
areas, cross-area comparisons remain challenging due to differing publication
volumes and citation practices.
To address this gap, we introduce the concept of ICLR points, defined as the
average effort required to produce one publication at top-tier machine learning
conferences such as ICLR, ICML, and NeurIPS. Leveraging comprehensive
publication data from DBLP (2019--2023) and faculty information from
CSRankings, we quantitatively measure and compare the average publication
effort across 27 computer science sub-areas. Our analysis reveals significant
differences in average publication effort, validating anecdotal perceptions:
systems conferences generally require more effort per publication than AI
conferences.
We further demonstrate the utility of the ICLR points metric by evaluating
publication records of universities, current faculties and recent faculty
candidates. Our findings highlight how using this metric enables more
meaningful cross-area comparisons in academic evaluation processes. Lastly, we
discuss the metric's limitations and caution against its misuse, emphasizing
the necessity of holistic assessment criteria beyond publication metrics alone.
| [
{
"version": "v1",
"created": "Thu, 20 Mar 2025 18:23:35 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Mar 2025 00:17:56 GMT"
},
{
"version": "v3",
"created": "Wed, 26 Mar 2025 13:57:21 GMT"
},
{
"version": "v4",
"created": "Wed, 14 May 2025 16:27:13 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Luo",
"Zhongtang",
""
]
] |
2503.17591 | Pedro Linck Maciel | Pedro Linck Maciel and Nadja K. Bernardes | Quantum Computation based on Open Quantum Walks | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Open Quantum Walks (OQW) are a type of quantum walk governed by the system's
interaction with its environment. We explore the time evolution and the limit
behavior of the OQW framework for Quantum Computation and show how we can
represent random unitary quantum channels, such as the dephasing and
depolarizing channels, in this model. We also develop a simulation protocol
with circuit representation for this model, which is heavily inspired by the
fact that graphs represent OQW and are, thereby, local (meaning that the state
in a particular node interacts only with its neighborhood). We obtain
asymptotic advantages in the system's dimension, circuit depth, and CNOT count
compared to other simulation methods.
| [
{
"version": "v1",
"created": "Sat, 22 Mar 2025 00:26:50 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 00:05:14 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Maciel",
"Pedro Linck",
""
],
[
"Bernardes",
"Nadja K.",
""
]
] |
2503.17599 | Wenhao Jiang | Zheqing Li, Yiying Yang, Jiping Lang, Wenhao Jiang, Yuhang Zhao,
Shuang Li, Dingqian Wang, Zhu Lin, Xuanna Li, Yuze Tang, Jiexian Qiu, Xiaolin
Lu, Hongji Yu, Shuang Chen, Yuhua Bi, Xiaofei Zeng, Yixian Chen, Junrong
Chen, Lin Yao | Evaluating Clinical Competencies of Large Language Models with a General
Practice Benchmark | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) have demonstrated considerable potential in
general practice. However, existing benchmarks and evaluation frameworks
primarily depend on exam-style or simplified question-answer formats, lacking a
competency-based structure aligned with the real-world clinical
responsibilities encountered in general practice. Consequently, the extent to
which LLMs can reliably fulfill the duties of general practitioners (GPs)
remains uncertain. In this work, we propose a novel evaluation framework to
assess the capability of LLMs to function as GPs. Based on this framework, we
introduce a general practice benchmark (GPBench), whose data are meticulously
annotated by domain experts in accordance with routine clinical practice
standards. We evaluate ten state-of-the-art LLMs and analyze their
competencies. Our findings indicate that current LLMs are not yet ready for
deployment in such settings without human oversight, and further optimization
specifically tailored to the daily responsibilities of GPs is essential.
| [
{
"version": "v1",
"created": "Sat, 22 Mar 2025 01:02:44 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 10:25:11 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Li",
"Zheqing",
""
],
[
"Yang",
"Yiying",
""
],
[
"Lang",
"Jiping",
""
],
[
"Jiang",
"Wenhao",
""
],
[
"Zhao",
"Yuhang",
""
],
[
"Li",
"Shuang",
""
],
[
"Wang",
"Dingqian",
""
],
[
"Lin",
"Zhu",
""
],
[
"Li",
"Xuanna",
""
],
[
"Tang",
"Yuze",
""
],
[
"Qiu",
"Jiexian",
""
],
[
"Lu",
"Xiaolin",
""
],
[
"Yu",
"Hongji",
""
],
[
"Chen",
"Shuang",
""
],
[
"Bi",
"Yuhua",
""
],
[
"Zeng",
"Xiaofei",
""
],
[
"Chen",
"Yixian",
""
],
[
"Chen",
"Junrong",
""
],
[
"Yao",
"Lin",
""
]
] |
2503.18599 | Minsu Kim | Minsu Kim, Seongmin Hong, RyeoWook Ko, Soongyu Choi, Hunjong Lee,
Junsoo Kim, Joo-Young Kim, Jongse Park | Oaken: Fast and Efficient LLM Serving with Online-Offline Hybrid KV
Cache Quantization | 16 pages, 14 figures, and 4 tables | null | 10.1145/3695053.3731019 | null | cs.AR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Modern Large Language Model serving system batches multiple requests to
achieve high throughput, while batching attention operations is challenging,
rendering memory bandwidth a critical bottleneck. The community relies on
high-end GPUs with multiple high-bandwidth memory channels. Unfortunately,
HBM's high bandwidth often comes at the expense of limited memory capacity,
which reduces core utilization and increases costs. Recent advancements
enabling longer contexts for LLMs have substantially increased the key-value
cache size, further intensifying the pressures on memory capacity. The
literature has explored KV cache quantization techniques, which commonly use
low bitwidth for most values, selectively using higher bitwidth for outlier
values. While this approach helps achieve high accuracy and low bitwidth
simultaneously, it comes with the limitation that cost for online outlier
detection is excessively high, negating the advantages. We propose Oaken, an
acceleration solution that achieves high accuracy and high performance
simultaneously through co-designing algorithm and hardware. To effectively find
a sweet spot in the accuracy-performance trade-off space of KV cache
quantization, Oaken employs an online-offline hybrid approach, setting outlier
thresholds offline, which are then used to determine the quantization scale
online. To translate the proposed algorithmic technique into tangible
performance gains, Oaken also comes with custom quantization engines and memory
management units that can be integrated with any LLM accelerators. We built an
Oaken accelerator on top of an LLM accelerator, LPU, and conducted a
comprehensive evaluation. Our experiments show that for a batch size of 256,
Oaken achieves up to 1.58x throughput improvement over NVIDIA A100 GPU,
incurring a minimal accuracy loss of only 0.54\% on average, compared to
state-of-the-art KV cache quantization techniques.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 11:56:50 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 04:22:24 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Kim",
"Minsu",
""
],
[
"Hong",
"Seongmin",
""
],
[
"Ko",
"RyeoWook",
""
],
[
"Choi",
"Soongyu",
""
],
[
"Lee",
"Hunjong",
""
],
[
"Kim",
"Junsoo",
""
],
[
"Kim",
"Joo-Young",
""
],
[
"Park",
"Jongse",
""
]
] |
2503.18938 | Shenyuan Gao | Shenyuan Gao, Siyuan Zhou, Yilun Du, Jun Zhang, Chuang Gan | AdaWorld: Learning Adaptable World Models with Latent Actions | ICML 2025. Project page: https://adaptable-world-model.github.io/,
code: https://github.com/Little-Podi/AdaWorld, model:
https://huggingface.co/Little-Podi/AdaWorld | null | null | null | cs.AI cs.CV cs.LG cs.RO | http://creativecommons.org/licenses/by/4.0/ | World models aim to learn action-controlled future prediction and have proven
essential for the development of intelligent agents. However, most existing
world models rely heavily on substantial action-labeled data and costly
training, making it challenging to adapt to novel environments with
heterogeneous actions through limited interactions. This limitation can hinder
their applicability across broader domains. To overcome this limitation, we
propose AdaWorld, an innovative world model learning approach that enables
efficient adaptation. The key idea is to incorporate action information during
the pretraining of world models. This is achieved by extracting latent actions
from videos in a self-supervised manner, capturing the most critical
transitions between frames. We then develop an autoregressive world model that
conditions on these latent actions. This learning paradigm enables highly
adaptable world models, facilitating efficient transfer and learning of new
actions even with limited interactions and finetuning. Our comprehensive
experiments across multiple environments demonstrate that AdaWorld achieves
superior performance in both simulation quality and visual planning.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 17:58:15 GMT"
},
{
"version": "v2",
"created": "Mon, 12 May 2025 12:09:07 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 10:26:17 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Gao",
"Shenyuan",
""
],
[
"Zhou",
"Siyuan",
""
],
[
"Du",
"Yilun",
""
],
[
"Zhang",
"Jun",
""
],
[
"Gan",
"Chuang",
""
]
] |
2503.19024 | Kohki Uno | Kohki Uno, Keiichi Maeda, Takashi Nagao, Giorgos Leloudas, Panos
Charalampopoulos, Seppo Mattila, Kentaro Aoki, Kenta Taguchi, Miho Kawabata,
Javier Moldon, Miguel P\'erez-Torres, Miika Pursiainen and Thomas Reynolds | Spectropolarimetry of A Nuclear Transient AT2023clx: Revealing The
Geometrical Alignment between The Transient Outflow and The Nuclear Dusty
Region | 15 pages, 5 figures and 1 table. Accepted for Publication in the ApJ
Letters | null | null | null | astro-ph.HE astro-ph.GA | http://creativecommons.org/publicdomain/zero/1.0/ | AT2023clx, which occurred in NGC3799 with a Low-Ionization Nuclear
Emission-Line Region (LINER), is one of the most nearby nuclear transients
classified as a tidal disruption event (TDE). We present three-epoch
spectropolarimetric follow-up observations of AT2023clx. We detected two
polarization components; one is a constant polarization of $\sim 1\%$
originated from an aspherical outflow associated with the transient, while the
other is a blue-excess polarization toward $\sim 2\%$ originated from a nuclear
dusty environment via light echoes. The polarization angle flipped by 90
degrees between the two epochs, indicating that the outflow direction was
perpendicular to the dust plane. Furthermore, the polarized flux might suggest
that the nuclear dust favors relatively large grains, potentially offering
constraints on its physical properties. Such polarization features -- the blue
excess and the 90-degree flip -- have never been observed in previous TDE
polarization samples, highlighting unique mechanisms behind AT2023clx. We
propose possible scenarios: the disruption of a star formed within or captured
by a nuclear dusty cloud. Given the LINER nature of NGC3799, the dusty region
may possibly be linked to a torus or disk associated with a weak Active
Galactic Nucleus (AGN). Furthermore, as a more speculative scenario, the event
might have been triggered by AGN-like activity, potentially linked to
changing-look AGNs or ambiguous nuclear transients. These findings highlight
the power of time-series spectropolarimetry of TDEs, not only in probing the
origins of nuclear transients, but also in investigating the physical
properties of nuclear dust.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 18:00:02 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 11:07:32 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Uno",
"Kohki",
""
],
[
"Maeda",
"Keiichi",
""
],
[
"Nagao",
"Takashi",
""
],
[
"Leloudas",
"Giorgos",
""
],
[
"Charalampopoulos",
"Panos",
""
],
[
"Mattila",
"Seppo",
""
],
[
"Aoki",
"Kentaro",
""
],
[
"Taguchi",
"Kenta",
""
],
[
"Kawabata",
"Miho",
""
],
[
"Moldon",
"Javier",
""
],
[
"Pérez-Torres",
"Miguel",
""
],
[
"Pursiainen",
"Miika",
""
],
[
"Reynolds",
"Thomas",
""
]
] |
2503.19863 | Tomas Vyhlidal PhD | Can Kutlu Y\"uksel, Tom\'a\v{s} Vyhl\'idal, Jaroslav Bu\v{s}ek, Martin
Hrom\v{c}\'ik and Silviu-Iulian Niculescu | A Spectrum-based Filter Design for Periodic Control of Systems with Time
Delay | 25 pages, 10 figures, accepted 15 January 2025, Available online 22
January 2025, Version of Record 28 January 2025 | Journal of Sound and Vibration, Volume 604, 2025, 118959 | 10.1016/j.jsv.2025.118959 | null | eess.SY cs.SY math.OC | http://creativecommons.org/licenses/by/4.0/ | A fully analytical controller design is proposed to tackle a periodic control
problem for stable linear systems with an input delay. Applying the internal
model control scheme, the controller design reduces to designing a filter,
which is done through the placement of poles and zeros. The zeros are placed to
compensate for the harmonics and to achieve the desired degree of properness
for the filter. For placing the poles, a quasi-optimal procedure is proposed
utilizing the standard LQR method. Given the high-dimensionality of the filter
due to targeting a large number of harmonics, the design, as well as controller
implementation, is performed over a state-space representation. A thorough
experimental case study is included to demonstrate both the practical
feasibility and effectiveness of the proposed control design. The experimental
validation is performed on a physical system, the goal of which is to reject
periodic vibrations acting on a mass-spring-damper setup where the sensor and
the actuator are non-collocated.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 17:31:35 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 13:34:35 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Yüksel",
"Can Kutlu",
""
],
[
"Vyhlídal",
"Tomáš",
""
],
[
"Bušek",
"Jaroslav",
""
],
[
"Hromčík",
"Martin",
""
],
[
"Niculescu",
"Silviu-Iulian",
""
]
] |
2503.19997 | Luyao Zhang | Luyao Zhang, Fabo Feng, Yicheng Rui, Guang-Yao Xiao, Wenting Wang | A Cartesian catalog of 30 million Gaia sources based on second-order and
Monte Carlo error propagation | null | Research in Astronomy and Astrophysics, 25, 055010 (2025) | 10.1088/1674-4527/adc6da | 055010 | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Accurate measurements of stellar positions and velocities are crucial for
studying galactic and stellar dynamics. We aim to create a Cartesian catalog
from {\it Gaia} DR3 to serve as a high-precision database for further research
using stellar coordinates and velocities. To avoid the negative parallax
values, we select 31,129,169 sources in {\it Gaia} DR3 with radial velocity,
where the fractional parallax error is less than 20\% ($0 <
\sigma_\varpi/\varpi < 0.2$). To select the most accurate and efficient method
of propagating mean and covariance, we use the Monte Carlo results with $10^7$
samples (MC7) as the benchmark, and compare the precision of linear,
second-order, and Monte Carlo error propagation methods. By assessing the
accuracy of propagated mean and covariance, we observe that second-order error
propagation exhibits mean deviations of at most 0.5\% compared to MC7, with
variance deviations of up to 10\%. Overall, this outperforms linear
transformation. Though Monte Carlo method with $10^4$ samples (MC4) is an order
of magnitude slower than second-order error propagation, its covariances
propagation accuracy reaches 1\% when $\sigma_\varpi/\varpi$ is below 15\%.
Consequently, we employ second-order error propagation to convert the mean
astrometry and radial velocity into Cartesian coordinates and velocities in
both equatorial and galactic systems for 30 million {\it Gaia} sources, and
apply MC4 for covariance propagation. The Cartesian catalog and source code are
provided for future applications in high-precision stellar and galactic
dynamics.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 18:29:38 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Zhang",
"Luyao",
""
],
[
"Feng",
"Fabo",
""
],
[
"Rui",
"Yicheng",
""
],
[
"Xiao",
"Guang-Yao",
""
],
[
"Wang",
"Wenting",
""
]
] |
2503.20899 | Chris Bard | Chris Bard, John Dorelli | High-Performance Computational Magnetohydrodynamics with Python | 21 pages, 8 figures; submitted to Computer Physics Communications;
AGATE open-source repository is located at:
https://git.smce.nasa.gov/marble/agate-open-source | null | null | null | physics.plasm-ph physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | We present the AGATE simulation code, a Python-based framework developed
primarily for solving the magnetohydrodynamics (MHD) equations while
maintaining adaptability to other equation sets. The code employs a modular,
object-oriented architecture that separates interface specifications from
numerical implementations, allowing users to customize numerical methods and
physics models. Built on a Godunov-type finite-volume scheme, AGATE currently
supports the ideal, Hall, and Chew-Goldberger-Low (CGL) MHD equations, with
multiple acceleration options ranging from Numpy to GPU-enabled computation via
NVIDIA CUDA. Performance testing demonstrates that our GPU implementations
achieve 40-60x speedups over CPU versions. Comprehensive validation through
established benchmarks confirms accurate reproduction of both linear and
nonlinear phenomena across different MHD regimes. This combination of
modularity, performance, and extensibility makes AGATE suitable for multiple
applications: from rapid prototyping to production simulations, and from
numerical algorithm development to physics education.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 18:16:17 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 16:46:32 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Bard",
"Chris",
""
],
[
"Dorelli",
"John",
""
]
] |
2503.21117 | Alfredo Raya Prof. | MPD Collaboration | MPD physics performance studies in Bi+Bi collisions at
$\sqrt{s_{NN}}=9.2$ GeV | 45 pages, 50 figures | null | null | null | nucl-ex hep-ph nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Multi-Purpose Detector (MPD) is one of the three experiments of the
Nuclotron Ion Collider-fAcility (NICA) complex, which is currently under
construction at the Joint Institute for Nuclear Research in Dubna. With
collisions of heavy ions in the collider mode, the MPD will cover the energy
range 4-11 GeV to scan the high baryon-density region of the QCD phase diagram.
With expected statistics of 50-100 million events collected during the first
run, MPD will be able to study a number of observables, including measurements
of light hadrons and hypernuclei production, particle flow, correlations and
fluctuations, have a first look at dielectron production, and modification of
vector-meson properties in dense matter. In this paper, we present selected
results of the physics feasibility studies for the MPD experiment in
Bismuth-Bismuth collisions at an energy of 9.2 GeV, the system considered as
one of the first available at the NICA collider.
| [
{
"version": "v1",
"created": "Thu, 27 Mar 2025 03:13:28 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 00:39:31 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"MPD Collaboration",
"",
""
]
] |
2503.21402 | Matilda Backholm | Shohreh Askari, Guillem Saldo Rubio, Anagha Datar, Hedi Harjunp\"a\"a,
Susanna C. Fagerholm, and Matilda Backholm | Soft matter mechanics of immune cell aggregates | null | null | null | null | physics.bio-ph | http://creativecommons.org/licenses/by/4.0/ | T-cells are a crucial subset of white blood cells that play a central role in
the immune system. When T-cells bind antigens, it leads to cell activation and
the induction of an immune response. If T-cells are activated by antigens in
vivo or artificially in vitro, they form multicellular aggregates. The
mechanical properties of such clusters provide valuable information on
different T-cell activation pathways. Furthermore, the aggregate mechanics
capture how T-cells are affected by mechanical forces and interact within
larger conglomerates, such as lymph nodes and tumours. However, an
understanding of collective T-cell adhesion and mechanics following cell
activation is currently lacking. Probing the mechanics of fragile and
microscopically small living samples is experimentally challenging. Here, the
micropipette force sensor technique was used to stretch T-cell aggregates and
directly measure their Young's modulus and ultimate tensile strength. A
mechanistic model was developed to correlate how the stiffness of the mesoscale
multicellular aggregate emerges from the mechanical response of the individual
microscopic cells within the cluster. We show how the aggregate elasticity is
affected by different activators and relate this to different activation
pathways in the cells. Our soft matter mechanics study of multicellular T-cell
aggregates contributes to our understanding of the biology behind immune cell
activation.
| [
{
"version": "v1",
"created": "Thu, 27 Mar 2025 11:47:58 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 07:57:09 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Askari",
"Shohreh",
""
],
[
"Rubio",
"Guillem Saldo",
""
],
[
"Datar",
"Anagha",
""
],
[
"Harjunpää",
"Hedi",
""
],
[
"Fagerholm",
"Susanna C.",
""
],
[
"Backholm",
"Matilda",
""
]
] |
2503.21620 | Zhengxi Lu | Zhengxi Lu, Yuxiang Chai, Yaxuan Guo, Xi Yin, Liang Liu, Hao Wang, Han
Xiao, Shuai Ren, Guanjing Xiong, Hongsheng Li | UI-R1: Enhancing Efficient Action Prediction of GUI Agents by
Reinforcement Learning | Updated UI-R1-E-3B | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | The recent DeepSeek-R1 has showcased the emergence of reasoning capabilities
in LLMs through reinforcement learning (RL) with rule-based rewards. Despite
its success in language models, its application in multi-modal domains,
particularly in graphic user interface (GUI) agent tasks, remains
under-explored. To address this issue, we propose UI-R1, the first framework to
explore how rule-based RL can enhance the reasoning capabilities of multimodal
large language models (MLLMs) for GUI action prediction tasks. Specifically,
UI-R1 introduces a novel rule-based action reward, enabling model optimization
via policy-based algorithms such as Group Relative Policy Optimization (GRPO).
For efficient training, we curate a small yet high-quality dataset of 136
challenging tasks, encompassing five common action types on mobile devices.
Experimental results demonstrate that our proposed UI-R1-3B achieves
significant improvements over the base model (i.e. Qwen2.5-VL-3B) on both
in-domain (ID) and out-of-domain (OOD) tasks, with average accuracy gains of
22.1% on ScreenSpot, 6.0% on ScreenSpot-Pro, and 12.7% on ANDROIDCONTROL.
Furthermore, UI-R1-3B delivers competitive performance compared to larger
models (e.g., OS-Atlas-7B) trained via supervised fine-tuning (SFT) on 76K
samples. We additionally develop an optimized version, UI-R1-E-3B, which
significantly improves both grounding efficiency and accuracy. These results
underscore the potential of rule-based reinforcement learning to advance GUI
understanding and control, paving the way for future research in this domain.
Code website: https://github.com/lll6gg/UI-R1.
| [
{
"version": "v1",
"created": "Thu, 27 Mar 2025 15:39:30 GMT"
},
{
"version": "v2",
"created": "Sun, 30 Mar 2025 13:05:16 GMT"
},
{
"version": "v3",
"created": "Wed, 16 Apr 2025 07:06:40 GMT"
},
{
"version": "v4",
"created": "Wed, 14 May 2025 11:56:07 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Lu",
"Zhengxi",
""
],
[
"Chai",
"Yuxiang",
""
],
[
"Guo",
"Yaxuan",
""
],
[
"Yin",
"Xi",
""
],
[
"Liu",
"Liang",
""
],
[
"Wang",
"Hao",
""
],
[
"Xiao",
"Han",
""
],
[
"Ren",
"Shuai",
""
],
[
"Xiong",
"Guanjing",
""
],
[
"Li",
"Hongsheng",
""
]
] |
2503.21696 | Wenqi Zhang | Wenqi Zhang, Mengna Wang, Gangao Liu, Xu Huixin, Yiwei Jiang,
Yongliang Shen, Guiyang Hou, Zhe Zheng, Hang Zhang, Xin Li, Weiming Lu, Peng
Li, Yueting Zhuang | Embodied-Reasoner: Synergizing Visual Search, Reasoning, and Action for
Embodied Interactive Tasks | Code: https://github.com/zwq2018/embodied_reasoner Dataset:
https://huggingface.co/datasets/zwq2018/embodied_reasoner | null | null | null | cs.CL cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in deep thinking models have demonstrated remarkable
reasoning capabilities on mathematical and coding tasks. However, their
effectiveness in embodied domains which require continuous interaction with
environments through image action interleaved trajectories remains largely
-unexplored. We present Embodied Reasoner, a model that extends o1 style
reasoning to interactive embodied search tasks. Unlike mathematical reasoning
that relies primarily on logical deduction, embodied scenarios demand spatial
understanding, temporal reasoning, and ongoing self-reflection based on
interaction history. To address these challenges, we synthesize 9.3k coherent
Observation-Thought-Action trajectories containing 64k interactive images and
90k diverse thinking processes (analysis, spatial reasoning, reflection,
planning, and verification). We develop a three-stage training pipeline that
progressively enhances the model's capabilities through imitation learning,
self-exploration via rejection sampling, and self-correction through reflection
tuning. The evaluation shows that our model significantly outperforms those
advanced visual reasoning models, e.g., it exceeds OpenAI o1, o3-mini, and
Claude-3.7 by +9\%, 24\%, and +13\%. Analysis reveals our model exhibits fewer
repeated searches and logical inconsistencies, with particular advantages in
complex long-horizon tasks. Real-world environments also show our superiority
while exhibiting fewer repeated searches and logical inconsistency cases.
| [
{
"version": "v1",
"created": "Thu, 27 Mar 2025 17:00:51 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 17:48:02 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Zhang",
"Wenqi",
""
],
[
"Wang",
"Mengna",
""
],
[
"Liu",
"Gangao",
""
],
[
"Huixin",
"Xu",
""
],
[
"Jiang",
"Yiwei",
""
],
[
"Shen",
"Yongliang",
""
],
[
"Hou",
"Guiyang",
""
],
[
"Zheng",
"Zhe",
""
],
[
"Zhang",
"Hang",
""
],
[
"Li",
"Xin",
""
],
[
"Lu",
"Weiming",
""
],
[
"Li",
"Peng",
""
],
[
"Zhuang",
"Yueting",
""
]
] |
2503.21813 | Zhangcheng Qiang | Zhangcheng Qiang, Kerry Taylor, Weiqing Wang, Jing Jiang | OAEI-LLM-T: A TBox Benchmark Dataset for Understanding Large Language
Model Hallucinations in Ontology Matching | 14 pages, 4 figures, 4 tables, 2 prompt templates | null | null | null | cs.CL cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hallucinations are often inevitable in downstream tasks using large language
models (LLMs). To tackle the substantial challenge of addressing hallucinations
for LLM-based ontology matching (OM) systems, we introduce a new benchmark
dataset OAEI-LLM-T. The dataset evolves from seven TBox datasets in the
Ontology Alignment Evaluation Initiative (OAEI), capturing hallucinations of
ten different LLMs performing OM tasks. These OM-specific hallucinations are
organised into two primary categories and six sub-categories. We showcase the
usefulness of the dataset in constructing an LLM leaderboard for OM tasks and
for fine-tuning LLMs used in OM tasks.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 18:20:04 GMT"
},
{
"version": "v2",
"created": "Wed, 7 May 2025 15:02:02 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 16:57:13 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Qiang",
"Zhangcheng",
""
],
[
"Taylor",
"Kerry",
""
],
[
"Wang",
"Weiqing",
""
],
[
"Jiang",
"Jing",
""
]
] |
2503.22453 | Si-Wei Liu | Si-Wei Liu, Ju-Jun Xie | Studying the $K\Lambda$ strong interaction with femtoscopic correlation
functions and the $N^*(1535)$ | null | null | null | null | hep-ph nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the $K \Lambda$ strong interaction dynamics around the energy
region of $N^*(1535)$ resonance through femtoscopic correlation functions
within the Koonin-Pratt formalism, where coupled-channel effects of $\pi N$,
$K\Lambda$, $K \Sigma$ and $\eta N$ channels are taken into account, by means
of which the $N^*(1535)$ resonance can be dynamically generated in the chiral
unitary approach. Two-body scattering amplitudes are calculated using the
chiral unitary approach, and the corresponding wave functions and correlation
functions are obtained. Our analysis shows that the $\pi N$ channel becomes
important to the $K^+\Lambda$ correlation function through coupled channel
effects, even if its mass threshold is far from the energy region of the
$N^*(1535)$. Furthermore, with the picture that the $N^*(1535)$ is a
dynamically generated state, we can get a good reproduction of the experimental
data on the $K^+\Lambda$ correlation function. This work establishes
femtoscopic correlation functions as a sensitive probe for coupled-channel
dynamics and validates the Koonin-Pratt framework in multi-channel hadronic
systems, offering critical insights into some exotic hadron states and strong
interaction mechanisms.
| [
{
"version": "v1",
"created": "Fri, 28 Mar 2025 14:06:02 GMT"
},
{
"version": "v2",
"created": "Thu, 3 Apr 2025 03:05:07 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 02:11:44 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Liu",
"Si-Wei",
""
],
[
"Xie",
"Ju-Jun",
""
]
] |
2503.24293 | Hayley Ross | Hayley Ross, Kathryn Davidson, Najoung Kim | Is analogy enough to draw novel adjective-noun inferences? | 9 pages (17 pages with appendix). Accepted to SCiL 2025 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Recent work (Ross et al., 2025, 2024) has argued that the ability of humans
and LLMs respectively to generalize to novel adjective-noun combinations shows
that they each have access to a compositional mechanism to determine the
phrase's meaning and derive inferences. We study whether these inferences can
instead be derived by analogy to known inferences, without need for
composition. We investigate this by (1) building a model of analogical
reasoning using similarity over lexical items, and (2) asking human
participants to reason by analogy. While we find that this strategy works well
for a large proportion of the dataset of Ross et al. (2025), there are novel
combinations for which both humans and LLMs derive convergent inferences but
which are not well handled by analogy. We thus conclude that the mechanism
humans and LLMs use to generalize in these cases cannot be fully reduced to
analogy, and likely involves composition.
| [
{
"version": "v1",
"created": "Mon, 31 Mar 2025 16:41:16 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 13:59:02 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Ross",
"Hayley",
""
],
[
"Davidson",
"Kathryn",
""
],
[
"Kim",
"Najoung",
""
]
] |
2504.00015 | Junde Wu | Edward B. Fel'dman, Alexander I. Zenchuk, Wentao Qi, Junde Wu | Controlled measurement, Hermitian conjugation and normalization in
matrix-manipulation algorithms | null | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | In this paper, we solve three important problems that are revealed, in
particular, in matrix-manipulation algorithms. The principal novelty is
introducing the concept of controlled measurement that solves the problem of
small access probability to the desired state of ancilla and possesses several
remarkable properties. We also introduce separate encoding of the real and
imaginary parts of a complex matrix that allows to include the Hermitian
conjugation into the list of matrix manipulations. Finally, we weaken the
constraints on the absolute values of matrix elements unavoidably imposed by
the normalization condition for a pure quantum state. The controlled
measurement together with both other extensions are implemented into the matrix
multiplication algorithm. The appropriate circuits are presented.
| [
{
"version": "v1",
"created": "Thu, 27 Mar 2025 08:49:59 GMT"
},
{
"version": "v2",
"created": "Wed, 2 Apr 2025 07:47:26 GMT"
},
{
"version": "v3",
"created": "Tue, 13 May 2025 23:41:03 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Fel'dman",
"Edward B.",
""
],
[
"Zenchuk",
"Alexander I.",
""
],
[
"Qi",
"Wentao",
""
],
[
"Wu",
"Junde",
""
]
] |
2504.00064 | Naresh Dadhich | Naresh Dadhich | Relativistic Virial Theorem, Limiting Compactness, and the end state of
gravitational collapse | 13 pages, latex double space, Essay written for GRF2025. Received
Honorable Mention in GRF 2025 | null | null | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is noteworthy that limiting compactness of a static bounded configuration
is characterized by a general principle: \textit{one, by equipartition of mass
between inside and outside, and the other by vanishing of energy inside.} The
former implies gravitational energy being half of mass leading to limiting
compactness $M/R = 4/9$ of Buchdahl star while for the latter, the two are
equal giving $M/R = 1/2$ of black hole with horizon. \emph{This is the
relativistic Virial theorem respectively for massive and massless particles.}
It is remarkable that it prescribes that there can exist only two equilibrium
states which also define limiting compactness of the object. Consequently, it
leads to a profound prediction that the ultimate endproduct of gravitational
collapse could only be one of the two, Buchdahl star or black hole.
| [
{
"version": "v1",
"created": "Mon, 31 Mar 2025 15:13:16 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 12:36:55 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Dadhich",
"Naresh",
""
]
] |
2504.01068 | Mateusz Bronikowski | M. Bronikowski, T. Petrushevska, J. D. R. Pierel, A. Acebron, D.
Donevski, B. Apostolova, N. Blagorodnova, T. Jankovi\v{c} | Cluster-lensed supernova yields from the Vera C. Rubin Observatory and
Nancy Grace Roman Space Telescope | 20 pages, 11 figures | A&A, 697, A146 (2025) | 10.1051/0004-6361/202451457 | null | astro-ph.GA astro-ph.CO astro-ph.HE astro-ph.SR | http://creativecommons.org/licenses/by/4.0/ | Through gravitational lensing, galaxy clusters can magnify supernovae (SNe)
and create multiple images of the same SN. This enables measurements of
cosmological parameters, which will be increasingly important in light of
upcoming telescopic surveys. We study the prospects of detecting strongly
lensed SNe in cluster fields with the Nancy Grace Roman Space Telescope
(Roman)'s High Latitude Time Domain Survey (HLTDS) and the Vera C. Rubin
Observatory's Legacy Survey of Space and Time (LSST). We employed two
approaches: one focusing on known multiply imaged galaxies behind clusters,
along with the SN rates specific to those galaxies, and another based on the
expected number of lensed SNe exploding in a given volume behind a galaxy
cluster. We collected all the clusters in the literature that feature a
well-constrained lens model and multiply imaged galaxies behind clusters with
high-quality data for the lensed galaxies. This allowed us to determine the SN
rate for each galaxy. We provide predictions for 46 clusters visible to the
Vera C. Rubin Observatory, as well as for 9 observable by Roman's HLTDS,
depending on whether the clusters fall within the survey's observing field. We
predict that the number of multiply imaged SNe discovered by LSST in its first
three years is $3.95 \pm 0.89$ from the first approach or $4.94 \pm 1.02$ from
the second. For the HLTDS, the expected number of multiply imaged SNe ranges
from $0.38 \pm 0.15$ to $5.2 \pm 2.2$, depending on the specific cluster
observed, however, the fields to be targeted remain a matter of discussion. We
conclude that LSST offers great prospects for detecting multiply imaged SNe.
Our predictions are effectively lower limits, as we only considered the most
massive and well-studied clusters. We provide a recommendation for HLTDS
observing field selection, namely: either MACS J0553.4-3342 or Abell 1758a
should be observed by the survey.
| [
{
"version": "v1",
"created": "Tue, 1 Apr 2025 18:00:01 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 09:36:35 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Bronikowski",
"M.",
""
],
[
"Petrushevska",
"T.",
""
],
[
"Pierel",
"J. D. R.",
""
],
[
"Acebron",
"A.",
""
],
[
"Donevski",
"D.",
""
],
[
"Apostolova",
"B.",
""
],
[
"Blagorodnova",
"N.",
""
],
[
"Jankovič",
"T.",
""
]
] |
2504.01335 | Atsushi Ito | Atsushi Ito | A remark on some punctual Quot schemes on smooth projective curves | 9 pages, v2: minor changes such as typo corrections, corrected the
grant information | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For a locally free sheaf $\mathcal{E}$ on a smooth projective curve, we can
define the punctual Quot scheme which parametrizes torsion quotients of
$\mathcal{E}$ of length $n$ supported at a fixed point. It is known that the
punctual Quot scheme is a normal projective variety with canonical Gorenstein
singularities. In this note, we show that the punctual Quot scheme is a
$\mathbb{Q}$-factorial Fano variety of Picard number one.
| [
{
"version": "v1",
"created": "Wed, 2 Apr 2025 03:43:27 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 03:08:14 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Ito",
"Atsushi",
""
]
] |
2504.01590 | Andrea Marina Lucero Manzano | Andrea M. Lucero Manzano, Esteban D. Cantero, Emanuel A. Mart\'inez,
F. Y. Bruno, Esteban A. S\'anchez, Oscar Grizzi | KTaO3(001) Preparation Methods in Vacuum: Effects on Surface
Stoichiometry, Crystallography, and in-gap States | 30 pages, 8 figures | null | null | null | cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | KTaO3 single crystals with different orientations are used as substrates for
the epitaxial growth of thin films and/or as hosts for two-dimensional electron
gases. Due to the polar nature of the KTaO3(001) surface, one can expect
difficulties and challenges to arise in its preparation. Maintaining good
insulating characteristics without adding undesirable in-gap electronic states,
obtaining good crystalline order up to the top surface layer, a sufficiently
flat surface, and complete cleanliness of the surface (without water, C or OH
contaminants), are in general difficult conditions to accomplish
simultaneously. Cleaving in vacuum is likely the best option for obtaining a
clean surface. However, since KTaO3 is cubic and lacks a well-defined cleavage
plane, this method is notsuitable for sample growth or reproducible device
fabrication. Here, we systematically evaluate the effect of typical preparation
methods applied on the surfaces of KTaO3(001) single crystals. In particular,
we used annealing in vacuum at different temperatures, light sputtering with
Ar+ ions at low energy (500 eV) followed by annealing, heavy Ar+ ion
bombardment and annealing, and grazing Ar+ ion bombardment under continuous
azimuthal rotation combined with both annealing in vacuum and in O2 atmosphere.
Possible side effects after each treatment are evaluated by a combination of
techniques, including low-energy ion scattering at forward angles, Auger
electron spectroscopy, low-energy electron energy loss, X-ray photoelectron
spectroscopy, low-energy electron diffraction, and time of flightsecondary ion
mass spectrometry. Advantages and shortcomings of each preparation method are
discussed in detail.
| [
{
"version": "v1",
"created": "Wed, 2 Apr 2025 10:52:57 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 13:28:53 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Manzano",
"Andrea M. Lucero",
""
],
[
"Cantero",
"Esteban D.",
""
],
[
"Martínez",
"Emanuel A.",
""
],
[
"Bruno",
"F. Y.",
""
],
[
"Sánchez",
"Esteban A.",
""
],
[
"Grizzi",
"Oscar",
""
]
] |
2504.01781 | Kartik Waghmare | Kartik Waghmare and Johanna Ziegel | Proper scoring rules for estimation and forecast evaluation | null | null | null | null | math.ST stat.ML stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Proper scoring rules have been a subject of growing interest in recent years,
not only as tools for evaluation of probabilistic forecasts but also as methods
for estimating probability distributions. In this article, we review the
mathematical foundations of proper scoring rules including general
characterization results and important families of scoring rules. We discuss
their role in statistics and machine learning for estimation and forecast
evaluation. Furthermore, we comment on interesting developments of their usage
in applications.
| [
{
"version": "v1",
"created": "Wed, 2 Apr 2025 14:46:14 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 22:12:47 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Waghmare",
"Kartik",
""
],
[
"Ziegel",
"Johanna",
""
]
] |
2504.02140 | \"Ozge Keskin | Ozge Keskin, Samuel K. Lander, and Ersin Gogus | Investigating the Recursive Short X-ray Burst Behavior of Magnetars
Through Crustal Interactions | Accepted for publication in ApJ | null | 10.3847/1538-4357/adc915 | null | astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Energetic bursts from strongly magnetized neutron stars, known as magnetars,
are typically detected in clusters. Once an active episode begins, anywhere
from a few to thousands of hard X-ray bursts can occur over durations ranging
from days to months. The temporal clustering of these recurrent bursts during
an active episode suggests an underlying mechanism that triggers multiple
bursts in rapid succession. These burst clusters are likely crucial for
understanding the processes driving magnetar activity. In this study, we
investigate the repetitive short X-ray burst behavior of magnetars through
crustal interactions, employing the cellular automaton model for the magnetar
crust proposed by Lander (2023). Our simulations, based on physically motivated
criteria, successfully reproduce burst clustering. Additionally, the durations
and energetics of active episodes in our simulations agree well with
observational data. We discuss the potential physical mechanisms underlying
burst clusters observed in numerous magnetars, as well as the reactivations of
an individual magnetar.
| [
{
"version": "v1",
"created": "Wed, 2 Apr 2025 21:33:39 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Keskin",
"Ozge",
""
],
[
"Lander",
"Samuel K.",
""
],
[
"Gogus",
"Ersin",
""
]
] |
2504.02538 | Bruce T. Draine | B. T. Draine, Karin Sandstrom, Daniel A. Dale, J.-D. T. Smith, Ryan
Chown, Grant P. Donnelly, Sara E. Duval, Cory M. Whitcomb, Angela Adamo, L.
Armus, Danielle A. Berg, Torsten B\"oker, Alberto D. Bolatto, Martha L.
Boyer, Daniela Calzetti, B. G. Elmegreen, Brandt A. L. Gaches, Karl D.
Gordon, L. K. Hunt, R. C. Kennicutt, Ralf S. Klessen, Thomas S.-Y. Lai, Adam
K. Leroy, Sean T. Linden, Alex Pedrini, Noah S. J. Rogers, Julia C.
Roman-Duval, Eva Schinnerer, Evan B. Skillman, Fabian Walter, Tony D.
Weinbeck, Benjamin F. Williams | Detection of Deuterated Hydrocarbon Nanoparticles in the Whirlpool
Galaxy, M51 | Revised. corresponding to published paper | Astrophys. J. Letters, 948:L42 (2025) | 10.3847/2041-8213/adc991 | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deuteration of hydrocarbon material, including polycyclic aromatic
hydrocarbons (PAHs), has been proposed to account for the low gas-phase
abundances of D in the interstellar medium. JWST spectra of four star-forming
regions in M51 show an emission feature, with central wavelength
$\sim$4.647$\mu$m and FWHM 0.0265$\mu$m, corresponding to the C-D stretching
mode in aliphatic hydrocarbons. The emitting aliphatic material is estimated to
have (D/H)$_{\rm aliph}\approx 0.17\pm0.02$ -- a factor $\sim$$10^4$ enrichment
relative to the overall interstellar medium (ISM). On $\sim$$50\,$pc scales,
deuteration levels toward four H$\,$II regions in M51 are 2-3 times higher than
in the Orion Bar photodissociation region (PDR), with implications for the
processes responsible for the formation and evolution of hydrocarbon
nanoparticles, including PAHs. The deuteration of the aliphatic material is
found to anticorrelate with helium ionization in the associated H$\,$II,
suggesting that harsh FUV radiation may act to lower the deuteration of
aliphatics in PDRs near massive stars. No evidence is found for deuteration of
aromatic material, with (D/H)$_{\rm arom} \lesssim 0.016$: deuteration of the
aliphatic material exceeds that of the aromatic material by at least a factor
10. The observed levels of deuteration may account for the depletion of D
observed in the Galactic interstellar medium. If so, the $4.65\mu$m feature may
be detectable in absorption.
| [
{
"version": "v1",
"created": "Thu, 3 Apr 2025 12:39:47 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 18:30:29 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Draine",
"B. T.",
""
],
[
"Sandstrom",
"Karin",
""
],
[
"Dale",
"Daniel A.",
""
],
[
"Smith",
"J. -D. T.",
""
],
[
"Chown",
"Ryan",
""
],
[
"Donnelly",
"Grant P.",
""
],
[
"Duval",
"Sara E.",
""
],
[
"Whitcomb",
"Cory M.",
""
],
[
"Adamo",
"Angela",
""
],
[
"Armus",
"L.",
""
],
[
"Berg",
"Danielle A.",
""
],
[
"Böker",
"Torsten",
""
],
[
"Bolatto",
"Alberto D.",
""
],
[
"Boyer",
"Martha L.",
""
],
[
"Calzetti",
"Daniela",
""
],
[
"Elmegreen",
"B. G.",
""
],
[
"Gaches",
"Brandt A. L.",
""
],
[
"Gordon",
"Karl D.",
""
],
[
"Hunt",
"L. K.",
""
],
[
"Kennicutt",
"R. C.",
""
],
[
"Klessen",
"Ralf S.",
""
],
[
"Lai",
"Thomas S. -Y.",
""
],
[
"Leroy",
"Adam K.",
""
],
[
"Linden",
"Sean T.",
""
],
[
"Pedrini",
"Alex",
""
],
[
"Rogers",
"Noah S. J.",
""
],
[
"Roman-Duval",
"Julia C.",
""
],
[
"Schinnerer",
"Eva",
""
],
[
"Skillman",
"Evan B.",
""
],
[
"Walter",
"Fabian",
""
],
[
"Weinbeck",
"Tony D.",
""
],
[
"Williams",
"Benjamin F.",
""
]
] |
2504.02692 | Yuhang Li | Yuhang Li, Ruokai Yin, Donghyun Lee, Shiting Xiao, Priyadarshini Panda | GPTAQ: Efficient Finetuning-Free Quantization for Asymmetric Calibration | ICML 2025 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce GPTAQ, a novel finetuning-free quantization method for
compressing large-scale transformer architectures. Unlike the previous GPTQ
method, which independently calibrates each layer, we always match the
quantized layer's output to the exact output in the full-precision model,
resulting in a scheme that we call asymmetric calibration. Such a scheme can
effectively reduce the quantization error accumulated in previous layers. We
analyze this problem using optimal brain compression to derive a close-formed
solution. The new solution explicitly minimizes the quantization error as well
as the accumulated asymmetry error. Furthermore, we utilize various techniques
to parallelize the solution calculation, including channel parallelization,
neuron decomposition, and Cholesky reformulation for matrix fusion. As a
result, GPTAQ is easy to implement, simply using 20 more lines of code than
GPTQ but improving its performance under low-bit quantization. Remarkably, on a
single GPU, we quantize a 405B language transformer as well as EVA-02, the rank
first vision transformer that achieves 90% pretraining Imagenet accuracy. Code
is available at Github.
| [
{
"version": "v1",
"created": "Thu, 3 Apr 2025 15:30:43 GMT"
},
{
"version": "v2",
"created": "Fri, 4 Apr 2025 11:31:54 GMT"
},
{
"version": "v3",
"created": "Tue, 13 May 2025 17:54:56 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Li",
"Yuhang",
""
],
[
"Yin",
"Ruokai",
""
],
[
"Lee",
"Donghyun",
""
],
[
"Xiao",
"Shiting",
""
],
[
"Panda",
"Priyadarshini",
""
]
] |
2504.02970 | Bruce T. Draine | B. T. Draine | The Spheroidal Analog Method for Modeling Irregular Porous Aggregates | Published in The Astrophysical Journal | Astrophysical Journal, 985:10 (2025 May 20) | 10.3847/1538-4357/adc57d | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is shown that the optical properties of an irregular porous grain with
effective radius $a_{\rm eff}\lesssim 3\lambda$ (where $\lambda$ is the
wavelength) can be well approximated by a ``spheroidal analog'': a spheroid
with appropriate axial ratio and size, with a dielectric function obtained from
an effective medium theory. Prescriptions for specifying the axial ratio and
porosity of the spheroidal analog, based on simple geometric properties, are
given. The accuracy of the spheroidal analogue method is studied for irregular
grains with a range of structures and porosities. Different effective medium
theories are compared; Bruggeman's theory is found to give the best results.
The accuracy of the spheroidal analog method justifies the use of spheroids for
modeling absorption, scattering and polarization by interstellar,
circumstellar, or interplanetary dust.
| [
{
"version": "v1",
"created": "Thu, 3 Apr 2025 18:45:39 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 18:56:35 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Draine",
"B. T.",
""
]
] |
2504.03390 | Ben Deitmar | Ben Deitmar | Eigen-inference by Marchenko-Pastur inversion | 44 pages, 12 figures | null | null | null | math.ST stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A new formula for Marchenko-Pastur inversion is derived and used for
inference of population linear spectral statistics. The formula allows for fast
and accurate estimation of the Stieltjes transform of the population spectral
distribution $s_H(z)$, when $z$ is sufficiently far from the support of the
population spectral distribution $H$. If the dimension $d$ and the sample size
$n$ go to infinity simultaneously such that $\frac{d}{n} \rightarrow c>0$, the
estimation error is shown to be asymptotically less than
$\frac{n^{\varepsilon}}{n}$ for arbitrary $\varepsilon > 0$. By integrating
along a curve around the support of $H$, estimators for population linear
spectral statistics are constructed, which benefit from this convergence speed
of $\frac{n^{\varepsilon}}{n}$.
The new method of estimating the Stieltjes transforms $s_H(z)$ is also
applied to the numerical construction of estimators for the population
eigenvalues, which in a simulation study are demonstrated to outperform
state-of-the-art Ledoit-Wolf estimators.
| [
{
"version": "v1",
"created": "Fri, 4 Apr 2025 12:03:11 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 15:13:39 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Deitmar",
"Ben",
""
]
] |
2504.04330 | Shota Takahashi Dr. | Shota Takahashi, Sebastian Pokutta, Akiko Takeda | Fast Frank--Wolfe Algorithms with Adaptive Bregman Step-Size for Weakly
Convex Functions | 46 pages, 10 figures | null | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a Frank--Wolfe (FW) algorithm with an adaptive Bregman step-size
strategy for smooth adaptable (also called: relatively smooth) (weakly-) convex
functions. This means that the gradient of the objective function is not
necessarily Lipschitz continuous, and we only require the smooth adaptable
property. Compared to existing FW algorithms, our assumptions are less
restrictive. We establish convergence guarantees in various settings, such as
sublinear to linear convergence rates, depending on the assumptions for convex
and nonconvex objective functions. Assuming that the objective function is
weakly convex and satisfies the local quadratic growth condition, we provide
both local sublinear and local linear convergence regarding the primal gap. We
also propose a variant of the away-step FW algorithm using Bregman distances
over polytopes. We establish global faster (up to linear) convergence for
convex optimization under the H\"{o}lder error bound condition and its local
linear convergence for nonconvex optimization under the local quadratic growth
condition. Numerical experiments demonstrate that our proposed FW algorithms
outperform existing methods.
| [
{
"version": "v1",
"created": "Sun, 6 Apr 2025 02:54:05 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 02:33:39 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Takahashi",
"Shota",
""
],
[
"Pokutta",
"Sebastian",
""
],
[
"Takeda",
"Akiko",
""
]
] |
2504.04717 | Yubo Li | Yubo Li, Xiaobin Shen, Xinyu Yao, Xueying Ding, Yidi Miao, Ramayya
Krishnan, Rema Padman | Beyond Single-Turn: A Survey on Multi-Turn Interactions with Large
Language Models | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent advancements in large language models (LLMs) have revolutionized their
ability to handle single-turn tasks, yet real-world applications demand
sophisticated multi-turn interactions. This survey provides a comprehensive
review of recent advancements in evaluating and enhancing multi-turn
interactions in LLMs. Focusing on task-specific scenarios, from instruction
following in diverse domains such as math and coding to complex conversational
engagements in roleplay, healthcare, education, and even adversarial jailbreak
settings, we systematically examine the challenges of maintaining context,
coherence, fairness, and responsiveness over prolonged dialogues. The paper
organizes current benchmarks and datasets into coherent categories that reflect
the evolving landscape of multi-turn dialogue evaluation. In addition, we
review a range of enhancement methodologies under multi-turn settings,
including model-centric strategies (contextual learning, supervised
fine-tuning, reinforcement learning, and new architectures), external
integration approaches (memory-augmented, retrieval-based methods, and
knowledge graph), and agent-based techniques for collaborative interactions.
Finally, we discuss open challenges and propose future directions for research
to further advance the robustness and effectiveness of multi-turn interactions
in LLMs. Related resources and papers are available at
https://github.com/yubol-cmu/Awesome-Multi-Turn-LLMs.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 04:00:08 GMT"
},
{
"version": "v2",
"created": "Tue, 8 Apr 2025 03:58:37 GMT"
},
{
"version": "v3",
"created": "Tue, 13 May 2025 13:14:08 GMT"
},
{
"version": "v4",
"created": "Wed, 14 May 2025 01:48:30 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Li",
"Yubo",
""
],
[
"Shen",
"Xiaobin",
""
],
[
"Yao",
"Xinyu",
""
],
[
"Ding",
"Xueying",
""
],
[
"Miao",
"Yidi",
""
],
[
"Krishnan",
"Ramayya",
""
],
[
"Padman",
"Rema",
""
]
] |
2504.05009 | Huw Cheston | Huw Cheston, Reuben Bance, Peter M. C. Harrison | Deconstructing Jazz Piano Style Using Machine Learning | Paper: 40 pages, 11 figures, 1 table; added information on training
time + computation cost, corrections to Table 1. Supplementary material: 33
pages, 48 figures, 6 tables; corrections to Table S.5 | null | null | null | cs.SD cs.IR cs.LG eess.AS | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Artistic style has been studied for centuries, and recent advances in machine
learning create new possibilities for understanding it computationally.
However, ensuring that machine-learning models produce insights aligned with
the interests of practitioners and critics remains a significant challenge.
Here, we focus on musical style, which benefits from a rich theoretical and
mathematical analysis tradition. We train a variety of supervised-learning
models to identify 20 iconic jazz musicians across a carefully curated dataset
of 84 hours of recordings, and interpret their decision-making processes. Our
models include a novel multi-input architecture that enables four musical
domains (melody, harmony, rhythm, and dynamics) to be analysed separately.
These models enable us to address fundamental questions in music theory and
also advance the state-of-the-art in music performer identification (94%
accuracy across 20 classes). We release open-source implementations of our
models and an accompanying web application for exploring musical styles.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 12:37:39 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 15:56:36 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Cheston",
"Huw",
""
],
[
"Bance",
"Reuben",
""
],
[
"Harrison",
"Peter M. C.",
""
]
] |
2504.05199 | Yanfeng Hang | Kezhu Guo, Yanfeng Hang | Equivalence Theorems and Double-Copy Structure in Scattering Amplitudes
of Massive Kaluza-Klein States with Matter Interactions | 31 pages. Incorporating general N-point discussion and new
references. The typos have been corrected and the conclusion remains
unchanged | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the scattering amplitudes of massive Kaluza-Klein (KK) states
in compactified five-dimensional warped gauge and gravity theories. Focusing on
tree-level $2\to2$ processes, we analyze the leading-order amplitudes involving
bulk KK matter fields and KK gauge/gravitational Goldstone bosons. By imposing
the gauge theory equivalence theorem (GAET) and the gravitational equivalence
theorem (GRET) within warped KK theories, we systematically reconstruct the
leading-order amplitudes for physical KK gauge bosons and gravitons, thereby
circumventing the intricate energy cancellations inherent in physical
amplitudes. Within this framework, the correspondence between GAET and GRET
arises as a direct manifestation of the leading-order double-copy relation in
the high-energy expansion. This connection provides a foundation for extending
the BCJ double-copy construction to four-point amplitudes involving bulk KK
matter fields, and further generalizes to arbitrary $N$-point cases, enabling a
systematic derivation of the corresponding gravitational amplitudes with
consistent incorporation of KK matter fields at leading order.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 15:50:19 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 15:35:20 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Guo",
"Kezhu",
""
],
[
"Hang",
"Yanfeng",
""
]
] |
2504.05235 | Sneh Pandya | Sneh Pandya, Yuanyuan Yang, Nicholas Van Alfen, Jonathan Blazek, Robin
Walters | IAEmu: Learning Galaxy Intrinsic Alignment Correlations | 16 pages, 10 figures, 1 table | null | null | null | astro-ph.CO astro-ph.GA cs.LG | http://creativecommons.org/licenses/by/4.0/ | The intrinsic alignments (IA) of galaxies, a key contaminant in weak lensing
analyses, arise from correlations in galaxy shapes driven by tidal interactions
and galaxy formation processes. Accurate IA modeling is essential for robust
cosmological inference, but current approaches rely on perturbative methods
that break down on nonlinear scales or on expensive simulations. We introduce
IAEmu, a neural network-based emulator that predicts the galaxy
position-position ($\xi$), position-orientation ($\omega$), and
orientation-orientation ($\eta$) correlation functions and their uncertainties
using mock catalogs based on the halo occupation distribution (HOD) framework.
Compared to simulations, IAEmu achieves ~3% average error for $\xi$ and ~5% for
$\omega$, while capturing the stochasticity of $\eta$ without overfitting. The
emulator provides both aleatoric and epistemic uncertainties, helping identify
regions where predictions may be less reliable. We also demonstrate
generalization to non-HOD alignment signals by fitting to IllustrisTNG
hydrodynamical simulation data. As a fully differentiable neural network, IAEmu
enables $\sim$10,000$\times$ speed-ups in mapping HOD parameters to correlation
functions on GPUs, compared to CPU-based simulations. This acceleration
facilitates inverse modeling via gradient-based sampling, making IAEmu a
powerful surrogate model for galaxy bias and IA studies with direct
applications to Stage IV weak lensing surveys.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 16:19:50 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 16:12:07 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Pandya",
"Sneh",
""
],
[
"Yang",
"Yuanyuan",
""
],
[
"Van Alfen",
"Nicholas",
""
],
[
"Blazek",
"Jonathan",
""
],
[
"Walters",
"Robin",
""
]
] |
2504.06005 | Abubakar Fadul | Abubakar M. A. Fadul, Kamber R. Schwarz, Merel L. R. van 't Hoff, Jane
Huang, Jennifer B. Bergner, Tushar Suhasaria, and Jenny K. Calahan | A deep search for Complex Organic Molecules toward the protoplanetary
disk of V883 Ori | 42 pages, 27 figures, accepted for publication in Astronomical
Journal (AJ) | The Astronomical Journal, Volume 169, Number 6, Page 307 (2025) | 10.3847/1538-3881/adc998 | null | astro-ph.SR astro-ph.EP astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Complex Organic Molecules (COMs) in the form of prebiotic molecules are
potentially building blocks of life. Using Atacama Large
Millimeter/submillimeter Array (ALMA) Band 7 observations in spectral scanning
mode, we carried out a deep search for COMs within the disk of V883 Ori,
covering frequency ranges of $\sim$ 348 - 366 GHz. V883 Ori is an FUor object
currently undergoing an accretion burst, which increases its luminosity and
consequently increases the temperature of the surrounding protoplanetary disk,
facilitating the detection of COMs in the gas phase. We identified 26
molecules, including 14 COMs and 12 other molecules, with first detection in
this source of the molecules: CH3OD, H2C17O, and H213CO. We searched for
multiple nitrogen-bearing COMs, as CH3CN had been the only nitrogen-bearing COM
that has been identified so far in this source. We also detected CH3CN, and
tentatively detect CH3CH2CN, CH2CHCN, CH3OCN, CH3NCO, and NH2CHO. We compared
the abundances relative to CH3OH with those in the handful of objects with
previous detections of these species: the Class 0 protostars IRAS 16293-2422 A,
IRAS 16293-2422 B and B1-c, the high-mass star-forming region Sagittarius B2
(North), the Solar System comet 67P/Churyumov-Gerasimenko, and the
protoplanetary disk of Oph-IRS 48. We report $\sim$ 1 to 3 orders of magnitude
higher abundances compared to Class 0 protostars and $\sim$ 1 to 3 orders of
magnitude lower abundances compared to the protoplanetary disk, Sagittarius B2
(North), and 67P/C-G. These results indicate that the protoplanetary disk phase
could contribute to build up of COMs.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 13:15:46 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Fadul",
"Abubakar M. A.",
""
],
[
"Schwarz",
"Kamber R.",
""
],
[
"Hoff",
"Merel L. R. van 't",
""
],
[
"Huang",
"Jane",
""
],
[
"Bergner",
"Jennifer B.",
""
],
[
"Suhasaria",
"Tushar",
""
],
[
"Calahan",
"Jenny K.",
""
]
] |
2504.06151 | Jacopo Tagliabue | Yifan Dai, Jacopo Tagliabue, Andrea Arpaci-Dusseau, Remzi
Arpaci-Dusseau, Tyler R. Caraza-Harter | Zerrow: True Zero-Copy Arrow Pipelines in Bauplan | Pre-print conf submission | null | null | null | cs.OS cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Bauplan is a FaaS-based lakehouse specifically built for data pipelines: its
execution engine uses Apache Arrow for data passing between the nodes in the
DAG. While Arrow is known as the "zero copy format", in practice, limited Linux
kernel support for shared memory makes it difficult to avoid copying entirely.
In this work, we introduce several new techniques to eliminate nearly all
copying from pipelines: in particular, we implement a new kernel module that
performs de-anonymization, thus eliminating a copy to intermediate data. We
conclude by sharing our preliminary evaluation on different workloads types, as
well as discussing our plan for future improvements.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 15:46:52 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 21:03:41 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Dai",
"Yifan",
""
],
[
"Tagliabue",
"Jacopo",
""
],
[
"Arpaci-Dusseau",
"Andrea",
""
],
[
"Arpaci-Dusseau",
"Remzi",
""
],
[
"Caraza-Harter",
"Tyler R.",
""
]
] |
2504.06513 | Xinyi Wang | Xinyi Wang, Taekyung Kim, Bardh Hoxha, Georgios Fainekos and Dimitra
Panagou | Safe Navigation in Uncertain Crowded Environments Using Risk Adaptive
CVaR Barrier Functions | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Robot navigation in dynamic, crowded environments poses a significant
challenge due to the inherent uncertainties in the obstacle model. In this
work, we propose a risk-adaptive approach based on the Conditional
Value-at-Risk Barrier Function (CVaR-BF), where the risk level is automatically
adjusted to accept the minimum necessary risk, achieving a good performance in
terms of safety and optimization feasibility under uncertainty. Additionally,
we introduce a dynamic zone-based barrier function which characterizes the
collision likelihood by evaluating the relative state between the robot and the
obstacle. By integrating risk adaptation with this new function, our approach
adaptively expands the safety margin, enabling the robot to proactively avoid
obstacles in highly dynamic environments. Comparisons and ablation studies
demonstrate that our method outperforms existing social navigation approaches,
and validate the effectiveness of our proposed framework.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 01:23:44 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 15:20:52 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Wang",
"Xinyi",
""
],
[
"Kim",
"Taekyung",
""
],
[
"Hoxha",
"Bardh",
""
],
[
"Fainekos",
"Georgios",
""
],
[
"Panagou",
"Dimitra",
""
]
] |
2504.06566 | Renyuan Xu | Minshuo Chen, Renyuan Xu, Yumin Xu and Ruixun Zhang | Diffusion Factor Models: Generating High-Dimensional Returns with Factor
Structure | null | null | null | null | q-fin.ST cs.LG q-fin.MF | http://creativecommons.org/licenses/by/4.0/ | Financial scenario simulation is essential for risk management and portfolio
optimization, yet it remains challenging especially in high-dimensional and
small data settings common in finance. We propose a diffusion factor model that
integrates latent factor structure into generative diffusion processes,
bridging econometrics with modern generative AI to address the challenges of
the curse of dimensionality and data scarcity in financial simulation. By
exploiting the low-dimensional factor structure inherent in asset returns, we
decompose the score function--a key component in diffusion models--using
time-varying orthogonal projections, and this decomposition is incorporated
into the design of neural network architectures. We derive rigorous statistical
guarantees, establishing nonasymptotic error bounds for both score estimation
at O(d^{5/2} n^{-2/(k+5)}) and generated distribution at O(d^{5/4}
n^{-1/2(k+5)}), primarily driven by the intrinsic factor dimension k rather
than the number of assets d, surpassing the dimension-dependent limits in the
classical nonparametric statistics literature and making the framework viable
for markets with thousands of assets. Numerical studies confirm superior
performance in latent subspace recovery under small data regimes. Empirical
analysis demonstrates the economic significance of our framework in
constructing mean-variance optimal portfolios and factor portfolios. This work
presents the first theoretical integration of factor structure with diffusion
models, offering a principled approach for high-dimensional financial
simulation with limited data. Our code is available at
https://github.com/xymmmm00/diffusion_factor_model.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 04:01:35 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 03:29:54 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Chen",
"Minshuo",
""
],
[
"Xu",
"Renyuan",
""
],
[
"Xu",
"Yumin",
""
],
[
"Zhang",
"Ruixun",
""
]
] |
2504.07136 | James Beattie | James R. Beattie, Christoph Federrath, Ralf S. Klessen, Salvatore
Cielo, Amitava Bhattacharjee | The spectrum of magnetized turbulence in the interstellar medium | 8 pages main text. 24 pages total. 3 main text figure. 7 figures
total. arXiv admin note: substantial text overlap with arXiv:2405.16626 | null | 10.1038/s41550-025-02551-5 | null | astro-ph.GA astro-ph.SR nlin.CD physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | The interstellar medium (ISM) of our Galaxy is magnetized, compressible and
turbulent, influencing many key ISM properties, like star formation, cosmic ray
transport, and metal and phase mixing. Yet, basic statistics describing
compressible, magnetized turbulence remain uncertain. Utilizing grid
resolutions up to $10,080^3$ cells, we simulate highly-compressible, magnetized
ISM-style turbulence with a magnetic field maintained by a small-scale dynamo.
We measure two coexisting kinetic energy cascades, $\mathcal{E}_{{\rm kin}}(k)
\propto k^{-n}$, in the turbulence, separating the plasma into scales that are
non-locally interacting, supersonic and weakly magnetized $(n=2.01\pm
0.03\approx 2)$ and locally interacting, subsonic and highly magnetized
$(n=1.465\pm 0.002\approx 3/2)$, where $k$ is the wavenumber. We show that the
$3/2$ spectrum can be explained with scale-dependent kinetic energy fluxes and
velocity-magnetic field alignment. On the highly magnetized modes, the magnetic
energy spectrum forms a local cascade $(n=1.798\pm 0.001\approx 9/5)$,
deviating from any known \textit{ab initio} theory. With a new generation of
radio telescopes coming online, these results provide a means to directly test
if the ISM in our Galaxy is maintained by the compressible turbulent motions
from within it.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 17:54:30 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Beattie",
"James R.",
""
],
[
"Federrath",
"Christoph",
""
],
[
"Klessen",
"Ralf S.",
""
],
[
"Cielo",
"Salvatore",
""
],
[
"Bhattacharjee",
"Amitava",
""
]
] |
2504.08088 | Chun Yin Hui | Chun-Yin Hui | On distribution of supersingular primes of abelian varieties and K3
surfaces | Revision | null | null | null | math.NT math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let X be an abelian variety or a K3 surface defined over a number field K. We
prove that the density of the supersingular primes of X is zero if X is non-CM.
By applying an effective Chebotarev density theorem of Serre, we obtain
asymptotic upper bounds of the counting function for these supersingular
primes.
| [
{
"version": "v1",
"created": "Thu, 10 Apr 2025 19:34:39 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 16:15:21 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Hui",
"Chun-Yin",
""
]
] |
2504.09204 | Patrick Polo | Patrick Polo | Classification of the root systems $R(m)$ | Minor typos corrected. For types $C_n, B_n, D_{n+1}$ the case where
$n<m$, omitted in the first version, has been added | null | null | null | math.RT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $R$ be a reduced irreducible root system, $h$ its Coxeter number and $m$
a positive integer smaller than $h$. Choose of base of $R$, whence a
corresponding height function, and let $R(m)$ be the set of roots whose height
is a multiple of $m$. In a recent paper, S. Nadimpalli, S. Pattanayak and D.
Prasad studied, for the purposes of character theory at torsion elements, the
root systems $R(m)$; in particular, they introduced a constant $d_m$ which is
always the dimension of a representation of the semisimple, simply-connected
group with root system dual to $R(m)$ and equals $1$ if the roots of height $m$
form a base of $R(m)$, and proved this property when $R$ is of type $A$ or $C$,
and also in type $B$ if $m$ is odd. In this paper, we complete their analysis
by determining a base of $R(m)$ and computing the constant $d_m$ in all cases.
| [
{
"version": "v1",
"created": "Sat, 12 Apr 2025 13:08:17 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 06:48:27 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Polo",
"Patrick",
""
]
] |
2504.09838 | Fei Sun | Hanchuan Chen, Fei Sun, Yichao Liu, Shuai Zhang, Hongming Fei, Zhihui
Chen | Broadband source-surrounded cloak for on-chip antenna radiation pattern
protection | 5 pages, 5 figures, 1 table | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As the frequency range of electromagnetic wave communication continues to
expand and the integration of integrated circuits increases, electromagnetic
waves emitted by on-chip antennas are prone to scattering from electronic
components, which limits further improvements in integration and the protection
of radiation patterns. Cloaks can be used to reduce electromagnetic scattering;
however, they cannot achieve both broadband and omnidirectional effectiveness
simultaneously. Moreover, their operating modes are typically designed for
scenarios where the source is located outside the cloak, making it difficult to
address this problem. In this work, we propose a dispersionless
air-impedance-matched metamaterial over the 2-8 GHz bandwidth that achieves an
adjustable effective refractive index ranging from 1.1 to 1.5, with
transmittance maintained above 93%. Based on this metamaterial, we introduce a
broadband source-surrounded cloak that can guide electromagnetic waves from a
broadband source surrounded by the cloak in any propagation direction to bypass
obstacles and reproduce the original wavefronts outside the cloak. Thereby
protecting the radiation pattern from distortion due to scattering caused by
obstacles. Our work demonstrates significant potential for enhancing the
integration density of integrated circuits and improving the operational
stability of communication systems.
| [
{
"version": "v1",
"created": "Mon, 14 Apr 2025 03:17:30 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 03:03:27 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Chen",
"Hanchuan",
""
],
[
"Sun",
"Fei",
""
],
[
"Liu",
"Yichao",
""
],
[
"Zhang",
"Shuai",
""
],
[
"Fei",
"Hongming",
""
],
[
"Chen",
"Zhihui",
""
]
] |
2504.10205 | Alexander Migdal | Alexander Migdal | Dual Theory of Turbulent Mixing | 9 pages, eight figures, fixed some errors in computations, including
new powers in the decay spectrum | null | null | null | physics.flu-dyn hep-th nlin.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present an exact analytic solution for incompressible turbulent mixing
described by 3D NS equations, with a passive scalar (concentration,
temperature, or other scalar field) driven by the turbulent velocity field.
Using our recent solution of decaying turbulence in terms of the Euler
ensemble, we represent the correlation functions of a passive scalar as
statistical averages over this ensemble. The statistical limit, corresponding
to decaying turbulence, can be computed in quadrature by a multi-dimensional
Mellin-Barnes integral representation. We find the decay spectrum and the
scaling functions of the pair correlation. Full comparison with physical and
real experiments is postponed till the future detailed computation of the
spectrum in our solution.
| [
{
"version": "v1",
"created": "Mon, 14 Apr 2025 13:13:36 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 14:38:31 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Migdal",
"Alexander",
""
]
] |
2504.10612 | Michal Balcerak | Michal Balcerak, Tamaz Amiranashvili, Antonio Terpin, Suprosanna Shit,
Sebastian Kaltenbach, Petros Koumoutsakos, Bjoern Menze | Energy Matching: Unifying Flow Matching and Energy-Based Models for
Generative Modeling | null | null | null | null | cs.LG cs.AI stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The most widely used generative models map noise and data distributions by
matching flows or scores. However, they struggle to incorporate partial
observations and additional priors--something energy-based models (EBMs) handle
elegantly by simply adding corresponding scalar energy terms. We address this
issue by proposing Energy Matching, a framework that endows flow-based
approaches with the flexibility of EBMs. Far from the data manifold, samples
move along curl-free, optimal transport paths from noise to data. As they
approach the data manifold, an entropic energy term guides the system into a
Boltzmann equilibrium distribution, explicitly capturing the underlying
likelihood structure of the data. We parameterize this dynamic with a single
time-independent scalar field, which serves as both a powerful generator and a
flexible prior for effective regularization of inverse problems. Our method
substantially outperforms existing EBMs on CIFAR-10 and ImageNet generation in
terms of fidelity, while retaining simulation-free training of transport-based
approaches away from the data manifold. Furthermore, we leverage the method's
flexibility to introduce an interaction energy that supports diverse mode
exploration, which we demonstrate in a controlled protein-generation setting.
Our approach focuses on learning a scalar potential energy--without
time-conditioning, auxiliary generators, or additional networks--which marks a
significant departure from recent EBM methods. We believe that this simplified
framework significantly advances EBMs capabilities and paves the way for their
wider adoption in generative modeling across diverse domains.
| [
{
"version": "v1",
"created": "Mon, 14 Apr 2025 18:10:58 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 12:10:11 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Balcerak",
"Michal",
""
],
[
"Amiranashvili",
"Tamaz",
""
],
[
"Terpin",
"Antonio",
""
],
[
"Shit",
"Suprosanna",
""
],
[
"Kaltenbach",
"Sebastian",
""
],
[
"Koumoutsakos",
"Petros",
""
],
[
"Menze",
"Bjoern",
""
]
] |
2504.11383 | Maryam Hakimzadeh | Wei Wang, Maryam Hakimzadeh, Haihui Ruan, Somdatta Goswami | Accelerating Multiscale Modeling with Hybrid Solvers: Coupling FEM and
Neural Operators with Domain Decomposition | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Numerical solvers for PDEs face challenges in balancing computational cost
and accuracy, particularly for multiscale and dynamical systems. Neural
operators (NOs) can significantly speed up simulations; however, they face
challenges such as error accumulation for dynamical systems and limited
generalization in multiphysics problems. This work introduces a novel hybrid
framework that integrates PI-NO with finite element method (FE) through domain
decomposition and leverages numerical analysis for time marching. The core
innovation lies in efficient coupling FE and NO subdomains via a Schwarz
alternating method: regions with complex, nonlinear, or high-gradient behavior
are resolved using a pretrained NO, while the remainder is handled by
conventional FE. To address the challenges of dynamic systems, we embed a
time-stepping scheme directly into the NO architecture, substantially reducing
long-term error propagation. Also, an adaptive subdomain evolution strategy
enables the ML resolved region to expand dynamically, capturing emerging fine
scale features without remeshing. The framework efficacy has been validated
across a range of problems, spanning static, quasi-static, and dynamic regimes
(e.g., linear elasticity, hyperelasticity, and elastodynamics), demonstrating
accelerated convergence (up to 20% improvement in convergence compared to
conventional FE coupling) while preserving solution fidelity with error margins
consistently below 1%. Our study shows that our hybrid solver: (1) maintains
solution continuity across subdomain interfaces, (2) reduces computational
costs by eliminating fine mesh requirements, (3) mitigates error accumulation
in time dependent simulations, and (4) enables automatic adaptation to evolving
physical phenomena. This work bridges the gap between numerical methods and
AI-driven surrogates, offering a scalable pathway for high-fidelity multiscale
simulations.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:54:04 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 12:26:48 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 15:14:24 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Wang",
"Wei",
""
],
[
"Hakimzadeh",
"Maryam",
""
],
[
"Ruan",
"Haihui",
""
],
[
"Goswami",
"Somdatta",
""
]
] |
2504.11436 | Eleanor Dillon | Eleanor Wiske Dillon, Sonia Jaffe, Nicole Immorlica, Christopher T.
Stanton | Shifting Work Patterns with Generative AI | null | null | null | null | econ.GN cs.LG q-fin.EC | http://creativecommons.org/licenses/by/4.0/ | We present evidence on how generative AI changes the work patterns of
knowledge workers using data from a 6-month-long, cross-industry, randomized
field experiment. Half of the 7,137 workers in the study received access to a
generative AI tool integrated into the applications they already used for
emails, document creation, and meetings. We find that access to the AI tool
during the first year of its release primarily impacted behaviors that workers
could change independently and not behaviors that require coordination to
change: workers who used the tool in more than half of the sample weeks spent
3.6 fewer hours, or 31% less time on email each week (intent to treat estimate
is 1.3 hours) and completed documents moderately faster, but did not
significantly change time spent in meetings.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:52:00 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 22:28:06 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Dillon",
"Eleanor Wiske",
""
],
[
"Jaffe",
"Sonia",
""
],
[
"Immorlica",
"Nicole",
""
],
[
"Stanton",
"Christopher T.",
""
]
] |
2504.12122 | Wenshuai Liu | Wenshuai Liu | Origin of the Moon's Earth-like isotopic composition from giant impact
on a differential rotating proto-Earth | 4 pages, 3 figures | null | null | null | astro-ph.EP astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | According to the giant impact theory, the Moon formed by accreting the
circum-terrestrial debris disk produced by Theia colliding with the
proto-Earth. The giant impact theory can explain most of the properties of the
Earth-Moon system, however, simulations of giant impact between a planetary
embryo and the growing proto-Earth indicate that the materials in the
circum-terrestrial debris disk produced by the impact originate mainly from the
impactor. Thus, the giant impact theory has difficulty explaining the Moon's
Earth-like isotopic compositions. More materials from the proto-Earth could be
delivered to the circum-terrestrial debris disk when a slightly sub-Mars-sized
body collides with a fast rotating planet of rigid rotation but the resulting
angular momentum is too large compared with that of the current Earth-Moon
system. Since planetesimals accreted by the proto-Earth hit the surface of the
proto-Earth, enhancing the rotation rate of the surface of the proto-Earth. The
surface's fast rotation rate relative to the slow rotation rate of the inner
region of the proto-Earth leads to transfer of angular momentum from surface to
inner, resulting in the differential rotation. Here, we show that the giant
impact of a sub-Mars-sized body on a differential rotating proto-Earth with a
fast rotating outer region and a relative slow rotating inner region could
result in a circum-terrestrial debris disk with materials predominately from
the proto-Earth without violating the angular momentum constraint. The theory
proposed here may provide a viable way of explaining the similarity in the
isotopic compositions of the Earth and Moon.
| [
{
"version": "v1",
"created": "Wed, 16 Apr 2025 14:34:51 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 14:15:00 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Liu",
"Wenshuai",
""
]
] |
2504.12693 | Jing Yu | Jing Yu and Jie-Xiang Zhu | Counting degree-constrained orientations | 9 pages. Fixed minor typos | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the enumeration of graph orientations under local degree
constraints. Given a finite graph $G = (V, E)$ and a family of admissible sets
$\{\mathsf P_v \subseteq \mathbb{Z} : v \in V\}$, let $\mathcal N (G; \prod_{v
\in V} \mathsf P_v)$ denote the number of orientations in which the out-degree
of each vertex $v$ lies in $P_v$. We prove a general duality formula expressing
$\mathcal N(G; \prod_{v \in V} \mathsf P_v)$ as a signed sum over edge subsets,
involving products of coefficient sums associated with $\{\mathsf P_v\}_{v \in
V}$, from a family of polynomials. Our approach employs gauge transformations,
a technique rooted in statistical physics and holographic algorithms. We also
present a probabilistic derivation of the same identity, interpreting the
orientation-generating polynomial as the expectation of a random polynomial
product. As applications, we obtain explicit formulas for the number of even
orientations and for mixed Eulerian-even orientations on general graphs. Our
formula generalizes a result of Borb\'enyi and Csikv\'ari on Eulerian
orientations of graphs.
| [
{
"version": "v1",
"created": "Thu, 17 Apr 2025 06:40:50 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 20:34:41 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Yu",
"Jing",
""
],
[
"Zhu",
"Jie-Xiang",
""
]
] |
2504.12777 | James Rudd-Jones | James Rudd-Jones, Mirco Musolesi, Mar\'ia P\'erez-Ortiz | Multi-Agent Reinforcement Learning Simulation for Environmental Policy
Synthesis | Published in AAMAS'25 Blue Sky Ideas Track | null | null | null | cs.MA cs.AI | http://creativecommons.org/licenses/by/4.0/ | Climate policy development faces significant challenges due to deep
uncertainty, complex system dynamics, and competing stakeholder interests.
Climate simulation methods, such as Earth System Models, have become valuable
tools for policy exploration. However, their typical use is for evaluating
potential polices, rather than directly synthesizing them. The problem can be
inverted to optimize for policy pathways, but the traditional optimization
approaches often struggle with non-linear dynamics, heterogeneous agents, and
comprehensive uncertainty quantification. We propose a framework for augmenting
climate simulations with Multi-Agent Reinforcement Learning (MARL) to address
these limitations. We identify key challenges at the interface between climate
simulations and the application of MARL in the context of policy synthesis,
including reward definition, scalability with increasing agents and state
spaces, uncertainty propagation across linked systems, and solution validation.
Additionally, we discuss challenges in making MARL-derived solutions
interpretable and useful for policy-makers. Our framework provides a foundation
for more sophisticated climate policy exploration while acknowledging important
limitations and areas for future research.
| [
{
"version": "v1",
"created": "Thu, 17 Apr 2025 09:18:04 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 16:44:01 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Rudd-Jones",
"James",
""
],
[
"Musolesi",
"Mirco",
""
],
[
"Pérez-Ortiz",
"María",
""
]
] |
2504.13118 | Denis Burgarella | Denis Burgarella, V\'eronique Buat, Patrice Theul\'e, Jorge Zavala,
Mark Dickinson, Pablo Arrabal Haro, Micaela B. Bagley, M\'ed\'eric Boquien,
Nikko Cleri, Tim Dewachter, Henry C. Ferguson, Vital Fern\`andez, Steven L.
Finkelstein, Eric Gawiser, Andrea Grazian, Norman Grogin, Benne W. Holwerda,
Jeyhan S. Kartaltepe, Lisa Kewley, Allison Kirkpatrick, Dale Kocevski, Anton
M. Koekemoer, Arianna Long, Jennifer Lotz, Ray A. Lucas, Bahram Mobasher,
Casey Papovich, Pablo G. P\'erez-Gonz\`alez, Nor Pirzkal, Swara Ravindranath,
Giulia Rodighiero, Yannick Roehlly, Caitlin Rose, Lise-Marie Seill\'e, Rachel
Somerville, Steve Wilkins, Guang Yang, and L. Y. Aaron Yung | CEERS: Forging the First Dust Grains in the Universe? A Population of
Galaxies with spectroscopically-derived Extremely Low Dust Attenuation
(GELDA) at 4.0<z<11.4 | Accepted in A&A (May 2025) | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We investigate the coevolution of metals and dust in 173 galaxies at
$4.0<z<11.4$ using NIRSpec spectroscopy. Focusing on galaxies with extremely
low dust attenuation, we explore their physical processes using a new CIGALE
version that incorporates spectroscopic and photometric data. Comparing
observations with models, we derive key physical parameters. We identify 49
galaxies with extremely low dust attenuation (GELDAs), characterized by
$A_{FUV}=0$ within $2\sigma$ and $M_{star}<10^9 M_\odot$. The stacked spectra
reveal a very blue UV slope ($\beta_{FUV} =-2.451\pm-0.066$) and Balmer
decrement H$\alpha$/H$\beta = 2.932\pm-0.660$, consistent with no dust
attenuation. GELDAs are more common at $z>8.8$ (83.3%) than at $z<8.8$ (26.3%)
suggesting they dominate the early Universe. Assuming a prior FIR dust spectrum
(from ALPINE), we examine dust-stellar mass trends. The $M_{dust}$ vs.
$M_{star}$ diagram shows upper and lower sequences linked by possible
transitional galaxies. A transition at $M_{star}=10^8.5 M_\odot$
($Z_{crit}=12+\log_{10}(O/H)=7.60$ or $Z/Z_\odot=0.1$) may mark the shift from
stellar dust production to ISM grain growth, in agreement with theoretical
predictions. Our full sample has a high mean gas fraction ($f_{gas}>0.9$),
indicating retained gas across all galaxies. Their small sizes and large gas
masses imply high gas surface densities but relatively low star formation
efficiency. High-redshift GELDAs may naturally explain the observed excess of
bright galaxies at $z>9$ compared to theoretical expectations.
| [
{
"version": "v1",
"created": "Thu, 17 Apr 2025 17:37:14 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 15:10:09 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Burgarella",
"Denis",
""
],
[
"Buat",
"Véronique",
""
],
[
"Theulé",
"Patrice",
""
],
[
"Zavala",
"Jorge",
""
],
[
"Dickinson",
"Mark",
""
],
[
"Haro",
"Pablo Arrabal",
""
],
[
"Bagley",
"Micaela B.",
""
],
[
"Boquien",
"Médéric",
""
],
[
"Cleri",
"Nikko",
""
],
[
"Dewachter",
"Tim",
""
],
[
"Ferguson",
"Henry C.",
""
],
[
"Fernàndez",
"Vital",
""
],
[
"Finkelstein",
"Steven L.",
""
],
[
"Gawiser",
"Eric",
""
],
[
"Grazian",
"Andrea",
""
],
[
"Grogin",
"Norman",
""
],
[
"Holwerda",
"Benne W.",
""
],
[
"Kartaltepe",
"Jeyhan S.",
""
],
[
"Kewley",
"Lisa",
""
],
[
"Kirkpatrick",
"Allison",
""
],
[
"Kocevski",
"Dale",
""
],
[
"Koekemoer",
"Anton M.",
""
],
[
"Long",
"Arianna",
""
],
[
"Lotz",
"Jennifer",
""
],
[
"Lucas",
"Ray A.",
""
],
[
"Mobasher",
"Bahram",
""
],
[
"Papovich",
"Casey",
""
],
[
"Pérez-Gonzàlez",
"Pablo G.",
""
],
[
"Pirzkal",
"Nor",
""
],
[
"Ravindranath",
"Swara",
""
],
[
"Rodighiero",
"Giulia",
""
],
[
"Roehlly",
"Yannick",
""
],
[
"Rose",
"Caitlin",
""
],
[
"Seillé",
"Lise-Marie",
""
],
[
"Somerville",
"Rachel",
""
],
[
"Wilkins",
"Steve",
""
],
[
"Yang",
"Guang",
""
],
[
"Yung",
"L. Y. Aaron",
""
]
] |
2504.13410 | Ning Wang | Ning Wang | Refinement of an analytical capture cross section formula | 5 figures, 1 table | null | null | null | nucl-th nucl-ex | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | An analytical formula with high accuracy is proposed for a systematic
description of the capture cross sections from light to superheavy reaction
systems. Based on the empirical barrier distribution (EBD) method, three key
input quantities are refined by introducing nuclear surface correction to the
Coulomb parameter $z$ for calculating the barrier height, incorporating the
$Q$-value and shell correction into the barrier distribution width
calculations, and considering the deep inelastic scattering effects of
superheavy systems on the barrier radius. With these refinements, the model
accuracy of the EBD method, not only the accuracy of the calculated barrier
height but also the accuracy of the predicted capture cross sections, is
substantially improved. The average deviation (in logarithmic scale) between
the predicted cross sections and the data for a total of 426 reaction systems
with $ 35 < Z_1 Z_2 < 2600$ is sharply reduced from 3.485 to 0.113.
| [
{
"version": "v1",
"created": "Fri, 18 Apr 2025 02:12:39 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 10:12:01 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Wang",
"Ning",
""
]
] |
2504.13807 | Zhengtong Xu | Zhengtong Xu, Zichen Miao, Qiang Qiu, Zhe Zhang, Yu She | DiffOG: Differentiable Policy Trajectory Optimization with
Generalizability | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Imitation learning-based visuomotor policies excel at manipulation tasks but
often produce suboptimal action trajectories compared to model-based methods.
Directly mapping camera data to actions via neural networks can result in jerky
motions and difficulties in meeting critical constraints, compromising safety
and robustness in real-world deployment. For tasks that require high robustness
or strict adherence to constraints, ensuring trajectory quality is crucial.
However, the lack of interpretability in neural networks makes it challenging
to generate constraint-compliant actions in a controlled manner. This paper
introduces differentiable policy trajectory optimization with generalizability
(DiffOG), a learning-based trajectory optimization framework designed to
enhance visuomotor policies. By leveraging the proposed differentiable
formulation of trajectory optimization with transformer, DiffOG seamlessly
integrates policies with a generalizable optimization layer. DiffOG refines
action trajectories to be smoother and more constraint-compliant while
maintaining alignment with the original demonstration distribution, thus
avoiding degradation in policy performance. We evaluated DiffOG across 11
simulated tasks and 2 real-world tasks. The results demonstrate that DiffOG
significantly enhances the trajectory quality of visuomotor policies while
having minimal impact on policy performance, outperforming trajectory
processing baselines such as greedy constraint clipping and penalty-based
trajectory optimization. Furthermore, DiffOG achieves superior performance
compared to existing constrained visuomotor policy. Please visit the project
website for more details: https://zhengtongxu.github.io/diffog-website/.
| [
{
"version": "v1",
"created": "Fri, 18 Apr 2025 17:20:27 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 03:17:11 GMT"
},
{
"version": "v3",
"created": "Tue, 13 May 2025 23:33:09 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Xu",
"Zhengtong",
""
],
[
"Miao",
"Zichen",
""
],
[
"Qiu",
"Qiang",
""
],
[
"Zhang",
"Zhe",
""
],
[
"She",
"Yu",
""
]
] |
2504.14219 | Alara Dirik | Alara Dirik, Tuanfeng Wang, Duygu Ceylan, Stefanos Zafeiriou, Anna
Fr\"uhst\"uck | PRISM: A Unified Framework for Photorealistic Reconstruction and
Intrinsic Scene Modeling | null | null | null | null | cs.GR cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present PRISM, a unified framework that enables multiple image generation
and editing tasks in a single foundational model. Starting from a pre-trained
text-to-image diffusion model, PRISM proposes an effective fine-tuning strategy
to produce RGB images along with intrinsic maps (referred to as X layers)
simultaneously. Unlike previous approaches, which infer intrinsic properties
individually or require separate models for decomposition and conditional
generation, PRISM maintains consistency across modalities by generating all
intrinsic layers jointly. It supports diverse tasks, including text-to-RGBX
generation, RGB-to-X decomposition, and X-to-RGBX conditional generation.
Additionally, PRISM enables both global and local image editing through
conditioning on selected intrinsic layers and text prompts. Extensive
experiments demonstrate the competitive performance of PRISM both for intrinsic
image decomposition and conditional image generation while preserving the base
model's text-to-image generation capability.
| [
{
"version": "v1",
"created": "Sat, 19 Apr 2025 08:03:06 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 12:50:18 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Dirik",
"Alara",
""
],
[
"Wang",
"Tuanfeng",
""
],
[
"Ceylan",
"Duygu",
""
],
[
"Zafeiriou",
"Stefanos",
""
],
[
"Frühstück",
"Anna",
""
]
] |
2504.14988 | Hongtao Yu | Hong-Tao Yu and Xiu-Shen Wei and Yuxin Peng and Serge Belongie | Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A
Comprehensive Evaluation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated
remarkable multimodal perception capabilities, garnering significant attention.
While numerous evaluation studies have emerged, assessing LVLMs both
holistically and on specialized tasks, fine-grained image tasks-fundamental to
computer vision-remain largely unexplored. To fill this gap, we introduce a
comprehensive fine-grained evaluation benchmark, i.e., FG-BMK, comprising 1.01
million questions and 0.33 million images. Our evaluation systematically
examines LVLMs from both human-oriented and machine-oriented perspectives,
focusing on their semantic recognition and fine-grained feature representation
capabilities. Through extensive experiments on twelve representative
LVLMs/VLMs, we uncover key findings regarding the influence of training
paradigms, modality alignment, perturbation susceptibility, and fine-grained
category reasoning on task performance. This work provides critical insights
into the limitations of current LVLMs and offers guidance for future data
construction and model design in the development of more advanced LVLMs. Our
code is open-source and available at https://github.com/SEU-VIPGroup/FG-BMK.
| [
{
"version": "v1",
"created": "Mon, 21 Apr 2025 09:30:41 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 02:33:48 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Yu",
"Hong-Tao",
""
],
[
"Wei",
"Xiu-Shen",
""
],
[
"Peng",
"Yuxin",
""
],
[
"Belongie",
"Serge",
""
]
] |
2504.15245 | Chunjian Zhang | Chunjian Zhang, Jiangyong Jia, Jinhui Chen, Chun Shen, Lumeng Liu | Probing the octupole deformation of $^{238}$U in high-energy nuclear
collisions | 6 pages, 4 figures | null | null | null | nucl-th nucl-ex | http://creativecommons.org/licenses/by/4.0/ | Some atomic nuclei exhibit ``pear" shapes arising from octupole deformation
($\beta_3$), though direct experimental evidences for such exotic shapes
remains scarce. Low-energy model studies suggest $^{238}$U may have a modest
octupole deformation arising from collective vibrational degrees of freedom, in
addition to a large prolate shape. We investigated the impact of this modest
octupole shape on observables involving triangular flow ($v_3$) in high-energy
nuclear collisions. Using a hydrodynamic framework, we show $v_3$ and its
correlation with mean transverse momentum, $\langle v_3^2 \delta p_{\rm{T}}
\rangle$, exhibit strong sensitivity to $\beta_3$. We found that $\langle
v_3^2\rangle$ follows a linear increase with $\beta_3^2$, while the
characteristic anticorrelation in $\langle v_3^2 \delta p_{\rm{T}} \rangle$
shows a pronounced $\beta_3$-dependent suppression. Our findings demonstrate
that the collective-flow-assisted nuclear imaging method in high-energy nuclear
collisions, when compared with experimental data, can provide unique
quantitative constraints on higher-order deformations.
| [
{
"version": "v1",
"created": "Mon, 21 Apr 2025 17:17:33 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 15:45:02 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Zhang",
"Chunjian",
""
],
[
"Jia",
"Jiangyong",
""
],
[
"Chen",
"Jinhui",
""
],
[
"Shen",
"Chun",
""
],
[
"Liu",
"Lumeng",
""
]
] |
2504.16472 | Mark Harman | Mark Harman, Peter O'Hearn, Shubho Sengupta | Harden and Catch for Just-in-Time Assured LLM-Based Software Testing:
Open Research Challenges | To Appear as keynote paper at FSE 2025 | null | null | null | cs.SE cs.AI | http://creativecommons.org/licenses/by/4.0/ | Despite decades of research and practice in automated software testing,
several fundamental concepts remain ill-defined and under-explored, yet offer
enormous potential real-world impact. We show that these concepts raise
exciting new challenges in the context of Large Language Models for software
test generation. More specifically, we formally define and investigate the
properties of hardening and catching tests. A hardening test is one that seeks
to protect against future regressions, while a catching test is one that
catches such a regression or a fault in new functionality introduced by a code
change. Hardening tests can be generated at any time and may become catching
tests when a future regression is caught. We also define and motivate the
Catching 'Just-in-Time' (JiTTest) Challenge, in which tests are generated
'just-in-time' to catch new faults before they land into production. We show
that any solution to Catching JiTTest generation can also be repurposed to
catch latent faults in legacy code. We enumerate possible outcomes for
hardening and catching tests and JiTTests, and discuss open research problems,
deployment options, and initial results from our work on automated LLM-based
hardening at Meta. This paper was written to accompany the keynote by the
authors at the ACM International Conference on the Foundations of Software
Engineering (FSE) 2025. Author order is alphabetical. The corresponding author
is Mark Harman.
| [
{
"version": "v1",
"created": "Wed, 23 Apr 2025 07:32:43 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 10:55:20 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Harman",
"Mark",
""
],
[
"O'Hearn",
"Peter",
""
],
[
"Sengupta",
"Shubho",
""
]
] |
2504.16686 | Simon Johann Klaus Lang | Simon J. K. Lang (1), Ignaz Eisele (1 and 2), Johannes Weber (1),
Alexandra Schewski (1 and 3), Emir Music (1), Alwin Maiwald (1), Martin Heigl
(1), Daniela Zahn (1), Zhen Luo (3), Lars Nebrich (1), Benedikt Schoof (3),
Thomas Mayer (1), Leonhard Sturm-Rogon (1), Wilfried Lerch (1), Rui N.
Pereira (1) and Christoph Kutter (1 and 2)((1) The Fraunhofer Institute for
Electronic Microsystems and Solid State Technologies EMFT, Munich, Germany,
(2) Center Integrated Sensor Systems (SENS), Universitaet der Bundeswehr
Muenchen, Munich, Germany, (3) Technical University of Munich, Munich,
Germany) | Wafer-Scale Characterization of Al/AlxOy/Al Josephson Junctions at Room
Temperature | 8 pages, 10 figures, 1 tables | null | null | null | quant-ph cond-mat.mtrl-sci physics.app-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Josephson junctions (JJs) are the key element of many devices operating at
cryogenic temperatures. Development of time-efficient wafer-scale JJ
characterization for process optimization and control of JJ fabrication is
essential. Such statistical characterization has to rely on room temperature
techniques since cryogenic measurements typically used for JJs are too time
consuming and unsuitable for wafer-scale characterization. In this work, we
show that from room temperature capacitance and current-voltage measurements,
with proper data analysis, we can independently obtain useful parameters of the
JJs on wafer-scale, like oxide thickness, tunnel coefficient, and interfacial
defect densities. Moreover, based on detailed analysis of current vs voltage
characteristics, different charge transport mechanisms across the junctions can
be distinguished. We exemplary demonstrate the worth of these methods by
studying junctions fabricated on 200 mm wafers with an industrially scale-able
concept based on subtractive processing using only CMOS compatible tools. From
these studies, we find that our subtractive fabrication approach yields
junctions with quite homogeneous average oxide thickness across the full
wafers, with a spread of less then 3$\,$%. The analysis also revealed a
variation of the tunnel coefficient with oxide thickness, pointing to a
stoichiometry gradient across the junctions' oxide width. Moreover, we
estimated relatively low interfacial defect densities in the range of 70 -
5000$\,$defects/cm$^2$ for our junctions and established that the density
increased with decreasing oxide thickness, indicating that the wet etching
process applied in the JJs fabrication for oxide thickness control leads to
formation of interfacial trap state
| [
{
"version": "v1",
"created": "Wed, 23 Apr 2025 13:15:01 GMT"
},
{
"version": "v2",
"created": "Wed, 7 May 2025 15:13:00 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 07:41:04 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Lang",
"Simon J. K.",
"",
"1 and 2"
],
[
"Eisele",
"Ignaz",
"",
"1 and 2"
],
[
"Weber",
"Johannes",
"",
"1 and 3"
],
[
"Schewski",
"Alexandra",
"",
"1 and 3"
],
[
"Music",
"Emir",
"",
"1 and 2"
],
[
"Maiwald",
"Alwin",
"",
"1 and 2"
],
[
"Heigl",
"Martin",
"",
"1 and 2"
],
[
"Zahn",
"Daniela",
"",
"1 and 2"
],
[
"Luo",
"Zhen",
"",
"1 and 2"
],
[
"Nebrich",
"Lars",
"",
"1 and 2"
],
[
"Schoof",
"Benedikt",
"",
"1 and 2"
],
[
"Mayer",
"Thomas",
"",
"1 and 2"
],
[
"Sturm-Rogon",
"Leonhard",
"",
"1 and 2"
],
[
"Lerch",
"Wilfried",
"",
"1 and 2"
],
[
"Pereira",
"Rui N.",
"",
"1 and 2"
],
[
"Kutter",
"Christoph",
"",
"1 and 2"
]
] |
2504.17363 | Fabien Baeriswyl | Fabien Baeriswyl (LPSM (UMR\_8001), DO), Olivier Wintenberger (LPSM
(UMR\_8001)) | Sample-Path Large Deviations for Functionals of Poisson Cluster
Processes | null | null | null | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We establish sample-path large deviation principles for the centered
cumulative functional of marked Poisson cluster processes in the Skorokhod
space equipped with the M1 topology, under joint regular variation assumptions
on the marks and the offspring distributions governing the propagation
mechanism. These findings can also be interpreted as hidden regular variation
of the cluster processes' functionals, extending the results in Dombry et al.
(2022) to cluster processes with heavy-tailed characteristics, including mixed
Binomial Poisson cluster processes and Hawkes processes. Notably, by
restricting to the adequate subspace of measures on D([0, 1], R+), and applying
the correct normalization and scaling to the paths of the centered cumulative
functional, the limit measure concentrates on paths with multiple large jumps.
| [
{
"version": "v1",
"created": "Thu, 24 Apr 2025 08:26:45 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 09:08:51 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 09:46:44 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Baeriswyl",
"Fabien",
"",
"LPSM"
],
[
"Wintenberger",
"Olivier",
"",
"LPSM"
]
] |
2504.17388 | Fangfang Du | Zhi-Guo Fan, Zhuo-Ya Bai, Qiu-Lin Tan, and Fang-Fang Du | Computing-heightened low-cost high-dimensional controlled-SUM gates | 16 pages, 7 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Qudit-based quantum gates offer several advantages over qubit-based
counterparts, such as higher information density, the ability to address more
complex problems, and richer quantum operations. In this paper, we present
three realistic protocols for implementing a 4$\times$4-dimensional (16D)
two-qudit controlled-SUM (CSUM) gate, where the 4D control qudit and 4D target
qudit are encoded in the polarization degree of freedom (DoF) and spatial DoF
of two photons, respectively. The first protocol is implemented exclusively
using linear optical elements without auxiliary resources, making it feasible
with current optical technologies and achieving an efficiency of 1/9. The
second protocol utilizes photon scattering by a microcavity-quantum-dot system,
enabling the 16D CSUM gate to operate deterministically without postselection.
The third protocol introduces an error-heralded mechanism based on the second
protocol, theoretically achieving unity fidelity. Moreover, all protocols
operate without ancillary photons, offering the advantages of compact circuits
and low cost while further promoting the development of high-dimensional
quantum computation.
| [
{
"version": "v1",
"created": "Thu, 24 Apr 2025 09:13:26 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 08:23:42 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Fan",
"Zhi-Guo",
""
],
[
"Bai",
"Zhuo-Ya",
""
],
[
"Tan",
"Qiu-Lin",
""
],
[
"Du",
"Fang-Fang",
""
]
] |
2504.17938 | Raza Ul Mustafa | Raza Ul Mustafa, Sesha Dassanayake, Noman Ashraf | Machine Learning-Based Prediction of Quality Shifts on Video Streaming
Over 5G | null | null | null | null | cs.MM cs.LG | http://creativecommons.org/licenses/by/4.0/ | The Quality of Experience (QoE) is the users satisfaction while streaming a
video session over an over-the-top (OTT) platform like YouTube. QoE of YouTube
reflects the smooth streaming session without any buffering and quality shift
events. One of the most important factors nowadays affecting QoE of YouTube is
frequent shifts from higher to lower resolutions and vice versa. These shifts
ensure a smooth streaming session; however, it might get a lower mean opinion
score. For instance, dropping from 1080p to 480p during a video can preserve
continuity but might reduce the viewers enjoyment. Over time, OTT platforms are
looking for alternative ways to boost user experience instead of relying on
traditional Quality of Service (QoS) metrics such as bandwidth, latency, and
throughput. As a result, we look into the relationship between quality shifting
in YouTube streaming sessions and the channel metrics RSRP, RSRQ, and SNR. Our
findings state that these channel metrics positively correlate with shifts.
Thus, in real-time, OTT can only rely on them to predict video streaming
sessions into lower- and higher-resolution categories, thus providing more
resources to improve user experience. Using traditional Machine Learning (ML)
classifiers, we achieved an accuracy of 77-percent, while using only RSRP,
RSRQ, and SNR. In the era of 5G and beyond, where ultra-reliable, low-latency
networks promise enhanced streaming capabilities, the proposed methodology can
be used to improve OTT services.
| [
{
"version": "v1",
"created": "Thu, 24 Apr 2025 21:00:43 GMT"
},
{
"version": "v2",
"created": "Tue, 6 May 2025 05:25:16 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 16:55:50 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Mustafa",
"Raza Ul",
""
],
[
"Dassanayake",
"Sesha",
""
],
[
"Ashraf",
"Noman",
""
]
] |
2504.18014 | Weixiang Sun | Weixiang Sun, Han Shen, Biwei Jiang, Xiaowei Liu | The Kinematic Signature of the Warp and Waves in the Milky Way Disk | 6 pages, 4 figures, 2 tables, accepted for publication in ApJ | null | null | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Using over 170,000 red clump (RC) stars selected from LAMOST and APOGEE, we
conduct a detailed analysis of the stellar $V_{Z}$ as a function of $L_{Z}$ (or
$R_{g}$) across different $\phi$ bins for various disk populations. The $V_{Z}$
of the whole RC sample stars exhibits a wave-like pattern superimposed on an
exponentially increasing trend, indicating the contribution from disk warp,
disk flare and disk waves. Our results across various populations suggest that
the thin disk is similar to the whole RC sample behavior, while the thick disk
displays a wave-like pattern superimposed on a linearly increasing trend,
meaning that the features of disk warp and waves are present in both thin and
thick disks, and the disk flare feature is only present in the thin disk. These
results indicate that the disk warp is potentially driven by secular processes
like disk perturbations from intergalactic magnetic fields and a misaligned
dark halo. The line-of-node of the disk warp of various populations displays a
slight difference, with $\phi_{0}$ = 5.68 $\pm$ 2.91 degree for the whole RC
sample stars, $\phi_{0}$ = 5.78 $\pm$ 2.89 degree for the thin disk stars, and
$\phi_{0}$ = 4.10 $\pm$ 3.43 degree for the thick disk stars.
| [
{
"version": "v1",
"created": "Fri, 25 Apr 2025 01:50:09 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 14:27:11 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Sun",
"Weixiang",
""
],
[
"Shen",
"Han",
""
],
[
"Jiang",
"Biwei",
""
],
[
"Liu",
"Xiaowei",
""
]
] |
2504.19617 | Luca Maxia | Luca Maxia, Dani\"el Boer, Jelle Bor | The impact of the TMD shape function on matching the transverse momentum
spectrum in $J/\psi$ production at the EIC | 17 pages, 8 figures, 1 table | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | The impact of the inclusion of TMD shape functions on the transverse momentum
spectrum in $J/\psi$ production at the EIC is investigated by considering the
matching of the TMD factorization description at low transverse momentum with
the collinear factorization description at high transverse momentum by means of
the inverse-error weighting method. Despite large uncertainties from scale
variations and the $J/\psi$ long-distance matrix elements, predictions for the
differential cross section and its $\cos(2\phi_\psi)$ modulation are obtained.
We find that physical constraints are satisfied in case a process-dependent
term is included for color octet production, but not in all cases when it is
excluded. These numerical results support the analytic calculations in
\cite{Boer:2023zit}. Future experimental data can thus test the validity of TMD
factorization in $J/\psi$ production and explore the presence of nontrivial
process-dependent effects in the soft-gluon resummation. This will be crucial
in the extraction of gluon unpolarized and linearly polarized TMD distributions
of the proton. In addition, we suggest how the sign of the latter can be
determined by investigating the presence of a node in the $\cos(2\phi_\psi)$
modulation as a function of transverse momentum.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 09:26:21 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 16:14:43 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Maxia",
"Luca",
""
],
[
"Boer",
"Daniël",
""
],
[
"Bor",
"Jelle",
""
]
] |
2504.19837 | Madhumita Kabiraj | Madhumita Kabiraj, Raka Dasgupta | Finite temperature phase diagram of the extended Bose-Hubbard model in
the presence of disorder | null | null | null | null | cond-mat.quant-gas cond-mat.stat-mech quant-ph | http://creativecommons.org/licenses/by/4.0/ | We study the finite and non-zero temperature phase diagram of the Extended
Bose-Hubbard Model for both pure and disordered systems. Such a system can be
experimentally realized by trapping ultracold Rydberg atoms in optical
lattices. By regulating the Rydberg excitation level and the lattice spacing,
the system can be engineered to effectively have (i) only the nearest-neighbor
interaction and (ii) both nearest-neighbor and next-nearest-neighbor
interactions. For both of these situations, we construct the mean-field phase
diagrams. It is found that the presence of a non-zero temperature significantly
changes the phase diagram because now there is a competition between quantum
and thermal fluctuations. We observe that conventional Mott insulator (MI) or
charge-density-wave (CDW) lobes vanish at higher temperatures. In a pure
system, they melt into a normal fluid (NF). In contrast, the insulating phases
that survive at high temperatures in the presence of disorder are the Bose
glass and the normal fluid. It is evident that the CDW lobes melt at a lower
temperature and the Mott lobes melt at higher temperatures. These transition
temperatures depend on the on-site and nearest-neighbor interaction strengths,
respectively. It is also found that, with the addition of disorder, the
insulating lobes are destroyed at a relatively lower temperature. The
mathematical framework that we present here is capable of treating long-range
interactions, disorder, and finite temperature simultaneously, and versatile
enough so that it can be extended to study different forms of disorder or
longer-range interactions.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:38:51 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 16:36:27 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Kabiraj",
"Madhumita",
""
],
[
"Dasgupta",
"Raka",
""
]
] |
2504.20404 | Gen Kimura | Gen Kimura, Aina Mayumi, Hiromichi Ohno, Jaeha Lee, Dariusz
Chru\'sci\'nski | Beyond Robertson-Schr\"odinger: A General Uncertainty Relation Unveiling
Hidden Noncommutative Trade-offs | 18 pages, Revised the proof of Lemma 2 | null | null | null | quant-ph cond-mat.stat-mech hep-th math-ph math.MP physics.ed-ph | http://creativecommons.org/licenses/by/4.0/ | We report a universal strengthening of the Robertson-Schr\''odinger
uncertainty relation, revealing a previously overlooked trade-off of genuinely
quantum origin, particularly as the state becomes more mixed. Remarkably, this
generalized bound supplements the standard commutator term and the covariance
term with an additional positive contribution that depends on the commutator of
observables. The relation also rigorously proves and extends a conjectured
uncertainty relation previously proposed in [Phys. Rev. A 110, 062215 (2024)].
For two-level quantum systems, the inequality becomes an exact equality for any
state and any pair of observables, establishing that the bound is tight in the
strongest possible sense.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 04:00:02 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Apr 2025 00:34:50 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 10:29:04 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Kimura",
"Gen",
""
],
[
"Mayumi",
"Aina",
""
],
[
"Ohno",
"Hiromichi",
""
],
[
"Lee",
"Jaeha",
""
],
[
"Chruściński",
"Dariusz",
""
]
] |
2504.20412 | Alex Mathai | Alex Mathai, Chenxi Huang, Suwei Ma, Jihwan Kim, Hailie Mitchell,
Aleksandr Nogikh, Petros Maniatis, Franjo Ivan\v{c}i\'c, Junfeng Yang,
Baishakhi Ray | CrashFixer: A crash resolution agent for the Linux kernel | null | null | null | null | cs.SE cs.AI cs.OS | http://creativecommons.org/licenses/by/4.0/ | Code large language models (LLMs) have shown impressive capabilities on a
multitude of software engineering tasks. In particular, they have demonstrated
remarkable utility in the task of code repair. However, common benchmarks used
to evaluate the performance of code LLMs are often limited to small-scale
settings. In this work, we build upon kGym, which shares a benchmark for
system-level Linux kernel bugs and a platform to run experiments on the Linux
kernel.
This paper introduces CrashFixer, the first LLM-based software repair agent
that is applicable to Linux kernel bugs. Inspired by the typical workflow of a
kernel developer, we identify the key capabilities an expert developer
leverages to resolve a kernel crash. Using this as our guide, we revisit the
kGym platform and identify key system improvements needed to practically run
LLM-based agents at the scale of the Linux kernel (50K files and 20M lines of
code). We implement these changes by extending kGym to create an improved
platform - called kGymSuite, which will be open-sourced. Finally, the paper
presents an evaluation of various repair strategies for such complex kernel
bugs and showcases the value of explicitly generating a hypothesis before
attempting to fix bugs in complex systems such as the Linux kernel. We also
evaluated CrashFixer's capabilities on still open bugs, and found at least two
patch suggestions considered plausible to resolve the reported bug.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 04:18:51 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 18:59:15 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Mathai",
"Alex",
""
],
[
"Huang",
"Chenxi",
""
],
[
"Ma",
"Suwei",
""
],
[
"Kim",
"Jihwan",
""
],
[
"Mitchell",
"Hailie",
""
],
[
"Nogikh",
"Aleksandr",
""
],
[
"Maniatis",
"Petros",
""
],
[
"Ivančić",
"Franjo",
""
],
[
"Yang",
"Junfeng",
""
],
[
"Ray",
"Baishakhi",
""
]
] |
2504.20533 | Anirban Dey | Anirban Dey, Davide Lonigro, Kazuya Yuasa, and Daniel Burgarth | Error bounds for the Floquet-Magnus expansion and their application to
the semiclassical quantum Rabi model | 13 pages, 4 figures, improved error bounds | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a general, nonperturbative method for deriving effective
Hamiltonians of arbitrary order for periodically driven systems, based on an
iterated integration by parts technique. The resulting family of effective
Hamiltonians reproduces the well-known Floquet-Magnus expansion, now enhanced
with explicit error bounds that quantify the distance between the exact and
approximate dynamics at each order, even in cases where the Floquet-Magnus
series fails to converge. We apply the method to the semiclassical Rabi model
and provide explicit error bounds for both the Bloch-Siegert Hamiltonian and
its third-order refinement. Our analysis shows that, while the rotating-wave
approximation more accurately captures the true dynamics than the Bloch-Siegert
Hamiltonian in most regimes, the third-order approximation ultimately
outperforms both.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 08:27:21 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 13:54:45 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Dey",
"Anirban",
""
],
[
"Lonigro",
"Davide",
""
],
[
"Yuasa",
"Kazuya",
""
],
[
"Burgarth",
"Daniel",
""
]
] |
2504.21227 | Omid Halimi Milani | Omid Halimi Milani, Amanda Nikho, Lauren Mills, Marouane Tliba, Ahmet
Enis Cetin, Mohammed H. Elnagar | Gradient Attention Map Based Verification of Deep Convolutional Neural
Networks with Application to X-ray Image Datasets | 13 pages, 7 figures, accepted at IEEE VLSI Test Symposium (VTS) 2025 | null | null | null | eess.IV cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning models have great potential in medical imaging, including
orthodontics and skeletal maturity assessment. However, applying a model to
data different from its training set can lead to unreliable predictions that
may impact patient care. To address this, we propose a comprehensive
verification framework that evaluates model suitability through multiple
complementary strategies. First, we introduce a Gradient Attention Map
(GAM)-based approach that analyzes attention patterns using Grad-CAM and
compares them via similarity metrics such as IoU, Dice Similarity, SSIM, Cosine
Similarity, Pearson Correlation, KL Divergence, and Wasserstein Distance.
Second, we extend verification to early convolutional feature maps, capturing
structural mis-alignments missed by attention alone. Finally, we incorporate an
additional garbage class into the classification model to explicitly reject
out-of-distribution inputs. Experimental results demonstrate that these
combined methods effectively identify unsuitable models and inputs, promoting
safer and more reliable deployment of deep learning in medical imaging.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 23:41:37 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 13:30:48 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Milani",
"Omid Halimi",
""
],
[
"Nikho",
"Amanda",
""
],
[
"Mills",
"Lauren",
""
],
[
"Tliba",
"Marouane",
""
],
[
"Cetin",
"Ahmet Enis",
""
],
[
"Elnagar",
"Mohammed H.",
""
]
] |
2505.00288 | Tri P. Nguyen | Tri P. Nguyen, Ilon Joseph, Mayya Tokman | Nystr\"om Type Exponential Integrators for Strongly Magnetized Charged
Particle Dynamics | null | null | null | null | physics.comp-ph cs.NA math.NA physics.plasm-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Calculating the dynamics of charged particles in electromagnetic fields (i.e.
the particle pushing problem) is one of the most computationally intensive
components of particle-in-cell (PIC) methods for plasma physics simulations.
This task is especially challenging when the plasma is strongly magnetized,
since in this case the particle motion consists of a wide range of temporal
scales from highly oscillatory fast gyromotion to slow macroscopic behavior and
the resulting numerical model is very stiff. Current state-of-the-art time
integrators used to simulate particle motion have limitations given the severe
numerical stiffness of the problem and more efficient methods are of interest.
Recently, exponential integrators have been proposed as a promising new
approach for these simulations and shown to offer computational advantages over
commonly used schemes. Exponential methods can solve linear problems exactly
and are $A$-stable. In this paper, the standard exponential algorithms
framework is extended to derive Nystr\"om-type exponential methods that
integrate the Newtonian equations of motion as a second-order differential
equation. Specific Nystr\"om-type schemes of second and third orders are
derived and applied to strongly magnetized particle pushing problems. Numerical
experiments are presented to demonstrate that the Nystr\"om-type exponential
integrators can provide significant improvement in computational efficiency
over the standard exponential methods.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 04:22:49 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 01:56:48 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Nguyen",
"Tri P.",
""
],
[
"Joseph",
"Ilon",
""
],
[
"Tokman",
"Mayya",
""
]
] |
2505.00476 | Michael Hite | Michael Hite | Simplified Fermionic Scattering State Preparation for the NISQ Era | null | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | With quantum computers steadily improving, large volume quantum scattering
simulations is getting very close. Yet, in the noisy intermediate scale quantum
(NISQ) era, we are limited to shallow circuits on the order of a thousand
layers. Thus, accurate and efficient state preparation methods are needed. We
introduce a simplified fermionic scattering state preparation method that
reduces circuit depth significantly by partially relaxing the fermionic
condition. Using the 1+1D transverse field Ising model as our test bed with
exact diagonalization and time evolving block decimation with matrix product
states, we show that our simplified states retain nearly all of the behavior of
the true fermionic state while being prepared on just a handful of qubits. We
also show promising early results on IonQ Forte 1.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 12:03:58 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 18:38:37 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Hite",
"Michael",
""
]
] |
2505.00887 | Xi Chen | Xi Chen, Yateng Tang, Jiarong Xu, Jiawei Zhang, Siwei Zhang, Sijia
Peng, Xuehao Zheng, Yun Xiong | Rethinking Time Encoding via Learnable Transformation Functions | 26 pages, 19 figures, 10 tables | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Effectively modeling time information and incorporating it into applications
or models involving chronologically occurring events is crucial. Real-world
scenarios often involve diverse and complex time patterns, which pose
significant challenges for time encoding methods. While previous methods focus
on capturing time patterns, many rely on specific inductive biases, such as
using trigonometric functions to model periodicity. This narrow focus on
single-pattern modeling makes them less effective in handling the diversity and
complexities of real-world time patterns. In this paper, we investigate to
improve the existing commonly used time encoding methods and introduce
Learnable Transformation-based Generalized Time Encoding (LeTE). We propose
using deep function learning techniques to parameterize non-linear
transformations in time encoding, making them learnable and capable of modeling
generalized time patterns, including diverse and complex temporal dynamics. By
enabling learnable transformations, LeTE encompasses previous methods as
specific cases and allows seamless integration into a wide range of tasks.
Through extensive experiments across diverse domains, we demonstrate the
versatility and effectiveness of LeTE.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 22:04:18 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 14:30:56 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Chen",
"Xi",
""
],
[
"Tang",
"Yateng",
""
],
[
"Xu",
"Jiarong",
""
],
[
"Zhang",
"Jiawei",
""
],
[
"Zhang",
"Siwei",
""
],
[
"Peng",
"Sijia",
""
],
[
"Zheng",
"Xuehao",
""
],
[
"Xiong",
"Yun",
""
]
] |
2505.00949 | Jiaqi Zeng | Akhiad Bercovich, Itay Levy, Izik Golan, Mohammad Dabbah, Ran
El-Yaniv, Omri Puny, Ido Galil, Zach Moshe, Tomer Ronen, Najeeb Nabwani, Ido
Shahaf, Oren Tropp, Ehud Karpas, Ran Zilberstein, Jiaqi Zeng, Soumye Singhal,
Alexander Bukharin, Yian Zhang, Tugrul Konuk, Gerald Shen, Ameya Sunil
Mahabaleshwarkar, Bilal Kartal, Yoshi Suhara, Olivier Delalleau, Zijia Chen,
Zhilin Wang, David Mosallanezhad, Adi Renduchintala, Haifeng Qian, Dima
Rekesh, Fei Jia, Somshubra Majumdar, Vahid Noroozi, Wasi Uddin Ahmad, Sean
Narenthiran, Aleksander Ficek, Mehrzad Samadi, Jocelyn Huang, Siddhartha
Jain, Igor Gitman, Ivan Moshkov, Wei Du, Shubham Toshniwal, George Armstrong,
Branislav Kisacanin, Matvei Novikov, Daria Gitman, Evelina Bakhturina, Jane
Polak Scowcroft, John Kamalu, Dan Su, Kezhi Kong, Markus Kliegl, Rabeeh
Karimi, Ying Lin, Sanjeev Satheesh, Jupinder Parmar, Pritam Gundecha, Brandon
Norick, Joseph Jennings, Shrimai Prabhumoye, Syeda Nahida Akter, Mostofa
Patwary, Abhinav Khattar, Deepak Narayanan, Roger Waleffe, Jimmy Zhang,
Bor-Yiing Su, Guyue Huang, Terry Kong, Parth Chadha, Sahil Jain, Christine
Harvey, Elad Segal, Jining Huang, Sergey Kashirsky, Robert McQueen, Izzy
Putterman, George Lam, Arun Venkatesan, Sherry Wu, Vinh Nguyen, Manoj Kilaru,
Andrew Wang, Anna Warno, Abhilash Somasamudramath, Sandip Bhaskar, Maka Dong,
Nave Assaf, Shahar Mor, Omer Ullman Argov, Scot Junkin, Oleksandr Romanenko,
Pedro Larroy, Monika Katariya, Marco Rovinelli, Viji Balas, Nicholas Edelman,
Anahita Bhiwandiwalla, Muthu Subramaniam, Smita Ithape, Karthik Ramamoorthy,
Yuting Wu, Suguna Varshini Velury, Omri Almog, Joyjit Daw, Denys Fridman,
Erick Galinkin, Michael Evans, Shaona Ghosh, Katherine Luna, Leon Derczynski,
Nikki Pope, Eileen Long, Seth Schneider, Guillermo Siman, Tomasz Grzegorzek,
Pablo Ribalta, Monika Katariya, Chris Alexiuk, Joey Conway, Trisha Saar, Ann
Guan, Krzysztof Pawelec, Shyamala Prayaga, Oleksii Kuchaiev, Boris Ginsburg,
Oluwatobi Olabiyi, Kari Briski, Jonathan Cohen, Bryan Catanzaro, Jonah Alben,
Yonatan Geifman, Eric Chung | Llama-Nemotron: Efficient Reasoning Models | null | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | We introduce the Llama-Nemotron series of models, an open family of
heterogeneous reasoning models that deliver exceptional reasoning capabilities,
inference efficiency, and an open license for enterprise use. The family comes
in three sizes -- Nano (8B), Super (49B), and Ultra (253B) -- and performs
competitively with state-of-the-art reasoning models such as DeepSeek-R1 while
offering superior inference throughput and memory efficiency. In this report,
we discuss the training procedure for these models, which entails using neural
architecture search from Llama 3 models for accelerated inference, knowledge
distillation, and continued pretraining, followed by a reasoning-focused
post-training stage consisting of two main parts: supervised fine-tuning and
large scale reinforcement learning. Llama-Nemotron models are the first
open-source models to support a dynamic reasoning toggle, allowing users to
switch between standard chat and reasoning modes during inference. To further
support open research and facilitate model development, we provide the
following resources: 1. We release the Llama-Nemotron reasoning models --
LN-Nano, LN-Super, and LN-Ultra -- under the commercially permissive NVIDIA
Open Model License Agreement. 2. We release the complete post-training dataset:
Llama-Nemotron-Post-Training-Dataset. 3. We also release our training
codebases: NeMo, NeMo-Aligner, and Megatron-LM.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 01:35:35 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 21:03:44 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 16:47:23 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Bercovich",
"Akhiad",
""
],
[
"Levy",
"Itay",
""
],
[
"Golan",
"Izik",
""
],
[
"Dabbah",
"Mohammad",
""
],
[
"El-Yaniv",
"Ran",
""
],
[
"Puny",
"Omri",
""
],
[
"Galil",
"Ido",
""
],
[
"Moshe",
"Zach",
""
],
[
"Ronen",
"Tomer",
""
],
[
"Nabwani",
"Najeeb",
""
],
[
"Shahaf",
"Ido",
""
],
[
"Tropp",
"Oren",
""
],
[
"Karpas",
"Ehud",
""
],
[
"Zilberstein",
"Ran",
""
],
[
"Zeng",
"Jiaqi",
""
],
[
"Singhal",
"Soumye",
""
],
[
"Bukharin",
"Alexander",
""
],
[
"Zhang",
"Yian",
""
],
[
"Konuk",
"Tugrul",
""
],
[
"Shen",
"Gerald",
""
],
[
"Mahabaleshwarkar",
"Ameya Sunil",
""
],
[
"Kartal",
"Bilal",
""
],
[
"Suhara",
"Yoshi",
""
],
[
"Delalleau",
"Olivier",
""
],
[
"Chen",
"Zijia",
""
],
[
"Wang",
"Zhilin",
""
],
[
"Mosallanezhad",
"David",
""
],
[
"Renduchintala",
"Adi",
""
],
[
"Qian",
"Haifeng",
""
],
[
"Rekesh",
"Dima",
""
],
[
"Jia",
"Fei",
""
],
[
"Majumdar",
"Somshubra",
""
],
[
"Noroozi",
"Vahid",
""
],
[
"Ahmad",
"Wasi Uddin",
""
],
[
"Narenthiran",
"Sean",
""
],
[
"Ficek",
"Aleksander",
""
],
[
"Samadi",
"Mehrzad",
""
],
[
"Huang",
"Jocelyn",
""
],
[
"Jain",
"Siddhartha",
""
],
[
"Gitman",
"Igor",
""
],
[
"Moshkov",
"Ivan",
""
],
[
"Du",
"Wei",
""
],
[
"Toshniwal",
"Shubham",
""
],
[
"Armstrong",
"George",
""
],
[
"Kisacanin",
"Branislav",
""
],
[
"Novikov",
"Matvei",
""
],
[
"Gitman",
"Daria",
""
],
[
"Bakhturina",
"Evelina",
""
],
[
"Scowcroft",
"Jane Polak",
""
],
[
"Kamalu",
"John",
""
],
[
"Su",
"Dan",
""
],
[
"Kong",
"Kezhi",
""
],
[
"Kliegl",
"Markus",
""
],
[
"Karimi",
"Rabeeh",
""
],
[
"Lin",
"Ying",
""
],
[
"Satheesh",
"Sanjeev",
""
],
[
"Parmar",
"Jupinder",
""
],
[
"Gundecha",
"Pritam",
""
],
[
"Norick",
"Brandon",
""
],
[
"Jennings",
"Joseph",
""
],
[
"Prabhumoye",
"Shrimai",
""
],
[
"Akter",
"Syeda Nahida",
""
],
[
"Patwary",
"Mostofa",
""
],
[
"Khattar",
"Abhinav",
""
],
[
"Narayanan",
"Deepak",
""
],
[
"Waleffe",
"Roger",
""
],
[
"Zhang",
"Jimmy",
""
],
[
"Su",
"Bor-Yiing",
""
],
[
"Huang",
"Guyue",
""
],
[
"Kong",
"Terry",
""
],
[
"Chadha",
"Parth",
""
],
[
"Jain",
"Sahil",
""
],
[
"Harvey",
"Christine",
""
],
[
"Segal",
"Elad",
""
],
[
"Huang",
"Jining",
""
],
[
"Kashirsky",
"Sergey",
""
],
[
"McQueen",
"Robert",
""
],
[
"Putterman",
"Izzy",
""
],
[
"Lam",
"George",
""
],
[
"Venkatesan",
"Arun",
""
],
[
"Wu",
"Sherry",
""
],
[
"Nguyen",
"Vinh",
""
],
[
"Kilaru",
"Manoj",
""
],
[
"Wang",
"Andrew",
""
],
[
"Warno",
"Anna",
""
],
[
"Somasamudramath",
"Abhilash",
""
],
[
"Bhaskar",
"Sandip",
""
],
[
"Dong",
"Maka",
""
],
[
"Assaf",
"Nave",
""
],
[
"Mor",
"Shahar",
""
],
[
"Argov",
"Omer Ullman",
""
],
[
"Junkin",
"Scot",
""
],
[
"Romanenko",
"Oleksandr",
""
],
[
"Larroy",
"Pedro",
""
],
[
"Katariya",
"Monika",
""
],
[
"Rovinelli",
"Marco",
""
],
[
"Balas",
"Viji",
""
],
[
"Edelman",
"Nicholas",
""
],
[
"Bhiwandiwalla",
"Anahita",
""
],
[
"Subramaniam",
"Muthu",
""
],
[
"Ithape",
"Smita",
""
],
[
"Ramamoorthy",
"Karthik",
""
],
[
"Wu",
"Yuting",
""
],
[
"Velury",
"Suguna Varshini",
""
],
[
"Almog",
"Omri",
""
],
[
"Daw",
"Joyjit",
""
],
[
"Fridman",
"Denys",
""
],
[
"Galinkin",
"Erick",
""
],
[
"Evans",
"Michael",
""
],
[
"Ghosh",
"Shaona",
""
],
[
"Luna",
"Katherine",
""
],
[
"Derczynski",
"Leon",
""
],
[
"Pope",
"Nikki",
""
],
[
"Long",
"Eileen",
""
],
[
"Schneider",
"Seth",
""
],
[
"Siman",
"Guillermo",
""
],
[
"Grzegorzek",
"Tomasz",
""
],
[
"Ribalta",
"Pablo",
""
],
[
"Katariya",
"Monika",
""
],
[
"Alexiuk",
"Chris",
""
],
[
"Conway",
"Joey",
""
],
[
"Saar",
"Trisha",
""
],
[
"Guan",
"Ann",
""
],
[
"Pawelec",
"Krzysztof",
""
],
[
"Prayaga",
"Shyamala",
""
],
[
"Kuchaiev",
"Oleksii",
""
],
[
"Ginsburg",
"Boris",
""
],
[
"Olabiyi",
"Oluwatobi",
""
],
[
"Briski",
"Kari",
""
],
[
"Cohen",
"Jonathan",
""
],
[
"Catanzaro",
"Bryan",
""
],
[
"Alben",
"Jonah",
""
],
[
"Geifman",
"Yonatan",
""
],
[
"Chung",
"Eric",
""
]
] |
2505.01058 | Sungwoo Kang | Sungwoo Kang | Coarse-grained graph architectures for all-atom force predictions | null | null | null | null | cond-mat.mtrl-sci cond-mat.soft | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce a machine-learning framework termed coarse-grained all-atom
force field (CGAA-FF), which incorporates coarse-grained message passing within
an all-atom force field using equivariant nature of graph models. The CGAA-FF
model employs grain embedding to encode atomistic coordinates into nodes
representing grains rather than individual atoms, enabling predictions of both
grain-level energies and atom-level forces. Tested on organic electrolytes,
CGAA-FF achieves root-mean-square errors of 4.96 meV atom-1 for energy and
0.201 eV A-1 for force predictions. CGAA-FF significantly reduces computational
costs, achieving about 22- and 14-fold improvements in simulation speed and
memory efficiency, respectively, compared to the all-atom potential
(SevenNet-0). Since this CGAA framework can be integrated into any equivariant
architecture, we believe this work opens the door to efficient all-atom
simulations of soft-matter systems.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 07:07:25 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 17:26:25 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Kang",
"Sungwoo",
""
]
] |
2505.01278 | \'Etienne Ligout | Luc Blanchet, David Langlois, Etienne Ligout | Innermost stable circular orbit of arbitrary-mass compact binaries at
fourth post-Newtonian order | 30 pages, 3 figures | null | null | null | gr-qc astro-ph.HE hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We compute by means of post-Newtonian (PN) methods the innermost stable
circular orbit (ISCO) of arbitrary-mass (in particular, comparable-mass)
compact binaries. Two methods are used with equivalent results: dynamical
perturbation of the conservative equations of motion in harmonic coordinates,
and dynamical perturbation of the conservative Hamiltonian in ADM coordinates.
The perturbation of the non-local tail term at 4PN order in both approaches is
carefully investigated. Our final gauge invariant result for the location of
the ISCO at 4PN order is close to the numerical value of the ISCO shift
computed by the gravitational self-force (GSF) approach in the small mass-ratio
limit, and is also in good agreement with the full numerical-relativity
calculation in the case of equal masses. The PN method followed here is
considered in standard Taylor-expanded form, without any resummation techniques
applied. As a complement, we also compute explicitly the gauge transformation
from harmonic coordinates to ADM coordinates up to 4PN order, including the
tail contribution therein.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 13:55:13 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 13:14:23 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Blanchet",
"Luc",
""
],
[
"Langlois",
"David",
""
],
[
"Ligout",
"Etienne",
""
]
] |
2505.01491 | Anjishnu Bose | Anjishnu Bose, Andrew Hardy, Naren Manjunath, Arun Paramekanti | Symmetry constrained field theories for chiral spin liquid to spin
crystal transitions | 24 pages, 3 figures, 4 tables | null | null | null | cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the spin rotationally invariant Kalmeyer-Laughlin chiral spin
liquid (CSL) in systems with broken time-reversal symmetry and explore symmetry
constraints on possible conventional spin crystal states accessible via a
direct transition. These constraints provide a framework to identify
topological invariants of the magnetically ordered state. We show that the
existence of a direct transition from a CSL requires a precise compatibility
condition between the topological invariants of the ordered state and the
anomaly of the CSL. The lattice symmetries also constrain the functional form
of the low-energy theory to describe these transitions. This allows us to
construct explicit Chern-Simons-matter field theories for the transition into a
class of noncoplanar orders identified as candidates directly accessible from
the CSL, including the octahedral spin crystal on the kagom\'e lattice, and the
tetrahedral order on the triangular and honeycomb lattice. These transitions
can either be described using coupled fractionalized $ \mathbb{CP}^1 $ theories
or fractionalized matrix principal chiral models. We also discuss extensions to
more general magnetic ordering transitions out of the CSL.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 21:34:14 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Bose",
"Anjishnu",
""
],
[
"Hardy",
"Andrew",
""
],
[
"Manjunath",
"Naren",
""
],
[
"Paramekanti",
"Arun",
""
]
] |
2505.01618 | Nolan Dey | Nolan Dey and Bin Claire Zhang and Lorenzo Noci and Mufan Li and Blake
Bordelon and Shane Bergsma and Cengiz Pehlevan and Boris Hanin and Joel
Hestness | Don't be lazy: CompleteP enables compute-efficient deep transformers | 10 main pages, 16 appendix pages, 13 figures | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | We study compute efficiency of LLM training when using different
parameterizations, i.e., rules for adjusting model and optimizer
hyperparameters (HPs) as model size changes. Some parameterizations fail to
transfer optimal base HPs (such as learning rate) across changes in model
depth, requiring practitioners to either re-tune these HPs as they scale up
(expensive), or accept sub-optimal training when re-tuning is prohibitive. Even
when they achieve HP transfer, we develop theory to show parameterizations may
still exist in the lazy learning regime where layers learn only features close
to their linearization, preventing effective use of depth and nonlinearity.
Finally, we identify and adopt the parameterization we call CompleteP that
achieves both depth-wise HP transfer and non-lazy learning in all layers.
CompleteP enables a wider range of model width/depth ratios to remain
compute-efficient, unlocking shapes better suited for different hardware
settings and operational contexts. Moreover, CompleteP enables 12-34% compute
efficiency improvements over the prior state-of-the-art.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 22:45:14 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 17:09:58 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Dey",
"Nolan",
""
],
[
"Zhang",
"Bin Claire",
""
],
[
"Noci",
"Lorenzo",
""
],
[
"Li",
"Mufan",
""
],
[
"Bordelon",
"Blake",
""
],
[
"Bergsma",
"Shane",
""
],
[
"Pehlevan",
"Cengiz",
""
],
[
"Hanin",
"Boris",
""
],
[
"Hestness",
"Joel",
""
]
] |
2505.01649 | Ali \"Ovg\"un Assoc.Prof.Dr. | Ali \"Ovg\"un, Reggie C. Pantig | Black Hole Solutions in Dark Photon Models with Higher Order Corrections | 7 pages, 3 figures. Reference list expanded | null | null | null | gr-qc | http://creativecommons.org/licenses/by/4.0/ | In this work, we derive new analytic, static, symmetric black hole solutions
in theories involving dark photons with minimal and higher-order magnetic
dipole interactions. Starting from the effective non-relativistic potential
between fermions mediated by a dark photon, we derive explicit corrections to
the Schwarzschild geometry induced by dark photon and spin-dependent terms.
These corrections alter the metric significantly at short distances, modifying
the horizon radius, Hawking temperature, photon sphere, and consequently, the
black hole shadow. Employing perturbative expansions, we provide analytic
expressions for the deviations from the Schwarzschild solution, highlighting an
exponential suppression controlled by the dark photon mass. Our results
demonstrate that higher-order magnetic dipole interactions produce distinctive
spin-dependent curvature terms, amplifying gravitational effects near the
horizon. These findings provide a theoretical foundation for future
phenomenological tests of dark photon models through gravitational wave
astronomy and black hole imaging, while highlighting dark photons' role as
mediators of dark matter interactions that can influence structure formation
and direct detection experiments.
| [
{
"version": "v1",
"created": "Sat, 3 May 2025 01:54:59 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 08:21:17 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Övgün",
"Ali",
""
],
[
"Pantig",
"Reggie C.",
""
]
] |
2505.01742 | Yu Mao | Yu Mao, Jingzong Li, Jun Wang, Hong Xu, Tei-Wei Kuo, Nan Guan, Chun
Jason Xue | Easz: An Agile Transformer-based Image Compression Framework for
Resource-constrained IoTs | null | null | null | null | eess.IV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Neural image compression, necessary in various machine-to-machine
communication scenarios, suffers from its heavy encode-decode structures and
inflexibility in switching between different compression levels. Consequently,
it raises significant challenges in applying the neural image compression to
edge devices that are developed for powerful servers with high computational
and storage capacities. We take a step to solve the challenges by proposing a
new transformer-based edge-compute-free image coding framework called Easz.
Easz shifts the computational overhead to the server, and hence avoids the
heavy encoding and model switching overhead on the edge. Easz utilizes a
patch-erase algorithm to selectively remove image contents using a conditional
uniform-based sampler. The erased pixels are reconstructed on the receiver side
through a transformer-based framework. To further reduce the computational
overhead on the receiver, we then introduce a lightweight transformer-based
reconstruction structure to reduce the reconstruction load on the receiver
side. Extensive evaluations conducted on a real-world testbed demonstrate
multiple advantages of Easz over existing compression approaches, in terms of
adaptability to different compression levels, computational efficiency, and
image reconstruction quality.
| [
{
"version": "v1",
"created": "Sat, 3 May 2025 08:39:19 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 13:02:05 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Mao",
"Yu",
""
],
[
"Li",
"Jingzong",
""
],
[
"Wang",
"Jun",
""
],
[
"Xu",
"Hong",
""
],
[
"Kuo",
"Tei-Wei",
""
],
[
"Guan",
"Nan",
""
],
[
"Xue",
"Chun Jason",
""
]
] |
2505.01775 | Fabiano Simas | D. Bazeia, A. S. Lob\~ao, Fabiano C. Simas | Kinks in generalized scalar field models and their scattering properties | version to appear in Physics Letters A | Phys. Lett. A 551 (2025) 130619 | 10.1016/j.physleta.2025.130619 | null | hep-th nlin.PS | http://creativecommons.org/licenses/by/4.0/ | This work investigates kink solutions in one-dimensional scalar field
theories. We begin with a review of the formalism used to obtain these
solutions, presenting the BPS formalism and linear stability analysis. Next, we
explore new models involving real scalar fields, generated by distinct
potentials, with a focus on the topological structures responsible for the
formation of kinks. Finally, we study collisions between the solutions obtained
in two distinct models, analyzing their dynamic implications.
| [
{
"version": "v1",
"created": "Sat, 3 May 2025 10:15:49 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Bazeia",
"D.",
""
],
[
"Lobão",
"A. S.",
""
],
[
"Simas",
"Fabiano C.",
""
]
] |
2505.01814 | Duan She | Duan She, Yi-Wei Qiu, Ze-Fang Jiang, and Defu Hou | Generalized relativistic second-order spin hydrodynamics from Zubarev's
non-equilibrium statistical operator | null | null | null | null | nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Inspired by the work in Ref.[1], which considers the additional second-order
contributions arising from nonlocal corrections due to two-point correlation
functions of tensors of different ranks at distinct spacetime points, we
similarly employ the nonequilibrium statistical operator method to extend this
framework to include spin degrees of freedom. In addition to obtaining
analogous extra second-order terms in the shear stress tensor, bulk viscous
pressure, and charge diffusion currents resulting from such contributions, we
further derive additional second-order terms originating from the same
mechanism in the charge diffusion currents, rotational stress tensor and the
boost heat vector. Furthermore, we express all transport coefficients
represented by two-point or three-point correlations in terms of retarded
Green's functions.
| [
{
"version": "v1",
"created": "Sat, 3 May 2025 13:17:13 GMT"
},
{
"version": "v2",
"created": "Wed, 7 May 2025 09:18:08 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 02:51:02 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"She",
"Duan",
""
],
[
"Qiu",
"Yi-Wei",
""
],
[
"Jiang",
"Ze-Fang",
""
],
[
"Hou",
"Defu",
""
]
] |
2505.01899 | Sam Miller | Sam K. Miller | On functoriality and the tensor product property in noncommutative
tensor-triangular geometry | v2: new section on partial functoriality of noncommutative Balmer
spectrum added, minor edits. 36 pages | null | null | null | math.CT math.AG math.KT math.RT | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Two pertinent questions for any support theory of a monoidal triangulated
category are whether it is functorial and if the tensor product property holds.
To this end, we consider the complete prime spectrum of an essentially small
monoidal triangulated category, which we show is universal among support data
satisfying the tensor product property, even if it is empty. We show the
complete prime spectrum is functorial and give a characterization of when a
monoidal exact functor induces a well-defined map on noncommutative Balmer
spectra. We give criteria for when induced maps on complete prime spectra are
injective or surjective, and determine the complete prime spectrum for crossed
product categories. Finally, we determine the universal functorial support
theory for monoidal triangulated categories coinciding with the Balmer spectrum
on braided monoidal triangulated categories.
| [
{
"version": "v1",
"created": "Sat, 3 May 2025 19:12:30 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 00:33:59 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Miller",
"Sam K.",
""
]
] |
2505.01956 | Ganesh Sapkota | Ganesh Sapkota, Sanjay Madria | SafeNav: Safe Path Navigation using Landmark Based Localization in a
GPS-denied Environment | 10 pages, conference paper. arXiv admin note: text overlap with
arXiv:2402.14280 | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | In battlefield environments, adversaries frequently disrupt GPS signals,
requiring alternative localization and navigation methods. Traditional
vision-based approaches like Simultaneous Localization and Mapping (SLAM) and
Visual Odometry (VO) involve complex sensor fusion and high computational
demand, whereas range-free methods like DV-HOP face accuracy and stability
challenges in sparse, dynamic networks. This paper proposes LanBLoc-BMM, a
navigation approach using landmark-based localization (LanBLoc) combined with a
battlefield-specific motion model (BMM) and Extended Kalman Filter (EKF). Its
performance is benchmarked against three state-of-the-art visual localization
algorithms integrated with BMM and Bayesian filters, evaluated on synthetic and
real-imitated trajectory datasets using metrics including Average Displacement
Error (ADE), Final Displacement Error (FDE), and a newly introduced Average
Weighted Risk Score (AWRS). LanBLoc-BMM (with EKF) demonstrates superior
performance in ADE, FDE, and AWRS on real-imitated datasets. Additionally, two
safe navigation methods, SafeNav-CHull and SafeNav-Centroid, are introduced by
integrating LanBLoc-BMM(EKF) with a novel Risk-Aware RRT* (RAw-RRT*) algorithm
for obstacle avoidance and risk exposure minimization. Simulation results in
battlefield scenarios indicate SafeNav-Centroid excels in accuracy, risk
exposure, and trajectory efficiency, while SafeNav-CHull provides superior
computational speed.
| [
{
"version": "v1",
"created": "Sun, 4 May 2025 01:40:31 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 21:56:50 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Sapkota",
"Ganesh",
""
],
[
"Madria",
"Sanjay",
""
]
] |
2505.02004 | Suyun Borjigin | Suyun Borjigin | Triple-identity Authentication: The Future of Secure Access | 10 pages, 2 figures, | null | null | null | cs.CR cs.ET cs.HC cs.SY eess.SY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In a typical authentication process, the local system verifies the user's
identity using a stored hash value generated by a cross-system hash algorithm.
This article shifts the research focus from traditional password encryption to
the establishment of gatekeeping mechanisms for effective interactions between
a system and the outside world. Here, we propose a triple-identity
authentication system to achieve this goal. Specifically, this local system
opens the inner structure of its hash algorithm to all user credentials,
including the login name, login password, and authentication password. When a
login credential is entered, the local system hashes it and then creates a
unique identifier using intermediate hash elements randomly selected from the
open algorithm. Importantly, this locally generated unique identifier (rather
than the stored hash produced by the open algorithm) is utilized to verify the
user's combined identity, which is generated by combining the entered
credential with the International Mobile Equipment Identity and the
International Mobile Subscriber Identity. The verification process is
implemented at each interaction point: the login name field, the login password
field, and the server's authentication point. Thus, within the context of this
triple-identity authentication system, we establish a robust gatekeeping
mechanism for system interactions, ultimately providing a level of security
that is equivalent to multi-factor authentication.
| [
{
"version": "v1",
"created": "Sun, 4 May 2025 06:18:59 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 11:32:58 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Borjigin",
"Suyun",
""
]
] |
2505.02126 | Zhihao Tang | Zhihao Tang, Shenghao Yang, Hongtao Zhang, and Mingbo Zhao | GarmentGS: Point-Cloud Guided Gaussian Splatting for High-Fidelity
Non-Watertight 3D Garment Reconstruction | Accepted by ICMR 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Traditional 3D garment creation requires extensive manual operations,
resulting in time and labor costs. Recently, 3D Gaussian Splatting has achieved
breakthrough progress in 3D scene reconstruction and rendering, attracting
widespread attention and opening new pathways for 3D garment reconstruction.
However, due to the unstructured and irregular nature of Gaussian primitives,
it is difficult to reconstruct high-fidelity, non-watertight 3D garments. In
this paper, we present GarmentGS, a dense point cloud-guided method that can
reconstruct high-fidelity garment surfaces with high geometric accuracy and
generate non-watertight, single-layer meshes. Our method introduces a fast
dense point cloud reconstruction module that can complete garment point cloud
reconstruction in 10 minutes, compared to traditional methods that require
several hours. Furthermore, we use dense point clouds to guide the movement,
flattening, and rotation of Gaussian primitives, enabling better distribution
on the garment surface to achieve superior rendering effects and geometric
accuracy. Through numerical and visual comparisons, our method achieves fast
training and real-time rendering while maintaining competitive quality.
| [
{
"version": "v1",
"created": "Sun, 4 May 2025 14:23:45 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 04:15:45 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Tang",
"Zhihao",
""
],
[
"Yang",
"Shenghao",
""
],
[
"Zhang",
"Hongtao",
""
],
[
"Zhao",
"Mingbo",
""
]
] |
2505.02291 | Tao Pang | H.J. Terry Suh, Tao Pang, Tong Zhao and Russ Tedrake | Dexterous Contact-Rich Manipulation via the Contact Trust Region | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | What is a good local description of contact dynamics for contact-rich
manipulation, and where can we trust this local description? While many
approaches often rely on the Taylor approximation of dynamics with an
ellipsoidal trust region, we argue that such approaches are fundamentally
inconsistent with the unilateral nature of contact. As a remedy, we present the
Contact Trust Region (CTR), which captures the unilateral nature of contact
while remaining efficient for computation. With CTR, we first develop a
Model-Predictive Control (MPC) algorithm capable of synthesizing local
contact-rich plans. Then, we extend this capability to plan globally by
stitching together local MPC plans, enabling efficient and dexterous
contact-rich manipulation. To verify the performance of our method, we perform
comprehensive evaluations, both in high-fidelity simulation and on hardware, on
two contact-rich systems: a planar IiwaBimanual system and a 3D AllegroHand
system. On both systems, our method offers a significantly lower-compute
alternative to existing RL-based approaches to contact-rich manipulation. In
particular, our Allegro in-hand manipulation policy, in the form of a roadmap,
takes fewer than 10 minutes to build offline on a standard laptop using just
its CPU, with online inference taking just a few seconds. Experiment data,
video and code are available at ctr.theaiinstitute.com.
| [
{
"version": "v1",
"created": "Sun, 4 May 2025 23:20:40 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 04:35:33 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Suh",
"H. J. Terry",
""
],
[
"Pang",
"Tao",
""
],
[
"Zhao",
"Tong",
""
],
[
"Tedrake",
"Russ",
""
]
] |
2505.02306 | Jihyung Park | Junfeng Jiao, Jihyung Park, Yiming Xu, Kristen Sussman, Lucy Atkinson | SafeMate: A Modular RAG-Based Agent for Context-Aware Emergency Guidance | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Despite the abundance of public safety documents and emergency protocols,
most individuals remain ill-equipped to interpret and act on such information
during crises. Traditional emergency decision support systems (EDSS) are
designed for professionals and rely heavily on static documents like PDFs or
SOPs, which are difficult for non-experts to navigate under stress. This gap
between institutional knowledge and public accessibility poses a critical
barrier to effective emergency preparedness and response. We introduce
SafeMate, a retrieval-augmented AI assistant that delivers accurate,
context-aware guidance to general users in both preparedness and active
emergency scenarios. Built on the Model Context Protocol (MCP), SafeMate
dynamically routes user queries to tools for document retrieval, checklist
generation, and structured summarization. It uses FAISS with cosine similarity
to identify relevant content from trusted sources.
| [
{
"version": "v1",
"created": "Mon, 5 May 2025 01:09:02 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 15:03:27 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 14:30:36 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Jiao",
"Junfeng",
""
],
[
"Park",
"Jihyung",
""
],
[
"Xu",
"Yiming",
""
],
[
"Sussman",
"Kristen",
""
],
[
"Atkinson",
"Lucy",
""
]
] |
2505.02378 | Zheming Gou | Vinay Kumar, Zheming Gou, Roger Ghanem | Uncertainty Quantification and Flow Dynamics in Rotating Detonation
Engines | null | null | null | null | physics.flu-dyn | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Rotating detonation engines (RDEs) are a critical technology for advancing
combustion engines, particularly in applications requiring high efficiency and
performance. Understanding the supersonic detonation structure and how various
parameters influence these phenomena is essential for optimizing RDE design. In
this study, we perform detailed simulations of detonations in an RDE and
analyze how the flow patterns are affected by key parameters associated with
the droplet arrangement within the engine. To further explore the system's
sensitivity, we apply polynomial chaos expansion to investigate the propagation
of uncertainties from input parameters to quantities of interest (QOIs).
Additionally, we develop a framework to accurately characterize the joint
distributions of QOIs with a limited number of simulations. Our findings
indicate that the strategic release of droplets may be crucial for sustaining
continuous detonation waves in the engine, and accurate representations of the
solution (e..g via high-order chaos expansions) are essential to accurately
capture the dependence between QOIs and input uncertainties. These insights
provide a quantitative foundation for further optimization of RDE designs.
| [
{
"version": "v1",
"created": "Mon, 5 May 2025 05:38:04 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 21:33:19 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Kumar",
"Vinay",
""
],
[
"Gou",
"Zheming",
""
],
[
"Ghanem",
"Roger",
""
]
] |
2505.02553 | Shunji Matsuura | Masanori Hanada, Shunji Matsuura, Emanuele Mendicelli and Enrico
Rinaldi | Exponential improvement in quantum simulations of bosons | null | null | null | RIKEN-iTHEMS-Report-25 | quant-ph hep-lat hep-th | http://creativecommons.org/licenses/by/4.0/ | Hamiltonian quantum simulation of bosons on digital quantum computers
requires truncating the Hilbert space to finite dimensions. The method of
truncation and the choice of basis states can significantly impact the
complexity of the quantum circuit required to simulate the system. For example,
a truncation in the Fock basis where each boson is encoded with a register of
$Q$ qubits, can result in an exponentially large number of Pauli strings
required to decompose the truncated Hamiltonian. This, in turn, can lead to an
exponential increase in $Q$ in the complexity of the quantum circuit. For
lattice quantum field theories such as Yang-Mills theory and QCD, several
Hamiltonian formulations and corresponding truncations have been put forward in
recent years. There is no exponential increase in $Q$ when resorting to the
orbifold lattice Hamiltonian, while we do not know how to remove the
exponential complexity in $Q$ in the commonly used Kogut-Susskind Hamiltonian.
Specifically, when using the orbifold lattice Hamiltonian, the continuum limit,
or, in other words, the removal of the ultraviolet energy cutoff, is obtained
with circuits whose resources scale like $Q$, while they scale like
$\mathcal{O}(\exp(Q))$ for the Kogut-Susskind Hamiltonian: this can be seen as
an exponential speed up in approaching the physical continuum limit for the
orbifold lattice Hamiltonian formulation. We show that the universal framework,
advocated by three of the authors (M.~H., S.~M., and E.~R.) and collaborators
provides a natural avenue to solve the exponential scaling of circuit
complexity with $Q$, and it is the reason why using the orbifold lattice
Hamiltonian is advantageous. We also point out that Hamiltonian formulations
based on a gauge-invariant Hilbert space require an exponential increase in the
resource requirement with respect to using an extended Hilbert space.
| [
{
"version": "v1",
"created": "Mon, 5 May 2025 10:48:52 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 16:49:42 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Hanada",
"Masanori",
""
],
[
"Matsuura",
"Shunji",
""
],
[
"Mendicelli",
"Emanuele",
""
],
[
"Rinaldi",
"Enrico",
""
]
] |
2505.02612 | Ivan Christov | Ivan P. Christov | Entanglement islands in 1D and 2D lattices with defects | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the spatial structure of quantum entanglement in one- and
two-dimensional lattice systems containing structural defects, using the
Time-Dependent Quantum Monte Carlo (TDQMC) method. By constructing reduced
density matrices from ensembles of guide waves, we resolve spatial variations
in both Coulomb-mediated entanglement and coherence without requiring full
many-body wavefunctions. This approach reveals localized regions, entanglement
islands, where quantum correlations are enhanced or suppressed due to the
presence of vacancies or interaction inhomogeneities. In 1D systems,
entanglement tends to concentrate near defects, while in 2D we observe
bridge-like and radially symmetric domains. Our results demonstrate that TDQMC
offers a scalable and physically transparent framework for real-space quantum
information analysis, with implications for quantum materials,
entanglement-based sensing, and coherent state engineering.
| [
{
"version": "v1",
"created": "Mon, 5 May 2025 12:30:12 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 12:56:05 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Christov",
"Ivan P.",
""
]
] |
2505.02997 | Ivy Pannier-G\"unther | Ivy Pannier-G\"unther and Andrew Kolmer Forbes and Pablo M. Poggi and
Ivan H. Deutsch | Long-Range Interacting Many-Body Systems in the Irrep Basis | 15+5 pages, 11+3 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spin models featuring infinite-range, homogeneous all-to-all interactions can
be efficiently described due to the existence of a symmetry-restricted Hilbert
subspace and an underlying classical phase space structure. However, when the
permutation invariance of the system is weakly broken, such as by long- but
finite-range interactions, these tools become mathematically invalid. Here we
propose to approximately describe these scenarios by considering additional
many-body subspaces according to the hierarchy of their coupling to the
symmetric subspace, defined by leveraging the structure of irreducible
representations (irreps) of the group $SU(2)$. We put forward a procedure,
dubbed "irrep distillation," which defines these additional subspaces to
minimize their dimension at each order of approximation. We discuss the
validity of our method in connection with the occurrence of quantum many-body
scars, benchmark its utility by analyzing the dynamical and equilibrium phase
transitions, outline its phenomenology, and compare its use-cases against other
approximations of long-range many-body systems.
| [
{
"version": "v1",
"created": "Mon, 5 May 2025 19:52:03 GMT"
},
{
"version": "v2",
"created": "Wed, 7 May 2025 23:12:36 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 15:18:42 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Pannier-Günther",
"Ivy",
""
],
[
"Forbes",
"Andrew Kolmer",
""
],
[
"Poggi",
"Pablo M.",
""
],
[
"Deutsch",
"Ivan H.",
""
]
] |
2505.03101 | Chunlei Liu | Chunlei Liu | Multiplication of polynomials over the binary field | null | null | null | null | math.NT cs.DS | http://creativecommons.org/licenses/by/4.0/ | Additive Fourier Transform is sdudied. A fast multiplication algorithm for
polynomials over the binary field is given. The bit complexity of the algorithm
is $O(n(log n)(\log\log n)^2)$.
| [
{
"version": "v1",
"created": "Tue, 6 May 2025 01:23:10 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 04:53:30 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Liu",
"Chunlei",
""
]
] |
2505.03185 | Jun Fang | Jun Fang, Yanuo Zhou, Ka I Chan, Jiajin Li, Zeyi Sun, Zhengnan Li,
Zicong Fu, Hongjing Piao, Haodong Xu, Yuanchun Shi, Yuntao Wang | Behavioral Sensing and Intervention Paradigm: A Review of Closed-Loop
Approaches for Ingestion Health | null | null | null | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | Ingestive behavior plays a critical role in health, yet many existing
interventions remain limited to static guidance or manual self-tracking. With
the increasing integration of sensors and perceptual computing, recent systems
have begun to support closed-loop interventions that dynamically sense user
behavior and provide feedback during or around ingestion episodes. In this
survey, we review 136 studies that leverage sensor-enabled or
interaction-mediated approaches to influence eating behavior. We propose a
behavioral closed-loop paradigm comprising three core components: target
behaviors, sensing modalities, and feedback strategies. A taxonomy of sensing
and intervention modalities is presented, organized along human- and
environment-based dimensions. Our analysis also examines evaluation methods and
design trends across different modality-behavior pairings. This review reveals
prevailing patterns and critical gaps, offering design insights for future
adaptive and context-aware ingestion health interventions.
| [
{
"version": "v1",
"created": "Tue, 6 May 2025 05:06:44 GMT"
},
{
"version": "v2",
"created": "Wed, 7 May 2025 11:37:12 GMT"
},
{
"version": "v3",
"created": "Wed, 14 May 2025 06:17:46 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Fang",
"Jun",
""
],
[
"Zhou",
"Yanuo",
""
],
[
"Chan",
"Ka I",
""
],
[
"Li",
"Jiajin",
""
],
[
"Sun",
"Zeyi",
""
],
[
"Li",
"Zhengnan",
""
],
[
"Fu",
"Zicong",
""
],
[
"Piao",
"Hongjing",
""
],
[
"Xu",
"Haodong",
""
],
[
"Shi",
"Yuanchun",
""
],
[
"Wang",
"Yuntao",
""
]
] |
2505.03508 | David O'Ryan Dr | David O'Ryan and Pablo G\'omez | Identifying Astrophysical Anomalies in 99.6 Million Cutouts from the
Hubble Legacy Archive Using AnomalyMatch | 15 pages - 12 figures - Submitted to A&A | null | null | null | astro-ph.IM astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Astronomical archives contain vast quantities of unexplored data that
potentially harbour rare and scientifically valuable cosmic phenomena. We
leverage new semi-supervised methods to extract such objects from the Hubble
Legacy Archive. We have systematically searched approximately 100 million image
cutouts from the entire Hubble Legacy Archive using the recently developed
AnomalyMatch method, which combines semi-supervised and active learning
techniques for the efficient detection of astrophysical anomalies. This
comprehensive search rapidly uncovered a multitude of astrophysical anomalies
presented here that significantly expand the inventory of known rare objects.
Among our discoveries are 138 new candidate gravitational lenses, 18 jellyfish
galaxies, and 417 mergers or interacting galaxies. The efficiency and accuracy
of our iterative detection strategy allows us to trawl the complete archive
within just 2-3 days, highlighting its potential for large-scale astronomical
surveys. We present a detailed overview of these newly identified objects,
discuss their astrophysical significance, and demonstrate the considerable
potential of AnomalyMatch to efficiently explore extensive astronomical
datasets, including, e.g., upcoming Euclid data releases.
| [
{
"version": "v1",
"created": "Tue, 6 May 2025 13:19:14 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 08:05:45 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"O'Ryan",
"David",
""
],
[
"Gómez",
"Pablo",
""
]
] |
2505.03700 | Burak O\u{g}uz | Burak O\u{g}uz | Topological Manipulations On $\mathbb{R}$ Symmetries Of Abelian Gauge
Theory | 40 pages, comments are welcome. v2: minor changes | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Performing topological manipulations is a fruitful way to understand global
aspects of Quantum Field Theory (QFT). Such modifications are controlled by the
notion of Topological QFT (TQFT) coupling across different codimensions.
Motivated by the recent developments involving non-compact TQFTs as the
Symmetry Topological Field Theory (SymTFT) for continuous symmetries, we
explore topological manipulations on global $\mathbb{R}$ symmetries in the
simple contexts of abelian gauge theories. We do so by turning on $\mathbb{R}$
background fields in these models, which are then inserted into a BF theory
with a conjugate field valued either in $\mathbb{R}$ or $U(1)$, depending on
the manipulation. When the conjugate field is $\mathbb{R}$-valued, it gives a
flat gauging prescription of the $\mathbb{R}$ global symmetry. For certain
cases, this gauging can be used to establish path-integral dualities, with
which one can construct topological duality defects and condensation defects.
The defects being defined by employing self-duality under $\mathbb{R}$ gauging,
they are anticipated to obey a Tambara-Yamagami like fusion algebra. Due to the
challenges arising from the non-compactness of $\mathbb{R}$, we do not
rigorously establish this expectation. For the TQFT coupling with $U(1)$ valued
conjugate field, we demonstrate a $2\pi \mathbb{Z} \subset \mathbb{R}$ subgroup
gauging, which we connect to the results obtained from SymTFT recently.
Executing this gauging in the 2d real scalar and the 4d $\mathbb{R}$ Maxwell,
we respectively obtain the 2d compact scalar and the 4d $U(1)$ Maxwell theory.
Finally, we give an exposition on the $\mathbb{R}$ gauging ideas applied to
$(-1)-$form and $(d-1)-$form $\mathbb{R}$ symmetries. In particular, using the
SymTFT picture, we identify the possible manipulations and realize them
explicitly in the context of $p$-form gauge theory in $d=p+1$ dimensions.
| [
{
"version": "v1",
"created": "Tue, 6 May 2025 17:19:42 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 17:37:03 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Oğuz",
"Burak",
""
]
] |
2505.03707 | Ido Kaminer | Offek Tziperman, David Nabben, Ron Ruimy, Jacob Holder, Ethan
Nussinson, Yiqi Fang, Alexey Gorlach, Daniel Kazenwadel, Aviv Karnieli, Ido
Kaminer, Peter Baum | Two-electron quantum walks can probe entanglement and decoherence in an
electron microscope | null | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | Classical physics is often a good approximation for quantum systems composed
of many interacting particles, although wavepacket dispersion and scattering
processes continuously induce delocalization and entanglement. According to
decoherence theory, an entangled ensemble can appear classical when only a
subset of all particles is observed. This emergence of macroscopic phenomena
from quantum interactions is, for example, relevant for phase transitions,
quantum thermalization, hydrodynamics, spin liquids, or time crystals. However,
entanglement and decoherence in free electrons have not yet been explored,
although the electron is a fundamental elementary particle with extraordinary
technological relevance. Here, we investigate the degree of coherence and
entanglement in a free-space electron gas in the beam of an ultrafast electron
microscope. We introduce a two-electron quantum walk that transforms the
quantum state into different bases for quantum state tomography of entangled or
partially entangled electron-electron pairs. We apply this novel diagnostic to
study quantum effects in short pulses of hundreds of electrons under strong
Coulomb correlation. We observe a high contrast interference in the
electron-electron correlations but no significant signs of electron-electron
entanglement which we explain by limited purity of the initial states and
decoherence effects from unmeasured reservoir electrons. The ability to
characterize quantum states of multiple free electrons may allow verification
of electron-electron entanglement for use in fundamental studies and quantum
electron microscopy.
| [
{
"version": "v1",
"created": "Tue, 6 May 2025 17:27:36 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 18:48:07 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Tziperman",
"Offek",
""
],
[
"Nabben",
"David",
""
],
[
"Ruimy",
"Ron",
""
],
[
"Holder",
"Jacob",
""
],
[
"Nussinson",
"Ethan",
""
],
[
"Fang",
"Yiqi",
""
],
[
"Gorlach",
"Alexey",
""
],
[
"Kazenwadel",
"Daniel",
""
],
[
"Karnieli",
"Aviv",
""
],
[
"Kaminer",
"Ido",
""
],
[
"Baum",
"Peter",
""
]
] |
2505.03898 | Ying Yuan | Shuqi Wang, Ying Yuan, Suyu Liu | ROSE: Randomized Optimal Selection Design for Dose Optimization | null | null | null | null | stat.ME | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The U.S. Food and Drug Administration (FDA) launched Project Optimus to shift
the objective of dose selection from the maximum tolerated dose to the optimal
biological dose (OBD), optimizing the benefit-risk tradeoff. One approach
recommended by the FDA's guidance is to conduct randomized trials comparing
multiple doses. In this paper, using the selection design framework (Simon et
al., 1985), we propose a randomized optimal selection (ROSE) design, which
minimizes sample size while ensuring the probability of correct selection of
the OBD at prespecified accuracy levels. The ROSE design is simple to
implement, involving a straightforward comparison of the difference in response
rates between two dose arms against a predetermined decision boundary. We
further consider a two-stage ROSE design that allows for early selection of the
OBD at the interim when there is sufficient evidence, further reducing the
sample size. Simulation studies demonstrate that the ROSE design exhibits
desirable operating characteristics in correctly identifying the OBD. A sample
size of 15 to 40 patients per dosage arm typically results in a percentage of
correct selection of the optimal dose ranging from 60% to 70%.
| [
{
"version": "v1",
"created": "Tue, 6 May 2025 18:09:26 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 18:37:04 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Wang",
"Shuqi",
""
],
[
"Yuan",
"Ying",
""
],
[
"Liu",
"Suyu",
""
]
] |
2505.04260 | Jessica Bo | Jessica Y. Bo, Tianyu Xu, Ishan Chatterjee, Katrina Passarella-Ward,
Achin Kulshrestha, and D Shin | Steerable Chatbots: Personalizing LLMs with Preference-Based Activation
Steering | null | null | null | null | cs.HC cs.AI | http://creativecommons.org/licenses/by/4.0/ | As large language models (LLMs) improve in their capacity to serve as
personal AI assistants, their ability to output uniquely tailored, personalized
responses that align with the soft preferences of their users is essential for
enhancing user satisfaction and retention. However, untrained lay users have
poor prompt specification abilities and often struggle with conveying their
latent preferences to AI assistants. To address this, we leverage activation
steering to guide LLMs to align with interpretable preference dimensions during
inference. In contrast to memory-based personalization methods that require
longer user history, steering is extremely lightweight and can be easily
controlled by the user via an linear strength factor. We embed steering into
three different interactive chatbot interfaces and conduct a within-subjects
user study (n=14) to investigate how end users prefer to personalize their
conversations. The results demonstrate the effectiveness of preference-based
steering for aligning real-world conversations with hidden user preferences,
and highlight further insights on how diverse values around control, usability,
and transparency lead users to prefer different interfaces.
| [
{
"version": "v1",
"created": "Wed, 7 May 2025 09:10:51 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 21:19:59 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Bo",
"Jessica Y.",
""
],
[
"Xu",
"Tianyu",
""
],
[
"Chatterjee",
"Ishan",
""
],
[
"Passarella-Ward",
"Katrina",
""
],
[
"Kulshrestha",
"Achin",
""
],
[
"Shin",
"D",
""
]
] |
2505.04377 | Valeria Goicoechea | Paola Bermolen, Valeria Goicoechea, Jos\'e R. Le\'on | Large Deviations and the Peano Phenomenon in Stochastic Differential
Equations with Homogeneous Drift | 23 pages | null | null | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider a diffusion equation in $\mathbb{R}^d$ with drift equal to the
gradient of a homogeneous potential of degree $1+\gamma$, with $0<\gamma<1$,
and local variance equal to $\varepsilon^2$ with $\varepsilon\to 0$. The
associated deterministic system for $\varepsilon=0$ has a potential that is not
a Lipschitz function at the origin. Therefore, an infinite number of solutions
exist, known as the Peano phenomenon. In this work, we study large deviations
of first and second order for the system with noise, generalizing previous
results for the particular potential $b(x)=x |x|^{\gamma-1}$. For the
first-order large deviations, we recover the rate function from the well-known
Freidlin-Wentzell work. For the second-order large deviation, we use a
refinement of Carmona-Simon bounds for the eigenfunctions of a Schr\"odinger
operator and prove that the exponential behavior of the process depends only on
the ground state of such an operator. Moreover, a refined study of the ground
state allows us to obtain the large deviation rate function explicitly and to
deduce that the family of diffusions converges to the set of extreme solutions
of the deterministic system.
| [
{
"version": "v1",
"created": "Wed, 7 May 2025 12:57:45 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 12:57:41 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Bermolen",
"Paola",
""
],
[
"Goicoechea",
"Valeria",
""
],
[
"León",
"José R.",
""
]
] |
2505.04489 | Francesco Coghi | Francesco Coghi, Romain Duvezin, John S. Wettlaufer | Accelerated escape dynamics in non-Markovian stochastic feedback | 17 pages, 6 figures, v2: fixed typos | null | null | null | cond-mat.stat-mech math.PR | http://creativecommons.org/licenses/by/4.0/ | We study the escape dynamics of a non-Markovian stochastic process with
time-averaged feedback, which we model as a one-dimensional Ornstein--Uhlenbeck
process wherein the drift is modified by the empirical mean of its trajectory.
This process maps onto a class of self-interacting diffusions. Using weak-noise
large deviation theory, we derive the most probable escape paths and quantify
their likelihood via the action functional. We compute the feedback-modified
Kramers rate and its inverse, which approximates the mean escape time, and show
that the feedback accelerates escape by storing finite-time fluctuations
thereby lowering the effective energy barrier, and shifting the optimal escape
time from infinite to finite. Although we identify alternative mechanisms, such
as slingshot and ballistic escape trajectories, we find that they remain
sub-optimal and hence do not accelerate escape. These results show how memory
feedback reshapes rare event statistics, thereby offering a mechanism to
potentially control escape dynamics.
| [
{
"version": "v1",
"created": "Wed, 7 May 2025 15:04:23 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 10:18:47 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Coghi",
"Francesco",
""
],
[
"Duvezin",
"Romain",
""
],
[
"Wettlaufer",
"John S.",
""
]
] |
2505.05079 | Dandan Chen | Dandan Chen, Rong Chen, Mengjie Zhao | Combinatorial proofs for partitions with repeated smallest part | null | null | null | null | math.CO math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, Andrews and El Bachraoui considered the number of integer
partitions whose smallest part is repeated exactly $k$ times and the remaining
parts are not repeated. They presented several interesting results and posed
questions regarding combinatorial proofs for these identities. In this paper,
we establish bijections to provide combinatorial proofs for these results.
| [
{
"version": "v1",
"created": "Thu, 8 May 2025 09:22:15 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 09:17:45 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Chen",
"Dandan",
""
],
[
"Chen",
"Rong",
""
],
[
"Zhao",
"Mengjie",
""
]
] |
2505.05084 | Xiaowei Zhu | Xiaowei Zhu, Yubing Ren, Yanan Cao, Xixun Lin, Fang Fang, Yangxi Li | Reliably Bounding False Positives: A Zero-Shot Machine-Generated Text
Detection Framework via Multiscaled Conformal Prediction | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | The rapid advancement of large language models has raised significant
concerns regarding their potential misuse by malicious actors. As a result,
developing effective detectors to mitigate these risks has become a critical
priority. However, most existing detection methods focus excessively on
detection accuracy, often neglecting the societal risks posed by high false
positive rates (FPRs). This paper addresses this issue by leveraging Conformal
Prediction (CP), which effectively constrains the upper bound of FPRs. While
directly applying CP constrains FPRs, it also leads to a significant reduction
in detection performance. To overcome this trade-off, this paper proposes a
Zero-Shot Machine-Generated Text Detection Framework via Multiscaled Conformal
Prediction (MCP), which both enforces the FPR constraint and improves detection
performance. This paper also introduces RealDet, a high-quality dataset that
spans a wide range of domains, ensuring realistic calibration and enabling
superior detection performance when combined with MCP. Empirical evaluations
demonstrate that MCP effectively constrains FPRs, significantly enhances
detection performance, and increases robustness against adversarial attacks
across multiple detectors and datasets.
| [
{
"version": "v1",
"created": "Thu, 8 May 2025 09:32:38 GMT"
},
{
"version": "v2",
"created": "Wed, 14 May 2025 04:38:15 GMT"
}
] | 2025-05-15T00:00:00 | [
[
"Zhu",
"Xiaowei",
""
],
[
"Ren",
"Yubing",
""
],
[
"Cao",
"Yanan",
""
],
[
"Lin",
"Xixun",
""
],
[
"Fang",
"Fang",
""
],
[
"Li",
"Yangxi",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.