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
2025-03-29 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2503.18305 | Guangsheng Ou | Guangsheng Ou and Mingwei Liu and Yuxuan Chen and Xueying Du and
Shengbo Wang and Zekai Zhang and Xin Peng and Zibin Zheng | Enhancing LLM-based Code Translation in Repository Context via Triple
Knowledge-Augmented | null | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) have behaved well in function-level code
translation without repository-level context. However, the performance of LLMs
in repository-level context code translation remains suboptimal due to complex
dependencies and context, hindering their adoption in industrial settings. In
this work, we propose a novel LLM-based code translation technique K-Trans,
which leverages triple knowledge augmentation to enhance LLM's translation
quality under repository context in real-world software development. First,
K-Trans constructs a translation knowledge base by extracting relevant
information from target-language codebases, the repository being translated,
and prior translation results. Second, for each function to be translated,
K-Trans retrieves relevant triple knowledge, including target-language code
samples, dependency usage examples, and successful translation function pairs,
serving as references to enhance LLM for translation. Third, K-Trans constructs
a knowledge-augmented translation prompt using the retrieved triple knowledge
and employs LLMs to generate the translated code while preserving repository
context. It further leverages LLMs for self-debugging, enhancing translation
correctness.
The experiments show that K-Trans substantially outperforms the baseline
adapted from previous work by 19.4%/40.2% relative improvement in pass@1 and
0.138 in CodeBLEU. It is important to note that the results also demonstrate
that each knowledge significantly contributes to K-Trans's effectiveness in
handling repository-level context code translation, with dependency usage
examples making the most notable contribution. Moreover, as the self-evolution
process progresses, the knowledge base continuously enhances the LLM's
performance across various aspects of the repository-level code translation.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 03:10:34 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 07:16:23 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ou",
"Guangsheng",
""
],
[
"Liu",
"Mingwei",
""
],
[
"Chen",
"Yuxuan",
""
],
[
"Du",
"Xueying",
""
],
[
"Wang",
"Shengbo",
""
],
[
"Zhang",
"Zekai",
""
],
[
"Peng",
"Xin",
""
],
[
"Zheng",
"Zibin",
""
]
] |
2503.18615 | Piotr Szewczak | Micha{\l} Pawlikowski, Piotr Szewczak, Lyubomyr Zdomskyy | Scales, products and the second row of the Scheepers diagram | 23 pp | null | null | null | math.CO math.GN | http://creativecommons.org/licenses/by/4.0/ | We consider products of sets of reals with a combinatorial structure based on
scales parameterized by filters. This kind of sets were intensively
investigated in products of spaces with combinatorial covering properties as
Hurewicz, Scheepers, Menger and Rothberger. We will complete this picture with
focusing on properties from the second row of the Scheepers diagram. In
particular we show that in the Miller model a product space of two
$\mathfrak{d}$-concentrated sets has a strong covering property
$\mathsf{S}_1(\Gamma,\Omega)$. We provide also counterexamples in products to
demonstrate limitations of used methods.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 12:15:24 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 15:21:54 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Pawlikowski",
"Michał",
""
],
[
"Szewczak",
"Piotr",
""
],
[
"Zdomskyy",
"Lyubomyr",
""
]
] |
2503.18684 | Ruiqi Zhu | Ruiqi Zhu, Endong Sun, Guanhe Huang, Oya Celiktutan | Efficient Continual Adaptation of Pretrained Robotic Policy with Online
Meta-Learned Adapters | Project link: https://ricky-zhu.github.io/OMLA/ | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | Continual adaptation is essential for general autonomous agents. For example,
a household robot pretrained with a repertoire of skills must still adapt to
unseen tasks specific to each household. Motivated by this, building upon
parameter-efficient fine-tuning in language models, prior works have explored
lightweight adapters to adapt pretrained policies, which can preserve learned
features from the pretraining phase and demonstrate good adaptation
performances. However, these approaches treat task learning separately,
limiting knowledge transfer between tasks. In this paper, we propose Online
Meta-Learned adapters (OMLA). Instead of applying adapters directly, OMLA can
facilitate knowledge transfer from previously learned tasks to current learning
tasks through a novel meta-learning objective. Extensive experiments in both
simulated and real-world environments demonstrate that OMLA can lead to better
adaptation performances compared to the baseline methods. The project link:
https://ricky-zhu.github.io/OMLA/.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 13:55:47 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:19:46 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhu",
"Ruiqi",
""
],
[
"Sun",
"Endong",
""
],
[
"Huang",
"Guanhe",
""
],
[
"Celiktutan",
"Oya",
""
]
] |
2503.18769 | Alan Dao | Alan Dao (Gia Tuan Dao), Dinh Bach Vu, Bui Quang Huy | AlphaSpace: Enabling Robotic Actions through Semantic Tokenization and
Symbolic Reasoning | null | null | null | null | cs.CL cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents AlphaSpace, a novel methodology designed to enhance the
spatial reasoning capabilities of language models for robotic manipulation in
3D Cartesian space. AlphaSpace employs a hierarchical semantics-based
tokenization strategy that encodes spatial information at both coarse and
fine-grained levels. Our approach represents objects with their attributes,
positions, and height information through structured tokens, enabling precise
spatial reasoning without relying on traditional vision-based embeddings. This
approach enables LLMs to accurately manipulate objects by positioning them at
specific (x, y, z) coordinates. Experimental results suggest that AlphaSpace
demonstrates promising potential for improving manipulation tasks, achieving a
total accuracy of 66.67%, compared to 37.5% for GPT-4o and 29.17% for Claude
3.5 Sonnet. These results demonstrate the potential of structured spatial
encoding for manipulation tasks and warrant further exploration.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 15:16:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:39:47 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Dao",
"Alan",
"",
"Gia Tuan Dao"
],
[
"Vu",
"Dinh Bach",
""
],
[
"Huy",
"Bui Quang",
""
]
] |
2503.18869 | Rui Xie | Rui Xie, Asad Ul Haq, Linsen Ma, Yunhua Fang, Zirak Burzin Engineer,
Liu Liu, Tong Zhang | Reimagining Memory Access for LLM Inference: Compression-Aware Memory
Controller Design | 9 pages, 11 figures | null | null | null | cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The efficiency of Large Language Model~(LLM) inference is often constrained
by substantial memory bandwidth and capacity demands. Existing techniques, such
as pruning, quantization, and mixture of experts/depth, reduce memory capacity
and/or bandwidth consumption at the cost of slight degradation in inference
quality. This paper introduces a design solution that further alleviates memory
bottlenecks by enhancing the on-chip memory controller in AI accelerators to
achieve two main objectives: (1) significantly reducing memory capacity and
bandwidth usage through lossless block compression~(e.g., LZ4 and ZSTD) of
model weights and key-value (KV) cache without compromising inference quality,
and (2) enabling memory bandwidth and energy consumption to scale
proportionally with context-dependent dynamic quantization. These goals are
accomplished by equipping the on-chip memory controller with mechanisms to
improve fine-grained bit-level accessibility and compressibility of weights and
KV cache through LLM-aware configuration of in-memory placement and
representation. Experimental results on publicly available LLMs demonstrate the
effectiveness of this approach, showing memory footprint reductions of 25.2\%
for model weights and 46.9\% for KV cache. In addition, our hardware prototype
at 4\,GHz and 32 lanes (7\,nm) achieves 8\,TB/s throughput with a modest area
overhead (under 3.8\,mm\(^2\)), which underscores the viability of LLM-aware
memory control as a key to efficient large-scale inference.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 16:44:32 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:48:14 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Xie",
"Rui",
""
],
[
"Haq",
"Asad Ul",
""
],
[
"Ma",
"Linsen",
""
],
[
"Fang",
"Yunhua",
""
],
[
"Engineer",
"Zirak Burzin",
""
],
[
"Liu",
"Liu",
""
],
[
"Zhang",
"Tong",
""
]
] |
2503.18876 | Chao Wu | Chao Wu | Singularity formation for the 1D model of EMHD | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we study the singularity formation phenomenon of the 1D model
of Electron Magnetohydrodynamics (EMHD). we will construct a solution whose
$C^3$-norm blows up in finite time. In the end, we will show that the solution
is in $C^{\infty}(\mathbb{R}\backslash \{0\})\cap C^{3,s}(\mathbb{R})\cap
H^3(\mathbb{R})$ and is not asymptotically self-similar.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 16:52:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 07:32:20 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wu",
"Chao",
""
]
] |
2503.18885 | Dirk Schuricht | Hosho Katsura, Dirk Schuricht | A brief note on the G$_2$ Affleck-Kennedy-Lieb-Tasaki chain | In memory of Ian Affleck | null | null | null | cond-mat.str-el math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | We consider the valence bond solid (VBS) state built of singlet pairs of
fundamental representations and projected onto adjoint representations of the
exceptional Lie group G$_2$. The two-point correlation function in the VBS
state is non-vanishing only for nearest neighbours, but possesses finite string
order. We construct a parent Hamiltonian for the VBS state, which constitutes
the G$_2$ analog of the famous AKLT chain.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 16:59:06 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:56:17 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Katsura",
"Hosho",
""
],
[
"Schuricht",
"Dirk",
""
]
] |
2503.18936 | Edward Medina Guerra | E.Medina-Guerra, I.V. Gornyi, and Yuval Gefen | Phase transitions in a non-Hermitian Su-Schrieffer-Heeger model via
Krylov spread complexity | null | null | null | null | cond-mat.str-el cond-mat.mes-hall quant-ph | http://creativecommons.org/licenses/by/4.0/ | We investigate phase transitions in a non-Hermitian Su-Schrieffer-Heeger
(SSH) with an imaginary chemical potential via Krylov spread complexity and
Krylov fidelity. The spread witnesses the PT-transition for the non-Hermitian
Bogoliubov vacuum of the SSH Hamiltonian, where the spectrum goes from purely
real to complex (oscillatory dynamics to damped oscillations). In addition, it
also witnesses the transition occurring in the PT-broken phase, where the
spectrum goes from complex to purely imaginary (damped oscillations to sheer
decay). For a purely imaginary spectrum, the Krylov spread fidelity, which
measures how the time-dependent spread reaches its stationary state value,
serves as a probe of previously undetected dynamical phase transitions.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 17:56:56 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Medina-Guerra",
"E.",
""
],
[
"Gornyi",
"I. V.",
""
],
[
"Gefen",
"Yuval",
""
]
] |
2503.18940 | Ye Tian | Ye Tian, Xin Xia, Yuxi Ren, Shanchuan Lin, Xing Wang, Xuefeng Xiao,
Yunhai Tong, Ling Yang, Bin Cui | Training-free Diffusion Acceleration with Bottleneck Sampling | Project Page: https://tyfeld.github.io/BottleneckSampling.github.io/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion models have demonstrated remarkable capabilities in visual content
generation but remain challenging to deploy due to their high computational
cost during inference. This computational burden primarily arises from the
quadratic complexity of self-attention with respect to image or video
resolution. While existing acceleration methods often compromise output quality
or necessitate costly retraining, we observe that most diffusion models are
pre-trained at lower resolutions, presenting an opportunity to exploit these
low-resolution priors for more efficient inference without degrading
performance. In this work, we introduce Bottleneck Sampling, a training-free
framework that leverages low-resolution priors to reduce computational overhead
while preserving output fidelity. Bottleneck Sampling follows a high-low-high
denoising workflow: it performs high-resolution denoising in the initial and
final stages while operating at lower resolutions in intermediate steps. To
mitigate aliasing and blurring artifacts, we further refine the resolution
transition points and adaptively shift the denoising timesteps at each stage.
We evaluate Bottleneck Sampling on both image and video generation tasks, where
extensive experiments demonstrate that it accelerates inference by up to
3$\times$ for image generation and 2.5$\times$ for video generation, all while
maintaining output quality comparable to the standard full-resolution sampling
process across multiple evaluation metrics.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 17:59:02 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:05:19 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Tian",
"Ye",
""
],
[
"Xia",
"Xin",
""
],
[
"Ren",
"Yuxi",
""
],
[
"Lin",
"Shanchuan",
""
],
[
"Wang",
"Xing",
""
],
[
"Xiao",
"Xuefeng",
""
],
[
"Tong",
"Yunhai",
""
],
[
"Yang",
"Ling",
""
],
[
"Cui",
"Bin",
""
]
] |
2503.18943 | Mingze Xu | Mingze Xu, Mingfei Gao, Shiyu Li, Jiasen Lu, Zhe Gan, Zhengfeng Lai,
Meng Cao, Kai Kang, Yinfei Yang, Afshin Dehghan | SlowFast-LLaVA-1.5: A Family of Token-Efficient Video Large Language
Models for Long-Form Video Understanding | Technical report | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce SlowFast-LLaVA-1.5 (abbreviated as SF-LLaVA-1.5), a family of
video large language models (LLMs) offering a token-efficient solution for
long-form video understanding. We incorporate the two-stream SlowFast mechanism
into a streamlined training pipeline, and perform joint video-image training on
a carefully curated data mixture of only publicly available datasets. Our
primary focus is on highly efficient model scales (1B and 3B), demonstrating
that even relatively small Video LLMs can achieve state-of-the-art performance
on video understanding, meeting the demand for mobile-friendly models.
Experimental results demonstrate that SF-LLaVA-1.5 achieves superior
performance on a wide range of video and image tasks, with robust results at
all model sizes (ranging from 1B to 7B). Notably, SF-LLaVA-1.5 achieves
state-of-the-art results in long-form video understanding (e.g., LongVideoBench
and MLVU) and excels at small scales across various video benchmarks.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 17:59:07 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:34:06 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Xu",
"Mingze",
""
],
[
"Gao",
"Mingfei",
""
],
[
"Li",
"Shiyu",
""
],
[
"Lu",
"Jiasen",
""
],
[
"Gan",
"Zhe",
""
],
[
"Lai",
"Zhengfeng",
""
],
[
"Cao",
"Meng",
""
],
[
"Kang",
"Kai",
""
],
[
"Yang",
"Yinfei",
""
],
[
"Dehghan",
"Afshin",
""
]
] |
2503.19013 | Joheen Chakraborty | Joheen Chakraborty, Erin Kara, Riccardo Arcodia, Johannes Buchner,
Margherita Giustini, Lorena Hern\'andez-Garc\'ia, Itai Linial, Megan
Masterson, Giovanni Miniutti, Andrew Mummery, Christos Panagiotou, Erwan
Quintin, Paula S\'anchez-S\'aez | Discovery of Quasi-periodic Eruptions in the Tidal Disruption Event and
Extreme Coronal Line Emitter AT2022upj: implications for the QPE/TDE fraction
and a connection to ECLEs | Accepted in ApJL | null | null | null | astro-ph.HE astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Quasi-periodic eruptions (QPEs) are recurring soft X-ray transients emerging
from the vicinity of supermassive black holes (SMBHs) in nearby, low-mass
galaxy nuclei; about ten QPE hosts have been identified thus far. Here we
report the \textit{NICER} discovery of QPEs in the optically-selected Tidal
Disruption Event (TDE) and Extreme Coronal Line Emitter (ECLE) AT2022upj,
exhibiting a large spread in recurrence times from 0.5-3.5 days, durations from
0.3-1 days, peak luminosities from $10^{42.5-43.0}$ erg s$^{-1}$, and erratic
flare profiles. A wealth of evidence now links at least some QPEs to the
newly-formed accretion flows emerging from TDEs; AT2022upj is the third QPE
reported in an optically discovered TDE. Marginalizing over the uncertain
distributions of QPE peak luminosity, recurrence time, delay after TDE peak,
and lifetime, we use the burgeoning sample to make a Bayesian estimate that the
fraction of optical TDEs resulting in QPEs within 5 years post-disruption is
$9^{+9}_{-5}$\%. Along with AT2019qiz, AT2022upj also marks the second of the
three optical TDE+X-ray QPEs showing coronal line emission, suggesting ECLEs
may represent a subset of TDEs particularly efficient at forming QPEs and/or
that sustained QPE X-ray emission contributes to coronal line emission in some
galaxy nuclei.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 18:00:00 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chakraborty",
"Joheen",
""
],
[
"Kara",
"Erin",
""
],
[
"Arcodia",
"Riccardo",
""
],
[
"Buchner",
"Johannes",
""
],
[
"Giustini",
"Margherita",
""
],
[
"Hernández-García",
"Lorena",
""
],
[
"Linial",
"Itai",
""
],
[
"Masterson",
"Megan",
""
],
[
"Miniutti",
"Giovanni",
""
],
[
"Mummery",
"Andrew",
""
],
[
"Panagiotou",
"Christos",
""
],
[
"Quintin",
"Erwan",
""
],
[
"Sánchez-Sáez",
"Paula",
""
]
] |
2503.19082 | Felix Burt | Felix Burt, Kuan-Cheng Chen, Kin K. Leung | A Multilevel Framework for Partitioning Quantum Circuits | 42 pages, 30 figures | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | Executing quantum algorithms over distributed quantum systems requires
quantum circuits to be divided into sub-circuits which communicate via
entanglement-based teleportation. Naively mapping circuits to qubits over
multiple quantum processing units (QPUs) results in large communication
overhead, increasing both execution time and noise. This can be minimised by
optimising the assignment of qubits to QPUs and the methods used for covering
non-local operations. Formulations which are general enough to capture the
spectrum of teleportation possibilities lead to complex problem instances which
can be difficult to solve effectively. This highlights a need to exploit the
wide range of heuristic techniques used in the graph partitioning literature.
This paper formalises and extends existing constructions for graphical quantum
circuit partitioning and designs a new objective function which captures
further possibilities for non-local operations via nested state teleportation.
We adapt the well-known Fiduccia-Mattheyses heuristic to the constraints and
problem objective and explore multilevel techniques which coarsen hypergraphs
and partition at multiple levels of granularity. We find that this reduces
runtime and improves solution quality of standard partitioning. We place these
techniques within a larger framework, through which we can extract full
distributed quantum circuits including teleportation instructions. We compare
the entanglement requirements and runtimes with state-of-the-art methods,
finding that we can achieve the lowest entanglement costs in most cases, while
always being close to the best performing method. We achieve an average
improvement of 35% over the next best performing method across a wide range of
circuits. We also find that our techniques can scale to much larger circuit
sizes than state-of-the-art methods, provided the number of partitions is not
too large.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 19:13:55 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:11:06 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Burt",
"Felix",
""
],
[
"Chen",
"Kuan-Cheng",
""
],
[
"Leung",
"Kin K.",
""
]
] |
2503.19176 | Yizhu Wen | Yizhu Wen, Ashwin Innuganti, Aaron Bien Ramos, Hanqing Guo, Qiben Yan | SoK: How Robust is Audio Watermarking in Generative AI models? | null | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by/4.0/ | Audio watermarking is increasingly used to verify the provenance of
AI-generated content, enabling applications such as detecting AI-generated
speech, protecting music IP, and defending against voice cloning. To be
effective, audio watermarks must resist removal attacks that distort signals to
evade detection. While many schemes claim robustness, these claims are
typically tested in isolation and against a limited set of attacks. A
systematic evaluation against diverse removal attacks is lacking, hindering
practical deployment. In this paper, we investigate whether recent watermarking
schemes that claim robustness can withstand a broad range of removal attacks.
First, we introduce a taxonomy covering 22 audio watermarking schemes. Next, we
summarize their underlying technologies and potential vulnerabilities. We then
present a large-scale empirical study to assess their robustness. To support
this, we build an evaluation framework encompassing 22 types of removal attacks
(109 configurations) including signal-level, physical-level, and AI-induced
distortions. We reproduce 9 watermarking schemes using open-source code,
identify 8 new highly effective attacks, and highlight 11 key findings that
expose the fundamental limitations of these methods across 3 public datasets.
Our results reveal that none of the surveyed schemes can withstand all tested
distortions. This evaluation offers a comprehensive view of how current
watermarking methods perform under real-world threats. Our demo and code are
available at https://sokaudiowm.github.io/.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 21:57:59 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 00:51:02 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wen",
"Yizhu",
""
],
[
"Innuganti",
"Ashwin",
""
],
[
"Ramos",
"Aaron Bien",
""
],
[
"Guo",
"Hanqing",
""
],
[
"Yan",
"Qiben",
""
]
] |
2503.19180 | Calvin Deutschbein | Calvin Deutschbein, Aristotle Stassinopoulos | "Test, Build, Deploy" -- A CI/CD Framework for Open-Source Hardware
Designs | 6 pages, 3 figures, under submission at ICECET'25 | null | null | null | cs.AR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Addressing TedX, Amber Huffman made an impassioned case that "none of us is
as smart as all of us" and that open-source hardware is the future. A major
contribution to software quality, open source and otherwise, on the software
side, is the systems design methodology of Continuous Integration and Delivery
(CI/CD), which we propose to systematically bring to hardware designs and their
specifications. To do so, we automatically generate specifications using
specification mining, "a machine learning approach to discovering formal
specifications" which dramatically impacted the ability of software engineers
to achieve quality, verification, and security. Yet applying the same
techniques to hardware is non-trivial. We present a technique for generalized,
continuous integration (CI) of hardware specification designs that continually
deploys (CD) a hardware specification. As a proof-of-concept, we demonstrate
Myrtha, a cloud-based, specification generator based on established hardware
and software quality tools.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 22:06:09 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 22:36:23 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Deutschbein",
"Calvin",
""
],
[
"Stassinopoulos",
"Aristotle",
""
]
] |
2503.19222 | Tomoki Ohsawa | Tomoki Ohsawa and Andrea Richaud | Small-Mass Asymptotics of Massive Point Vortex Dynamics in
Bose--Einstein Condensates | 27 pages, 13 figures | null | null | null | cond-mat.quant-gas math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We perform an asymptotic analysis of the massive point vortex dynamics in
Bose--Einstein condensates in the small-mass limit $\varepsilon \to 0$. We show
that the orthogonal projection of the massive dynamics to a certain subspace
$\mathcal{K}$ (called kinematic subspace) in the phase space yields the
standard massless vortex dynamics or the Kirchhoff equations. We show that the
massless dynamics is the zeroth-order approximation to the massive equation as
$\varepsilon \to 0$, and also derive a first-order correction to the
zeroth-order massless dynamics. The massive dynamics near $\mathcal{K}$ is well
approximated by the corresponding massless dynamics. In fact, our main result
proves that the massive dynamics starting in $\mathcal{K}$ stays
$O(\varepsilon)$-close to $\mathcal{K}$ for short time. On the other hand, the
massive dynamics starting outside $\mathcal{K}$ exhibits an oscillatory initial
layer due to its inertia and deviates significantly from the massless dynamics.
We also perform an asymptotic analysis with a rescaled time to derive the inner
approximation to capture the initial layer. The 0th-order inner equations are
exactly solvable, and the 1st-order inner equations can be solved by
quadrature. The combined inner approximation captures the initial oscillatory
layer of the massive dynamics.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 00:06:44 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 02:29:19 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ohsawa",
"Tomoki",
""
],
[
"Richaud",
"Andrea",
""
]
] |
2503.19285 | Yubo Li | Yubo Li, Xinyu Yao, Rema Padman | No Black Box Anymore: Demystifying Clinical Predictive Modeling with
Temporal-Feature Cross Attention Mechanism | 10 pages, 3 figures, submitted to AMIA 2025 | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite the outstanding performance of deep learning models in clinical
prediction tasks, explainability remains a significant challenge. Inspired by
transformer architectures, we introduce the Temporal-Feature Cross Attention
Mechanism (TFCAM), a novel deep learning framework designed to capture dynamic
interactions among clinical features across time, enhancing both predictive
accuracy and interpretability. In an experiment with 1,422 patients with
Chronic Kidney Disease, predicting progression to End-Stage Renal Disease,
TFCAM outperformed LSTM and RETAIN baselines, achieving an AUROC of 0.95 and an
F1-score of 0.69. Beyond performance gains, TFCAM provides multi-level
explainability by identifying critical temporal periods, ranking feature
importance, and quantifying how features influence each other across time
before affecting predictions. Our approach addresses the "black box"
limitations of deep learning in healthcare, offering clinicians transparent
insights into disease progression mechanisms while maintaining state-of-the-art
predictive performance.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 02:35:08 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 22:09:44 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Li",
"Yubo",
""
],
[
"Yao",
"Xinyu",
""
],
[
"Padman",
"Rema",
""
]
] |
2503.19316 | Zhiping Xiao | Zhiping Xiao, Xinyu Wang, Yifang Qin, Zijie Huang, Mason A. Porter,
Yizhou Sun | A Social Dynamical System for Twitter Analysis | will be submitted to a journal soon | null | null | null | cs.SI | http://creativecommons.org/licenses/by/4.0/ | Understanding the evolution of public opinion is crucial for informed
decision-making in various domains, particularly public affairs. The rapid
growth of social networks, such as Twitter (now rebranded as X), provides an
unprecedented opportunity to analyze public opinion at scale without relying on
traditional surveys. With the rise of deep learning, Graph Neural Networks
(GNNs) have shown great promise in modeling online opinion dynamics. Notably,
classical opinion dynamics models, such as DeGroot, can be reformulated within
a GNN framework.
We introduce Latent Social Dynamical System (LSDS), a novel framework for
modeling the latent dynamics of social media users' opinions based on textual
content. Since expressed opinions may not fully reflect underlying beliefs,
LSDS first encodes post content into latent representations. It then leverages
a GraphODE framework, using a GNN-based ODE function to predict future
opinions. A decoder subsequently utilizes these predicted latent opinions to
perform downstream tasks, such as interaction prediction, which serve as
benchmarks for model evaluation. Our framework is highly flexible, supporting
various opinion dynamic models as ODE functions, provided they can be adapted
into a GNN-based form. It also accommodates different encoder architectures and
is compatible with diverse downstream tasks.
To validate our approach, we constructed dynamic datasets from Twitter data.
Experimental results demonstrate the effectiveness of LSDS, highlighting its
potential for future applications. We plan to publicly release our dataset and
code upon the publication of this paper.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 03:25:07 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 20:17:10 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Xiao",
"Zhiping",
""
],
[
"Wang",
"Xinyu",
""
],
[
"Qin",
"Yifang",
""
],
[
"Huang",
"Zijie",
""
],
[
"Porter",
"Mason A.",
""
],
[
"Sun",
"Yizhou",
""
]
] |
2503.19352 | Tomoaki Ishiyama | Tomoaki Ishiyama, Francisco Prada, Anatoly A. Klypin | Evolution of clustering in cosmological models with time-varying dark
energy | 11 pages, 8 figures, submitted to PRD. HDF5 halo catalogs are
available on https://skun.iaa.csic.es/SUsimulations/DDE/ | null | null | null | astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Observations favor cosmological models with a time-varying dark energy
component. But how does dynamical dark energy (DDE) influence the growth of
structure in an expanding Universe? We investigate this question using
high-resolution $N$-body simulations based on a DDE cosmology constrained by
first-year DESI data (DESIY1$+$DDE), characterized by a 4% lower Hubble
constant ($H_0$) and 10% higher matter density ($\Omega_0$) than the
Planck-2018 $\Lambda$CDM model. We examine the impact on the matter power
spectrum, halo abundances, clustering, and Baryonic Acoustic Oscillations
(BAO). We find that DESIY1$+$DDE exhibits a 10% excess in power at small scales
and a 15% suppression at large scales, driven primarily by its higher
$\Omega_0$. This trend is reflected in the halo mass function: DESIY1$+$DDE
predicts up to 70% more massive halos at $z = 2$ and a 40% excess at $z = 0.3$.
Clustering analysis reveals a 3.71% shift of the BAO peak towards smaller
scales in DESIY1$+$DDE, consistent with its reduced sound horizon compared to
Planck18 Measurements of the BAO dilation parameter $\alpha$, using halo
samples with DESI-like tracer number densities across $0 < z < 1.5$, agree with
the expected DESIY1$+$DDE-to-Planck18 sound horizon ratio. After accounting for
cosmology-dependent distances, the simulation-based observational dilation
parameter closely matches DESI Y1 data. We find that the impact of DDE is
severely limited by current observational constraints, which strongly favor
cosmological models -- whether including DDE or not -- with a tightly
constrained parameter $\Omega_0h^2\approx 0.143$, within 1-2% uncertainty.
Indeed, our results demonstrate that variations in cosmological parameters,
particularly $\Omega_0$, have a greater influence on structure formation than
the DDE component alone.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 05:02:44 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 03:24:08 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ishiyama",
"Tomoaki",
""
],
[
"Prada",
"Francisco",
""
],
[
"Klypin",
"Anatoly A.",
""
]
] |
2503.19388 | Xian Gong | Paul X. McCarthy, Xian Gong, Marieth Coetzer, Marian-Andrei Rizoiu,
Margaret L. Kern, John A. Johnson, Richard Holden, Fabian Braesemann | The economics of global personality diversity | null | null | null | null | econ.GN q-fin.EC | http://creativecommons.org/licenses/by/4.0/ | This study explores the relationship between personality diversity and
national economic performance, introducing the Global Personality Diversity
Index (${\Psi}$-GPDI) as a novel metric. Leveraging a dataset of 760,242
individuals across 135 countries, we quantify within-country diversity based on
the Big Five personality traits. Our findings reveal that personality diversity
accounts for 19.9% of the variance in GDP per capita and provides an additional
2.8% explanatory power beyond institutional quality and immigration,
underscoring its unique contribution to economic vitality. Through multi-factor
analysis, we demonstrate how personality diversity complements existing
economic frameworks, offering actionable insights for policymakers seeking to
enhance innovation, productivity, and resilience. This research positions
psychological diversity as a critical yet under explored factor in driving
economic growth, bridging the fields of psychology and economics.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 06:44:49 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 03:27:53 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 02:53:41 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"McCarthy",
"Paul X.",
""
],
[
"Gong",
"Xian",
""
],
[
"Coetzer",
"Marieth",
""
],
[
"Rizoiu",
"Marian-Andrei",
""
],
[
"Kern",
"Margaret L.",
""
],
[
"Johnson",
"John A.",
""
],
[
"Holden",
"Richard",
""
],
[
"Braesemann",
"Fabian",
""
]
] |
2503.19396 | Tanmoy Bhowmik | Arunangshu Bora, Anirban Mandal, Shreya Mittal, Mihir N Pandey, Rohan
Rana, Harsh Saxena, Priyajit Jana, Rijul Dhumane, Tanmoy Bhowmik and Martha
Constantinou | The Energy Spectrum of Kaon from Lattice QCD | 6 Pages, 3 Figures, 1 Table | null | null | null | hep-lat | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This study presents the analysis of data related to the two-point function of
kaon generated from lattice QCD simulations. Using gauge configurations of
twisted-mass fermions, we obtain the correlation functions for 6 values of
momentum for the kaon between 0 and 2GeV (both inclusive), we use statistical
techniques such as jackknife resampling to derive the energy of the particle.
The lattice results are compared to the continuum dispersion relation for the
particle, to assess systematic uncertainties in the lattice data. We establish
consistency with theory by comparing our results with the theoretical
predictions.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 07:05:07 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:57:52 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Bora",
"Arunangshu",
""
],
[
"Mandal",
"Anirban",
""
],
[
"Mittal",
"Shreya",
""
],
[
"Pandey",
"Mihir N",
""
],
[
"Rana",
"Rohan",
""
],
[
"Saxena",
"Harsh",
""
],
[
"Jana",
"Priyajit",
""
],
[
"Dhumane",
"Rijul",
""
],
[
"Bhowmik",
"Tanmoy",
""
],
[
"Constantinou",
"Martha",
""
]
] |
2503.19470 | Mingyang Chen | Mingyang Chen, Tianpeng Li, Haoze Sun, Yijie Zhou, Chenzheng Zhu,
Haofen Wang, Jeff Z. Pan, Wen Zhang, Huajun Chen, Fan Yang, Zenan Zhou,
Weipeng Chen | ReSearch: Learning to Reason with Search for LLMs via Reinforcement
Learning | Work in progress | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have shown remarkable capabilities in reasoning,
exemplified by the success of OpenAI-o1 and DeepSeek-R1. However, integrating
reasoning with external search processes remains challenging, especially for
complex multi-hop questions requiring multiple retrieval steps. We propose
ReSearch, a novel framework that trains LLMs to Reason with Search via
reinforcement learning without using any supervised data on reasoning steps.
Our approach treats search operations as integral components of the reasoning
chain, where when and how to perform searches is guided by text-based thinking,
and search results subsequently influence further reasoning. We train ReSearch
on Qwen2.5-7B(-Instruct) and Qwen2.5-32B(-Instruct) models and conduct
extensive experiments. Despite being trained on only one dataset, our models
demonstrate strong generalizability across various benchmarks. Analysis reveals
that ReSearch naturally elicits advanced reasoning capabilities such as
reflection and self-correction during the reinforcement learning process.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 09:00:58 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 05:56:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Mingyang",
""
],
[
"Li",
"Tianpeng",
""
],
[
"Sun",
"Haoze",
""
],
[
"Zhou",
"Yijie",
""
],
[
"Zhu",
"Chenzheng",
""
],
[
"Wang",
"Haofen",
""
],
[
"Pan",
"Jeff Z.",
""
],
[
"Zhang",
"Wen",
""
],
[
"Chen",
"Huajun",
""
],
[
"Yang",
"Fan",
""
],
[
"Zhou",
"Zenan",
""
],
[
"Chen",
"Weipeng",
""
]
] |
2503.19614 | Pu Liu | Pu Liu, Chaoxi Cui, Lei Li, Runze Li, Dong-Hui Xu, and Zhi-Ming Yu | Floquet control of topological phases and Hall effects in Z2 nodal line
semimetals | 9 pages, 5 figures | null | null | null | cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dynamic control of topological properties in materials is central to modern
condensed matter physics, and Floquet engineering, utilizing periodic light
fields, provides a promising avenue. Here, we use Floquet theory to
theoretically study the topological response of a Z2 nodal line semimetal
(NLSM) when driven by circularly polarized light (CPL). We demonstrate that the
direction of CPL irradiation critically dictates the resulting topological
phase transitions. Specifically, when light is incident perpendicular to the
nodal line plane, increasing the light amplitude induces two successive
topological phase transitions: first, from the Z2 NLSM to a vortex NLSM, a rare
and intriguing topological state; and second, a transition from the vortex NLSM
to a semimetal with a pair of Weyl points (WPs). In stark contrast, irradiation
along other directions directly transforms the Z2 nodal line into a pair of
WPs. We further investigate the transport properties of the Floquet Z2 NLSM,
focusing on the anomalous and planar Hall effects. The anomalous Hall effect
exhibits a direction-dependent amplitude variation, deviating from conventional
two-band NLSM behavior. Importantly, we reveal a significant and tunable planar
Hall effect, a phenomenon largely unexplored in Floquet topological materials,
which is highly sensitive to both light amplitude and direction. Our findings
not only present a novel route to realize the vortex NLSM, but also establish
an efficient way to control Hall transport phenomena in Z2 NLSMs.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 12:58:55 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:40:34 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Liu",
"Pu",
""
],
[
"Cui",
"Chaoxi",
""
],
[
"Li",
"Lei",
""
],
[
"Li",
"Runze",
""
],
[
"Xu",
"Dong-Hui",
""
],
[
"Yu",
"Zhi-Ming",
""
]
] |
2503.19654 | Mehdi Moshtaghi | Mehdi Moshtaghi, Siavash H. Khajavi, Joni Pajarinen | RGB-Th-Bench: A Dense benchmark for Visual-Thermal Understanding of
Vision Language Models | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | We introduce RGB-Th-Bench, the first benchmark designed to evaluate the
ability of Vision-Language Models (VLMs) to comprehend RGB-Thermal image pairs.
While VLMs have demonstrated remarkable progress in visual reasoning and
multimodal understanding, their evaluation has been predominantly limited to
RGB-based benchmarks, leaving a critical gap in assessing their capabilities in
infrared vision tasks. Existing visible-infrared datasets are either
task-specific or lack high-quality annotations necessary for rigorous model
evaluation. To address these limitations, RGB-Th-Bench provides a comprehensive
evaluation framework covering 14 distinct skill dimensions, with a total of
1,600+ expert-annotated Yes/No questions. The benchmark employs two accuracy
metrics: a standard question-level accuracy and a stricter skill-level
accuracy, which evaluates model robustness across multiple questions within
each skill dimension. This design ensures a thorough assessment of model
performance, including resilience to adversarial and hallucinated responses. We
conduct extensive evaluations on 19 state-of-the-art VLMs, revealing
significant performance gaps in RGB-Thermal understanding. Our results show
that even the strongest models struggle with thermal image comprehension, with
performance heavily constrained by their RGB-based capabilities. Additionally,
the lack of large-scale application-specific and expert-annotated
thermal-caption-pair datasets in pre-training is an important reason of the
observed performance gap. RGB-Th-Bench highlights the urgent need for further
advancements in multimodal learning to bridge the gap between visible and
thermal image understanding. The dataset is available through this link, and
the evaluation code will also be made publicly available.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 13:43:47 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 10:11:22 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Moshtaghi",
"Mehdi",
""
],
[
"Khajavi",
"Siavash H.",
""
],
[
"Pajarinen",
"Joni",
""
]
] |
2503.19690 | Ran Yu | Bo Leng, Ran Yu, Wei Han, Lu Xiong, Zhuoren Li and Hailong Huang | Risk-Aware Reinforcement Learning for Autonomous Driving: Improving
Safety When Driving through Intersection | 11 pages, 10 figures | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Applying reinforcement learning to autonomous driving has garnered widespread
attention. However, classical reinforcement learning methods optimize policies
by maximizing expected rewards but lack sufficient safety considerations, often
putting agents in hazardous situations. This paper proposes a risk-aware
reinforcement learning approach for autonomous driving to improve the safety
performance when crossing the intersection. Safe critics are constructed to
evaluate driving risk and work in conjunction with the reward critic to update
the actor. Based on this, a Lagrangian relaxation method and cyclic gradient
iteration are combined to project actions into a feasible safe region.
Furthermore, a Multi-hop and Multi-layer perception (MLP) mixed Attention
Mechanism (MMAM) is incorporated into the actor-critic network, enabling the
policy to adapt to dynamic traffic and overcome permutation sensitivity
challenges. This allows the policy to focus more effectively on surrounding
potential risks while enhancing the identification of passing opportunities.
Simulation tests are conducted on different tasks at unsignalized
intersections. The results show that the proposed approach effectively reduces
collision rates and improves crossing efficiency in comparison to baseline
algorithms. Additionally, our ablation experiments demonstrate the benefits of
incorporating risk-awareness and MMAM into RL.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 14:17:15 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:02:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Leng",
"Bo",
""
],
[
"Yu",
"Ran",
""
],
[
"Han",
"Wei",
""
],
[
"Xiong",
"Lu",
""
],
[
"Li",
"Zhuoren",
""
],
[
"Huang",
"Hailong",
""
]
] |
2503.19721 | Chengjie Ge | Chengjie Ge, Xueyang Fu, Peng He, Kunyu Wang, Chengzhi Cao, Zheng-Jun
Zha | EventMamba: Enhancing Spatio-Temporal Locality with State Space Models
for Event-Based Video Reconstruction | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Leveraging its robust linear global modeling capability, Mamba has notably
excelled in computer vision. Despite its success, existing Mamba-based vision
models have overlooked the nuances of event-driven tasks, especially in video
reconstruction. Event-based video reconstruction (EBVR) demands spatial
translation invariance and close attention to local event relationships in the
spatio-temporal domain. Unfortunately, conventional Mamba algorithms apply
static window partitions and standard reshape scanning methods, leading to
significant losses in local connectivity. To overcome these limitations, we
introduce EventMamba--a specialized model designed for EBVR tasks. EventMamba
innovates by incorporating random window offset (RWO) in the spatial domain,
moving away from the restrictive fixed partitioning. Additionally, it features
a new consistent traversal serialization approach in the spatio-temporal
domain, which maintains the proximity of adjacent events both spatially and
temporally. These enhancements enable EventMamba to retain Mamba's robust
modeling capabilities while significantly preserving the spatio-temporal
locality of event data. Comprehensive testing on multiple datasets shows that
EventMamba markedly enhances video reconstruction, drastically improving
computation speed while delivering superior visual quality compared to
Transformer-based methods.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 14:46:45 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:41:35 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ge",
"Chengjie",
""
],
[
"Fu",
"Xueyang",
""
],
[
"He",
"Peng",
""
],
[
"Wang",
"Kunyu",
""
],
[
"Cao",
"Chengzhi",
""
],
[
"Zha",
"Zheng-Jun",
""
]
] |
2503.19722 | Yifei Xiong | Yifei Xiong, Ning Jiang, Zhen Pan, Lei Hao, and Zhenzhen Li | Extended Emission Line Region in a Post-Starburst Galaxy Hosting TDE
AT2019qiz and QPE | 17 pages, 11 figures | null | null | null | astro-ph.GA astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a comprehensive analysis of the extended emission line region
(EELR) in the host galaxy of the tidal disruption event (TDE) AT2019qiz,
utilizing VLT/MUSE integral-field spectroscopy. The high spatial-resolution
data reveal a bi-conical emission structure approximately $3.7~\mathrm{kpc}$ in
scale within the galactic center, characterized by a prominent [OIII] line in
the nucleus and significant [NII] line emission extending into the EELR.
Spectral analysis of the EELR indicates line ratios consistent with Seyfert
ionization in the center and LINER-type ionization in the outer diffuse region,
suggesting ionization from galactic nuclear activity. The required ionizing
luminosity, estimated from the H$\rm{\alpha}$ and H$\rm{\beta}$ luminosities
based on the photoionization and recombination balance assumption, is
$10^{41.8}$ $\mathrm{erg\,s^{-1}}$ for all spaxels classified as AGN, and
$10^{40.7}$ $\mathrm{erg\,s^{-1}}$ for spaxels in the central
$0.9~\mathrm{kpc}$ Seyfert region. However, the current bolometric luminosity
of the nucleus $L_{\text{bol}} \leq 10^{40.8}\,\mathrm{erg\,s^{-1}}$, estimated
from quiescent-state soft X-ray observations, is insufficient to ionize the
entire EELR, implying a recently faded AGN or a delayed response to historical
activity. Stellar population analysis reveals a post-starburst characteristic
in the EELR, and the gas kinematics show disturbances and non-circular
components compared to the stellar kinematics. Notably, the recent detection of
quasi-periodic eruptions (QPEs) in the X-ray light curve of AT2019qiz confirms
the TDE-QPE association. Our findings provide direct evidence for an AGN-like
EELR in the host galaxy of the nearest TDE with QPE detection, offering new
insights into the complex interplay between TDEs, QPEs, AGN activity, and host
galaxy evolution.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 14:46:54 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 16:18:04 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Xiong",
"Yifei",
""
],
[
"Jiang",
"Ning",
""
],
[
"Pan",
"Zhen",
""
],
[
"Hao",
"Lei",
""
],
[
"Li",
"Zhenzhen",
""
]
] |
2503.19807 | Luis Irastorza Valera | Luis Irastorza-Valera, Luis Saucedo-Mora | Probabilistic combination of loads in topology optimization designs via
cumulative damage criteria | null | null | null | null | physics.comp-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Topology optimization (TO) is a well-established methodology for structural
design under user-defined constraints, e.g. minimum volume and maximum
stiffness. However, such methods have traditionally been applied to static,
deterministic loading, in which modulus, position and direction are known and
invariant. This is against the probabilistic load combination used in the
structural engineering designs, and entails two important shortcomings.
The first one is related to maintenance and reliability: static loading fails
to consider naturally occurring uncertainties in the loading process,
measurements or regular service; also ignoring (quantitatively) unforeseen
phenomena such as vibrations, and the material's behavior is assumed linear
isotropic, ignoring fatigue, plasticity and anisotropy in functionally-graded
materials. The second one concerns optimality itself: often times, the
structure presented as "optimal" in fact over-estimates loading and thus wastes
material by exceeding its real needs and/or distributing it poorly throughout
the design dominion.
In this article, a probabilistic framework is presented: uncertain and
pseudo-dynamic loading is introduced to create robust topologies via a
reinforced SIMP scheme with embedded penalization addressing fatigue damage,
layer direction, mechanical response (traction/compression) and yield limit
(von Mises equivalent stress). This computationally efficient framework is
applied to various loading scenarios, generating diverse designs for the same
volume fraction constraint with improved and more realistic performances. Under
the proposed method, if loads are permanent and damage isotropic, the
methodology converges to the traditional (deterministic) topological
optimization results. Future ramifications of this work are pondered,
especially regarding metamaterial design.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 16:17:32 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:55:45 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Irastorza-Valera",
"Luis",
""
],
[
"Saucedo-Mora",
"Luis",
""
]
] |
2503.19838 | Sarah McCarthy | Sarah E. McCarthy, Ali Anwar, Daniel K. L. Oi, Loyd J. McKnight | Compact and stable source of polarization-entangled photon-pairs based
on a folded linear displacement interferometer | 11 pages, 8 figures, submitted | null | null | null | quant-ph physics.optics | http://creativecommons.org/licenses/by/4.0/ | The realization of quantum networks requires the development of robust low
size, weight and power (SWaP) systems suitable for operation under harsh
environments in remote and mobile nodes such as satellites. We present a source
of polarization-entangled photon-pairs in a folded linear displacement
interferometer based on spontaneous parametric down conversion using a Type-0
periodically poled potassium titanyl phosphate crystal. Featuring a compact and
stable double-pass geometry using a corner-cube retroreflector, the source has
a detected pair rate of 2.5 M pairs/s/mW with a Bell state fidelity of 94.1%
+/- 2.1%. The qualities and demonstrated performance of the source make it
suitable for deployment in entanglement-based quantum networks.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 16:59:38 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 15:41:28 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"McCarthy",
"Sarah E.",
""
],
[
"Anwar",
"Ali",
""
],
[
"Oi",
"Daniel K. L.",
""
],
[
"McKnight",
"Loyd J.",
""
]
] |
2503.19862 | Armin Ilg | Jan-Hendrik Arling (1), Alexander Burgman (2), Christina Dimitriadi
(3), Ulrich Einhaus (4), Axel Gall\'en (5), Abdelhamid Haddad (6), Laura
Huhta (7), Armin Ilg (8), Jan Klamka (9), Elizabeth Long (10), Thomas
Madlener (1), Arnau Morancho Tard\`a (11), Emanuela Musumeci (12), Krzysztof
M\k{e}ka{\l}a (1 and 9), Elena Pompa Pacchi (13), Marvin Pfaff (14), Daniel
Reichelt (15), Leonhard Reichenbach (15 and 16), Birgit Stapf (15), Francesco
P. Ucci (17 and 18), Erik Wallin (19), Harriet Watson (20), Sagar Vidya
Addepalli (21), Bruno Alves (22), Robert Mihai Amarinei (23), Ricardo
Barru\'e (24), Lydia Brenner (25), Giacomo Da Molin (24), Arturo de Giorgi
(26), Bohdan Dudar (27), Francesco Giuli (28 and 29), Andrea Gurgone (30 and
31), C\'esar Jes\'us-Valls (32), Antoine Laudrain (1), Martin J. Losekamm
(33), Rafa{\l} Mase{\l}ek (34), Wrishik Naskar (35), Miquel Nebot-Guinot
(20), Marko Pesut (8), Thomas P\"oschl (15), Efrain P. Segarra (36), Rebecca
Taylor (14 and 15), Pavel Vana (10), Hannah Wakeling (37), Aidan R.
Wiederhold (38) ((1) Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany,
(2) Stockholm University, Stockholm, Sweden, (3) KTH Royal Institute of
Technology, Stockholm, Sweden, (4) Karlsruhe Institute for Technology,
Karlsruhe, Germany, (5) Uppsala University, Uppsala, Sweden, (6) Laboratoire
de Physique de Clermont Auvergne, CNRS/IN2P3, Universit\'e Clermont Auvergne,
France, (7) University of Jyv\"askyl\"a, Jyv\"askyl\"a, Finland, (8)
University of Z\"urich, Z\"urich, Switzerland, (9) University of Warsaw,
Warsaw, Poland, (10) Charles University, Prague, Czech Republic, (11) Niels
Bohr Institute, Copenhagen, Denmark, (12) Instituto de F\'isica Corpuscular
(IFIC), CSIC - Universitat de Val\`encia, Paterna (Val\`encia), Spain, (13)
The University of Oklahoma, Norman, Oklahoma, USA, (14) Imperial College
London, London, United Kingdom, (15) CERN, Geneva, Switzerland, (16)
University of Bonn, Bonn, Germany, (17) University of Pavia, Pavia, Italy,
(18) INFN - Sezione di Pavia, Pavia, Italy, (19) Lund University, Lund,
Sweden, (20) The University of Edinburgh, Edinburgh, United Kingdom, (21)
SLAC National Accelerator Laboratory, Menlo Park, USA, (22) Laboratoire
Leprince-Ringuet, CNRS/IN2P3, Ecole Polytechnique, Institut Polytechnique de
Paris, Palaiseau, France, (23) University of Geneva, Geneva, Switzerland,
(24) Laborat\'orio de Instrumenta\c{c}\~ao e F\'isica Experimental de
Part\'iculas (LIP), Lisbon, Portugal, (25) National Institute for Subatomic
Physics (NIKHEF), Amsterdam, Netherlands, (26) Institute for Particle Physics
Phenomenology, Durham University, Durham, United Kingdom, (27) University of
Mainz, Mainz, Germany, (28) Dipartimento di Fisica, Universit\`a degli Studi
di Roma Tor Vergata, Rome, Italy, (29) INFN - Sezione di Roma 2, Rome, Italy,
(30) Universit\`a di Pisa, Pisa, Italy, (31) INFN - Sezione di Pisa, Pisa,
Italy, (32) Kavli IPMU (WPI), UTIAS, The University of Tokyo, Kashiwa, Japan,
(33) Technical University of Munich, Munich, Germany, (34) Laboratoire de
physique subatomique et de cosmologie de Grenoble, Grenoble, France, (35)
University of Glasgow, Glasgow, United Kingdom, (36) Paul Scherrer Institute,
Villigen, Switzerland, (37) John Adams Institute for Accelerator Science,
University of Oxford, Oxford, United Kingdom, (38) University of Manchester,
Manchester, United Kingdom) | Early Career Researcher Input to the European Strategy for Particle
Physics Update: White Paper | Endorsed by the ECFA ECR Panel. Editor and author attribution in the
document | null | null | null | hep-ex hep-ph hep-th physics.acc-ph physics.ins-det | http://creativecommons.org/licenses/by/4.0/ | This document, written by early career researchers (ECRs) in particle
physics, aims to represent the perspectives of the European ECR community and
serves as input for the 2025--2026 update of the European Strategy for Particle
Physics. With input from a community-wide survey, it highlights key challenges
faced by ECRs -- career stability, funding access and long-term research
opportunities -- while proposing policy recommendations and targeted
initiatives. It underscores the importance of practices fostering diverse,
equitable, inclusive and healthy workplaces, as well as of stronger ECR
communities, and highlights how effective communication and interdisciplinary
collaborations reinforce the societal relevance of particle physics and promote
continued support for large-scale and long-term projects. Finally, the future
of both collider and beyond-collider experiments is addressed, emphasising the
critical role of ECRs in shaping future projects. The ECR contribution is
formed of two parts: the ten-page executive summary submitted as input to the
European Strategy for Particle Physics Update and, as backup document, this
extended white paper providing additional context.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 17:29:10 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:47:02 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Arling",
"Jan-Hendrik",
"",
"1 and 9"
],
[
"Burgman",
"Alexander",
"",
"1 and 9"
],
[
"Dimitriadi",
"Christina",
"",
"1 and 9"
],
[
"Einhaus",
"Ulrich",
"",
"1 and 9"
],
[
"Gallén",
"Axel",
"",
"1 and 9"
],
[
"Haddad",
"Abdelhamid",
"",
"1 and 9"
],
[
"Huhta",
"Laura",
"",
"1 and 9"
],
[
"Ilg",
"Armin",
"",
"1 and 9"
],
[
"Klamka",
"Jan",
"",
"1 and 9"
],
[
"Long",
"Elizabeth",
"",
"1 and 9"
],
[
"Madlener",
"Thomas",
"",
"1 and 9"
],
[
"Tardà",
"Arnau Morancho",
"",
"1 and 9"
],
[
"Musumeci",
"Emanuela",
"",
"1 and 9"
],
[
"Mękała",
"Krzysztof",
"",
"1 and 9"
],
[
"Pacchi",
"Elena Pompa",
"",
"15 and 16"
],
[
"Pfaff",
"Marvin",
"",
"15 and 16"
],
[
"Reichelt",
"Daniel",
"",
"15 and 16"
],
[
"Reichenbach",
"Leonhard",
"",
"15 and 16"
],
[
"Stapf",
"Birgit",
"",
"17 and 18"
],
[
"Ucci",
"Francesco P.",
"",
"17 and 18"
],
[
"Wallin",
"Erik",
"",
"28 and 29"
],
[
"Watson",
"Harriet",
"",
"28 and 29"
],
[
"Addepalli",
"Sagar Vidya",
"",
"28 and 29"
],
[
"Alves",
"Bruno",
"",
"28 and 29"
],
[
"Amarinei",
"Robert Mihai",
"",
"28 and 29"
],
[
"Barrué",
"Ricardo",
"",
"28 and 29"
],
[
"Brenner",
"Lydia",
"",
"28 and 29"
],
[
"Da Molin",
"Giacomo",
"",
"28 and 29"
],
[
"de Giorgi",
"Arturo",
"",
"28 and 29"
],
[
"Dudar",
"Bohdan",
"",
"28 and 29"
],
[
"Giuli",
"Francesco",
"",
"28 and 29"
],
[
"Gurgone",
"Andrea",
"",
"30 and\n 31"
],
[
"Jesús-Valls",
"César",
"",
"14 and 15"
],
[
"Laudrain",
"Antoine",
"",
"14 and 15"
],
[
"Losekamm",
"Martin J.",
"",
"14 and 15"
],
[
"Masełek",
"Rafał",
"",
"14 and 15"
],
[
"Naskar",
"Wrishik",
"",
"14 and 15"
],
[
"Nebot-Guinot",
"Miquel",
"",
"14 and 15"
],
[
"Pesut",
"Marko",
"",
"14 and 15"
],
[
"Pöschl",
"Thomas",
"",
"14 and 15"
],
[
"Segarra",
"Efrain P.",
"",
"14 and 15"
],
[
"Taylor",
"Rebecca",
"",
"14 and 15"
],
[
"Vana",
"Pavel",
""
],
[
"Wakeling",
"Hannah",
""
],
[
"Wiederhold",
"Aidan R.",
""
]
] |
2503.19920 | Angelo Nucciotti | B.K. Alpert, M. Balata, D.T. Becker, D.A. Bennett, M. Borghesi, P.
Campana, R. Carobene, M. De Gerone, W.B. Doriese, M. Faverzani, L. Ferrari
Barusso, E. Ferri, J.W. Fowler, G. Gallucci, S. Gamba, J.D. Gard, F. Gatti,
A. Giachero, M. Gobbo, U. K\"oster, D. Labranca, M. Lusignoli, P.
Manfrinetti, J.A.B. Mates, E. Maugeri, R. Moretti, S. Nisi, A. Nucciotti,
G.C. O'Neil, L. Origo, G. Pessina, S. Ragazzi, C.D. Reintsema, D.R. Schmidt,
D. Schumann, D.S. Swetz, Z. Talip, J.N. Ullom and L.R. Vale | Most stringent bound on electron neutrino mass obtained with a scalable
low temperature microcalorimeter array | 9 pages, 6 figures, for submission to PRL | null | null | null | hep-ex nucl-ex physics.ins-det | http://creativecommons.org/licenses/by/4.0/ | The determination of the absolute neutrino mass scale remains a fundamental
open question in particle physics, with profound implications for both the
Standard Model and cosmology. Direct kinematic measurements, independent of
model-dependent assumptions, provide the most robust approach to address this
challenge. In this Letter, we present the most stringent upper bound on the
effective electron neutrino mass ever obtained with a calorimetric measurement
of the electron capture decay of $^{163}$Ho. The HOLMES experiment employs an
array of ion-implanted transition-edge sensor (TES) microcalorimeters,
achieving an average energy resolution of 6 eV FWHM with a scalable,
multiplexed readout technique. With a total of $7\times10^7$ decay events
recorded over two months and a Bayesian statistical analysis, we derive an
upper limit of $m_{\beta}<27$ eV/c$^2$ at 90% credibility. These results
validate the feasibility of $^{163}$Ho calorimetry for next-generation neutrino
mass experiments and demonstrate the potential of a scalable TES-based
microcalorimetric technique to push the sensitivity of direct neutrino mass
measurements beyond the current state of the art.
| [
{
"version": "v1",
"created": "Mon, 10 Mar 2025 21:26:20 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 07:03:18 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Alpert",
"B. K.",
""
],
[
"Balata",
"M.",
""
],
[
"Becker",
"D. T.",
""
],
[
"Bennett",
"D. A.",
""
],
[
"Borghesi",
"M.",
""
],
[
"Campana",
"P.",
""
],
[
"Carobene",
"R.",
""
],
[
"De Gerone",
"M.",
""
],
[
"Doriese",
"W. B.",
""
],
[
"Faverzani",
"M.",
""
],
[
"Barusso",
"L. Ferrari",
""
],
[
"Ferri",
"E.",
""
],
[
"Fowler",
"J. W.",
""
],
[
"Gallucci",
"G.",
""
],
[
"Gamba",
"S.",
""
],
[
"Gard",
"J. D.",
""
],
[
"Gatti",
"F.",
""
],
[
"Giachero",
"A.",
""
],
[
"Gobbo",
"M.",
""
],
[
"Köster",
"U.",
""
],
[
"Labranca",
"D.",
""
],
[
"Lusignoli",
"M.",
""
],
[
"Manfrinetti",
"P.",
""
],
[
"Mates",
"J. A. B.",
""
],
[
"Maugeri",
"E.",
""
],
[
"Moretti",
"R.",
""
],
[
"Nisi",
"S.",
""
],
[
"Nucciotti",
"A.",
""
],
[
"O'Neil",
"G. C.",
""
],
[
"Origo",
"L.",
""
],
[
"Pessina",
"G.",
""
],
[
"Ragazzi",
"S.",
""
],
[
"Reintsema",
"C. D.",
""
],
[
"Schmidt",
"D. R.",
""
],
[
"Schumann",
"D.",
""
],
[
"Swetz",
"D. S.",
""
],
[
"Talip",
"Z.",
""
],
[
"Ullom",
"J. N.",
""
],
[
"Vale",
"L. R.",
""
]
] |
2503.19923 | Manuel De Castro Ribeiro Jardim | Cesare Maria Dalbagno, Manuel de Castro Ribeiro Jardim, Mihnea
Anghelu\c{t}\u{a} | Mapping fMRI Signal and Image Stimuli in an Artificial Neural Network
Latent Space: Bringing Artificial and Natural Minds Together | 4 pages, 3 figures | null | null | null | q-bio.NC cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The goal of this study is to investigate whether latent space representations
of visual stimuli and fMRI data share common information. Decoding and
reconstructing stimuli from fMRI data remains a challenge in AI and
neuroscience, with significant implications for understanding neural
representations and improving the interpretability of Artificial Neural
Networks (ANNs). In this preliminary study, we investigate the feasibility of
such reconstruction by examining the similarity between the latent spaces of
one autoencoder (AE) and one vision transformer (ViT) trained on fMRI and image
data, respectively. Using representational similarity analysis (RSA), we found
that the latent spaces of the two domains appear different. However, these
initial findings are inconclusive, and further research is needed to explore
this relationship more thoroughly.
| [
{
"version": "v1",
"created": "Wed, 12 Mar 2025 08:40:39 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 09:41:43 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Dalbagno",
"Cesare Maria",
""
],
[
"Jardim",
"Manuel de Castro Ribeiro",
""
],
[
"Angheluţă",
"Mihnea",
""
]
] |
2503.19924 | Athokpam Langlen Chanu | Athokpam Langlen Chanu, Youngjai Park, Younghwa Cha, UnCheol Lee,
Joon-Young Moon, and Jong-Min Park | Human Brain State Analysis Unveils Complexity of Brain Dynamics:
Insights from General Anesthesia and ADHD EEG signals | null | null | null | null | q-bio.NC nlin.AO | http://creativecommons.org/licenses/by/4.0/ | The human brain is a complex system exhibiting rich dynamical behavior across
various states, including those induced by anesthesia or neurological
disorders. Using electroencephalography (EEG) recordings, our study
investigates the underlying complexity and universal patterns in human brain
dynamics across states induced by general anesthesia and neurological disorder
of inattentive-type attention deficit hyperactivity disorder (ADHD). We extract
relative phase dynamics time series, $\beta(t)$, from EEG signals and compute
permutation entropy (PE) and statistical complexity across the different states
using the framework of ordinal patterns. Our results reveal several key
findings. First, different brain states exhibit distinct PE values, indicating
distinct signatures of information content across states. We find an inverse
correlation between entropy and the level of consciousness during general
anesthesia. Further, when mapped onto the complexity-entropy causality plane,
all brain states, regardless of condition, individual, or $\beta(t)$ time
series, align along a single curve, suggesting an underlying universal pattern
in brain dynamics. Moreover, compared to well-known stochastic processes
(linear underdamped Langevin dynamics, active Ornstein-Uhlenbeck process, and
fractional Brownian motion), brain data consistently exhibits higher complexity
for any given PE value. Multifractal analysis shows this enhanced complexity
likely arises from greater multifractal scaling properties compared to
stochastic processes. Our findings highlight the power of ordinal patterns in
distinguishing various dynamic brain states and uncovering hidden universal
patterns in brain dynamics. Our comprehensive characterization of human brain
complexity across states offers valuable insights for future research into
consciousness, attention disorders, and neural information processing.
| [
{
"version": "v1",
"created": "Wed, 12 Mar 2025 15:00:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 05:50:21 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chanu",
"Athokpam Langlen",
""
],
[
"Park",
"Youngjai",
""
],
[
"Cha",
"Younghwa",
""
],
[
"Lee",
"UnCheol",
""
],
[
"Moon",
"Joon-Young",
""
],
[
"Park",
"Jong-Min",
""
]
] |
2503.19969 | K-Ryan Hinds Mr | K-Ryan Hinds, Daniel Perley, Jesper Sollerman, Adam Miller,
Christoffer Fremling, Takashi Moriya, Kaustav Das, Yu-Jing Qin, Eric Bellm,
Xi Tracy Chen, Michael Coughlin, Wynn Jacobson-Gal\'an, Mansi Kasliwal,
Shrinivas Kulkarni, Ashish Mahabal, F. Masci, Priscila, J. Pessi, J. N.
Purdum, Reed Riddle, Avinash Singh, Roger Smith and Niharika Sravan | Inferring CSM Properties of Type II SNe Using a Magnitude-Limited ZTF
Sample | null | null | null | null | astro-ph.HE | http://creativecommons.org/licenses/by-sa/4.0/ | Although all Type II supernovae (SNe) originate from massive stars possessing
a hydrogen-rich envelope, their light curve morphology is diverse, reflecting
poorly characterised heterogeneity in the physical properties of their
progenitor systems. Here, we present a detailed light curve analysis of a
magnitude-limited sample of 639 Type II SNe from the Zwicky Transient Facility
Bright Transient Survey. Using Gaussian processes, we systematically measure
empirical light curve features (e.g. rise times, peak colours and luminosities)
in a robust sampling-independent manner. We focus on rise times as they are
highly sensitive to pre-explosion progenitor properties, especially the
presence of a dense circumstellar medium (CSM) shed by the progenitor in the
years immediately pre-explosion. By correlating our feature measurements with
physical parameters from an extensive grid of STELLA hydrodynamical models with
varying progenitor properties (CSM structure, $\dot M$, $R_{CSM}$ and
$M_{ZAMS}$), we quantify the proportion of events with sufficient pre-explosion
mass-loss to significantly alter the initial light curve (roughly $M_{CSM} \geq
10^{-2.5} M_{\odot}$) in a highly complete sample of 377 spectroscopically
classified Type II SNe. We find that 67 $\pm$ 6\% of observed SNe in our
magnitude-limited sample show evidence for substantial CSM ($M_{CSM} \geq
10^{-2.5} M_{\odot}$) close to the progenitor ($R_{CSM} <10^{15}$ cm) at the
time of explosion. After applying a volumetric-correction, we find
36$^{+5}_{-7}$\% of all Type II SN progenitors possess substantial CSM within
$10^{15}$ cm at the time of explosion. This high fraction of progenitors with
dense CSM, supported by photometric and spectroscopic evidence of previous SNe,
reveals mass-loss rates significantly exceeding those measured in local group
red supergiants or predicted by current theoretical models.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 18:00:03 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 09:09:10 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Hinds",
"K-Ryan",
""
],
[
"Perley",
"Daniel",
""
],
[
"Sollerman",
"Jesper",
""
],
[
"Miller",
"Adam",
""
],
[
"Fremling",
"Christoffer",
""
],
[
"Moriya",
"Takashi",
""
],
[
"Das",
"Kaustav",
""
],
[
"Qin",
"Yu-Jing",
""
],
[
"Bellm",
"Eric",
""
],
[
"Chen",
"Xi Tracy",
""
],
[
"Coughlin",
"Michael",
""
],
[
"Jacobson-Galán",
"Wynn",
""
],
[
"Kasliwal",
"Mansi",
""
],
[
"Kulkarni",
"Shrinivas",
""
],
[
"Mahabal",
"Ashish",
""
],
[
"Masci",
"F.",
""
],
[
"Priscila",
"",
""
],
[
"Pessi",
"J.",
""
],
[
"Purdum",
"J. N.",
""
],
[
"Riddle",
"Reed",
""
],
[
"Singh",
"Avinash",
""
],
[
"Smith",
"Roger",
""
],
[
"Sravan",
"Niharika",
""
]
] |
2503.20074 | Yahav Biran | Yahav Biran, Imry Kissos | Adaptive Orchestration for Large-Scale Inference on Heterogeneous
Accelerator Systems Balancing Cost, Performance, and Resilience | 14 pages, 7 figures | null | null | null | cs.PF cs.AI | http://creativecommons.org/licenses/by/4.0/ | The surge in generative AI workloads has created a need for scalable
inference systems that can flexibly harness both GPUs and specialized
accelerators while containing operational costs. This paper proposes a
hardware-agnostic control loop that adaptively allocates requests across
heterogeneous accelerators based on real-time cost and capacity signals. The
approach sustains low latency and high throughput by dynamically shifting
between cost-optimized and capacity-optimized modes, ensuring the most
efficient use of expensive compute resources under fluctuating availability.
Evaluated using the Stable Diffusion model, the framework consistently meets
latency targets, automatically redirects traffic during capacity shortfalls,
and capitalizes on lower-cost accelerators when possible. These results
highlight how a feedback-driven deployment strategy, spanning the entire
software and hardware stack, can help organizations efficiently scale
generative AI workloads while maintaining resilience in the face of limited
accelerator capacity.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 21:20:11 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:16:44 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Biran",
"Yahav",
""
],
[
"Kissos",
"Imry",
""
]
] |
2503.20083 | Benjamin Minixhofer | Benjamin Minixhofer, Ivan Vuli\'c, Edoardo Maria Ponti | Cross-Tokenizer Distillation via Approximate Likelihood Matching | Preprint | null | null | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Distillation has shown remarkable success in transferring knowledge from a
Large Language Model (LLM) teacher to a student LLM. However, current
distillation methods predominantly require the same tokenizer between the
teacher and the student, restricting their applicability to only a small subset
of teacher-student pairs. In this work, we develop a cross-tokenizer
distillation method to solve this crucial deficiency. Our method is the first
to enable cross-tokenizer distillation without a next-token prediction loss as
the main objective, instead purely maximizing the student predictions'
similarity to the teacher's predictions (known as pure distillation), while
also being robust to large mismatches between the teacher and the student
tokenizer function and vocabulary. Empirically, our method enables
substantially improved performance as tested on two use cases. First, we show
that viewing tokenizer transfer as self-distillation enables unprecedently
effective transfer across tokenizers. We transfer (subword-level) Llama and
Gemma models to byte-level tokenization more effectively than prior methods
transfer to a similar subword tokenizer under a comparable training budget.
Transferring different base models to the same tokenizer also enables
ensembling them (e.g., via averaging their predicted probabilities) which
boosts performance. Second, we use our cross-tokenizer distillation method to
distil a large maths-specialized LLM into a smaller model, achieving
competitive maths problem-solving performance. Overall, our results make
substantial strides toward better adaptability and enhanced interaction between
different LLMs.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 21:44:10 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 08:54:04 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Minixhofer",
"Benjamin",
""
],
[
"Vulić",
"Ivan",
""
],
[
"Ponti",
"Edoardo Maria",
""
]
] |
2503.20235 | Ahyun Seo | Ahyun Seo, Minsu Cho | Leveraging 3D Geometric Priors in 2D Rotation Symmetry Detection | Accepted to CVPR 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Symmetry plays a vital role in understanding structural patterns, aiding
object recognition and scene interpretation. This paper focuses on rotation
symmetry, where objects remain unchanged when rotated around a central axis,
requiring detection of rotation centers and supporting vertices. Traditional
methods relied on hand-crafted feature matching, while recent segmentation
models based on convolutional neural networks detect rotation centers but
struggle with 3D geometric consistency due to viewpoint distortions. To
overcome this, we propose a model that directly predicts rotation centers and
vertices in 3D space and projects the results back to 2D while preserving
structural integrity. By incorporating a vertex reconstruction stage enforcing
3D geometric priors -- such as equal side lengths and interior angles -- our
model enhances robustness and accuracy. Experiments on the DENDI dataset show
superior performance in rotation axis detection and validate the impact of 3D
priors through ablation studies.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 05:02:16 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 02:40:25 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Seo",
"Ahyun",
""
],
[
"Cho",
"Minsu",
""
]
] |
2503.20262 | Tawfiq Ammari | Tawfiq Ammari, Anna Gutowska, Jacob Ziff, Casey Randazzo, Harihan
Subramonyam | From the CDC to emerging infectious disease publics: The long-now of
polarizing and complex health crises | null | null | null | null | cs.HC cs.SI | http://creativecommons.org/licenses/by/4.0/ | As the COVID-19 pandemic evolved, the Center for Disease Control and
Prevention used Twitter to share updates about the virus and safety guidelines,
reaching millions instantly, in what we call the CDC public. We analyze two
years of tweets, from, to, and about the CDC using a mixed-methods approach to
characterize the nature and credibility of COVID-19 discourse and audience
engagement. We found that the CDC is not engaging in two-way communication with
the CDC publics and that discussions about COVID-19 reflected societal
divisions and political polarization. We introduce a crisis message journey
concept showing how the CDC public responds to the changing nature of the
crisis (e.g., new variants) using ``receipts'' of earlier, and at times
contradictory, guidelines. We propose design recommendations to support the CDC
in tailoring messages to specific users and publics (e.g., users interested in
racial equity) and in managing misinformation, especially in reaction to crisis
flashpoints.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 06:09:29 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 00:44:15 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ammari",
"Tawfiq",
""
],
[
"Gutowska",
"Anna",
""
],
[
"Ziff",
"Jacob",
""
],
[
"Randazzo",
"Casey",
""
],
[
"Subramonyam",
"Harihan",
""
]
] |
2503.20275 | Jing Wang | Jing Wang, Chao Li, Taolei Wang, Jinyang Guo, Hanzhang Yang, Yiming
Zhuansun, Minyi Guo | Survey of Disaggregated Memory: Cross-layer Technique Insights for
Next-Generation Datacenters | null | null | null | null | cs.DC cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The growing scale of data requires efficient memory subsystems with large
memory capacity and high memory performance. Disaggregated architecture has
become a promising solution for today's cloud and edge computing for its
scalability and elasticity. As a critical part of disaggregation, disaggregated
memory faces many design challenges in many dimensions, including hardware
scalability, architecture structure, software system design, application
programmability, resource allocation, power management, etc. These challenges
inspire a number of novel solutions at different system levels to improve
system efficiency. In this paper, we provide a comprehensive review of
disaggregated memory, including the methodology and technologies of
disaggregated memory system foundation, optimization, and management. We study
the technical essentials of disaggregated memory systems and analyze them from
the hardware, architecture, system, and application levels. Then, we compare
the design details of typical cross-layer designs on disaggregated memory.
Finally, we discuss the challenges and opportunities of future disaggregated
memory works that serve better for next-generation elastic and efficient
datacenters.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 06:49:36 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 04:16:38 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wang",
"Jing",
""
],
[
"Li",
"Chao",
""
],
[
"Wang",
"Taolei",
""
],
[
"Guo",
"Jinyang",
""
],
[
"Yang",
"Hanzhang",
""
],
[
"Zhuansun",
"Yiming",
""
],
[
"Guo",
"Minyi",
""
]
] |
2503.20286 | Zhenyu Liang | Zhenyu Liang, Hao Li, Naiwei Yu, Kebin Sun, Ran Cheng | Bridging Evolutionary Multiobjective Optimization and GPU Acceleration
via Tensorization | Accepted by IEEE TEVC | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Evolutionary multiobjective optimization (EMO) has made significant strides
over the past two decades. However, as problem scales and complexities
increase, traditional EMO algorithms face substantial performance limitations
due to insufficient parallelism and scalability. While most work has focused on
algorithm design to address these challenges, little attention has been given
to hardware acceleration, thereby leaving a clear gap between EMO algorithms
and advanced computing devices, such as GPUs. To bridge the gap, we propose to
parallelize EMO algorithms on GPUs via the tensorization methodology. By
employing tensorization, the data structures and operations of EMO algorithms
are transformed into concise tensor representations, which seamlessly enables
automatic utilization of GPU computing. We demonstrate the effectiveness of our
approach by applying it to three representative EMO algorithms: NSGA-III,
MOEA/D, and HypE. To comprehensively assess our methodology, we introduce a
multiobjective robot control benchmark using a GPU-accelerated physics engine.
Our experiments show that the tensorized EMO algorithms achieve speedups of up
to 1113x compared to their CPU-based counterparts, while maintaining solution
quality and effectively scaling population sizes to hundreds of thousands.
Furthermore, the tensorized EMO algorithms efficiently tackle complex
multiobjective robot control tasks, producing high-quality solutions with
diverse behaviors. Source codes are available at
https://github.com/EMI-Group/evomo.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 07:30:23 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 02:00:11 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Liang",
"Zhenyu",
""
],
[
"Li",
"Hao",
""
],
[
"Yu",
"Naiwei",
""
],
[
"Sun",
"Kebin",
""
],
[
"Cheng",
"Ran",
""
]
] |
2503.20308 | Lee Chae-Yeon | Lee Chae-Yeon, Oh Hyun-Bin, Han EunGi, Kim Sung-Bin, Suekyeong Nam,
Tae-Hyun Oh | Perceptually Accurate 3D Talking Head Generation: New Definitions,
Speech-Mesh Representation, and Evaluation Metrics | CVPR 2025 | null | null | null | cs.GR cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advancements in speech-driven 3D talking head generation have made
significant progress in lip synchronization. However, existing models still
struggle to capture the perceptual alignment between varying speech
characteristics and corresponding lip movements. In this work, we claim that
three criteria -- Temporal Synchronization, Lip Readability, and Expressiveness
-- are crucial for achieving perceptually accurate lip movements. Motivated by
our hypothesis that a desirable representation space exists to meet these three
criteria, we introduce a speech-mesh synchronized representation that captures
intricate correspondences between speech signals and 3D face meshes. We found
that our learned representation exhibits desirable characteristics, and we plug
it into existing models as a perceptual loss to better align lip movements to
the given speech. In addition, we utilize this representation as a perceptual
metric and introduce two other physically grounded lip synchronization metrics
to assess how well the generated 3D talking heads align with these three
criteria. Experiments show that training 3D talking head generation models with
our perceptual loss significantly improve all three aspects of perceptually
accurate lip synchronization. Codes and datasets are available at
https://perceptual-3d-talking-head.github.io/.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 08:18:57 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 04:19:30 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chae-Yeon",
"Lee",
""
],
[
"Hyun-Bin",
"Oh",
""
],
[
"EunGi",
"Han",
""
],
[
"Sung-Bin",
"Kim",
""
],
[
"Nam",
"Suekyeong",
""
],
[
"Oh",
"Tae-Hyun",
""
]
] |
2503.20313 | Size Zheng | Size Zheng, Jin Fang, Xuegui Zheng, Qi Hou, Wenlei Bao, Ningxin Zheng,
Ziheng Jiang, Dongyang Wang, Jianxi Ye, Haibin Lin, Li-Wen Chang, Xin Liu | TileLink: Generating Efficient Compute-Communication Overlapping Kernels
using Tile-Centric Primitives | null | null | null | null | cs.DC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Large deep learning models have achieved state-of-the-art performance in a
wide range of tasks. These models often necessitate distributed systems for
efficient training and inference. The fundamental building blocks for
distributed model execution are intra-layer parallel operators. The most
effective approach to enhancing the performance of intra-layer parallel
operators involves overlapping computation with communication. The overlapping
can be achieved through either operator decomposition or kernel fusion. While
decomposing operators is straightforward to implement, it often results in
suboptimal performance. On the other hand, fusing communication kernels with
compute kernels demands significant expertise and is error-prone.
In this paper, we propose TileLink to enable efficient compilation and
generation of overlapped compute-communication kernels. TileLink is composed of
frontend and backend. In the frontend, TileLink decouples the design space of
communication and computation, linking these two parts via tile-centric
primitives. In the backend, TileLink translates these primitives into low-level
communication instructions, integrating the communication and computation
components to achieve overlapped execution. In experiments, TileLink achieves
from $1.17\times$ to $20.76\times$ speedup to non-overlapping baseline and
achieves performance comparable to state-of-the-art overlapping libraries on
GPUs.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 08:25:12 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:13:46 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zheng",
"Size",
""
],
[
"Fang",
"Jin",
""
],
[
"Zheng",
"Xuegui",
""
],
[
"Hou",
"Qi",
""
],
[
"Bao",
"Wenlei",
""
],
[
"Zheng",
"Ningxin",
""
],
[
"Jiang",
"Ziheng",
""
],
[
"Wang",
"Dongyang",
""
],
[
"Ye",
"Jianxi",
""
],
[
"Lin",
"Haibin",
""
],
[
"Chang",
"Li-Wen",
""
],
[
"Liu",
"Xin",
""
]
] |
2503.20321 | Jaeah Lee | Jaeah Lee, Changwoon Choi, Young Min Kim, Jaesik Park | Recovering Dynamic 3D Sketches from Videos | Accepted to CVPR 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding 3D motion from videos presents inherent challenges due to the
diverse types of movement, ranging from rigid and deformable objects to
articulated structures. To overcome this, we propose Liv3Stroke, a novel
approach for abstracting objects in motion with deformable 3D strokes. The
detailed movements of an object may be represented by unstructured motion
vectors or a set of motion primitives using a pre-defined articulation from a
template model. Just as a free-hand sketch can intuitively visualize scenes or
intentions with a sparse set of lines, we utilize a set of parametric 3D curves
to capture a set of spatially smooth motion elements for general objects with
unknown structures. We first extract noisy, 3D point cloud motion guidance from
video frames using semantic features, and our approach deforms a set of curves
to abstract essential motion features as a set of explicit 3D representations.
Such abstraction enables an understanding of prominent components of motions
while maintaining robustness to environmental factors. Our approach allows
direct analysis of 3D object movements from video, tackling the uncertainty
that typically occurs when translating real-world motion into recorded footage.
The project page is accessible via: https://jaeah.me/liv3stroke_web
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 08:43:21 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:31:57 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Lee",
"Jaeah",
""
],
[
"Choi",
"Changwoon",
""
],
[
"Kim",
"Young Min",
""
],
[
"Park",
"Jaesik",
""
]
] |
2503.20334 | Kazuyuki Sakamoto | Kazuyuki Sakamoto, Hirotaka Ishikawa, Takashi Wake, Chie Ishimoto, Jun
Fujii, Hendrik Bentmann, Minoru Ohtaka, Kenta Kuroda, Natsu Inoue, Takuma
Hattori, Toshio Miyamachi, Fumio Komori, Isamu Yamamoto, Cheng Fan, Peter
Kr\"uger, Hiroshi Ota, Fumihiko Matsui, Friedrich Reinert, Jos\'e Avila, and
Maria C. Asensio | Spatial Control of Charge Doping in n-Type Topological Insulators | null | Nano Letters 21, 2021, 4415 | 10.1021/acs.nanolett.1c01100 | null | cond-mat.mes-hall cond-mat.mtrl-sci | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Spatially controlling the Fermi level of topological insulators and keeping
its electronic states stable are indispensable processes to put this material
into practical use for semiconductor spintronics devices. So far, however, such
a method has not been established yet. Here we show a novel method for doping
hole into n-type topological insulators Bi$_2$X$_3$ (X= Se, Te) that overcomes
the shortcomings of the previous reported methods. The key of this doping is to
adsorb H$_2$O on Bi$_2$X$_3$ decorated with a small amount of carbon, and its
trigger is the irradiation of photon with sufficient energy to excite
core-electrons of the outermost layer atoms. This method allows controlling the
doping amount by the irradiation time, and acts as photolithography. Such a
tunable doping makes it possible to design the electronic states at the
nanometer scale, and thus paves a promising avenue toward the realization of
novel spintronics devices based on topological insulators.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 08:59:01 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Sakamoto",
"Kazuyuki",
""
],
[
"Ishikawa",
"Hirotaka",
""
],
[
"Wake",
"Takashi",
""
],
[
"Ishimoto",
"Chie",
""
],
[
"Fujii",
"Jun",
""
],
[
"Bentmann",
"Hendrik",
""
],
[
"Ohtaka",
"Minoru",
""
],
[
"Kuroda",
"Kenta",
""
],
[
"Inoue",
"Natsu",
""
],
[
"Hattori",
"Takuma",
""
],
[
"Miyamachi",
"Toshio",
""
],
[
"Komori",
"Fumio",
""
],
[
"Yamamoto",
"Isamu",
""
],
[
"Fan",
"Cheng",
""
],
[
"Krüger",
"Peter",
""
],
[
"Ota",
"Hiroshi",
""
],
[
"Matsui",
"Fumihiko",
""
],
[
"Reinert",
"Friedrich",
""
],
[
"Avila",
"José",
""
],
[
"Asensio",
"Maria C.",
""
]
] |
2503.20336 | Fuchao He | Yaru Fu, Fuchao He, Zheng Shi, and Haijun Zhang | Power Minimization for NOMA-assisted Pinching Antenna Systems With
Multiple Waveguides | null | null | null | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The integration of pinching antenna systems with non-orthogonal multiple
access (NOMA) has emerged as a promising technique for future 6G applications.
This paper is the first to investigate power minimization for NOMA-assisted
pinching antenna systems utilizing multiple dielectric waveguides. We formulate
a total power minimization problem constrained by each user's minimum data
requirements, addressing a classical challenge. To efficiently solve the
non-convex optimization problem, we propose an iterative algorithm.
Furthermore, we demonstrate that the interference function of this algorithm is
standard, ensuring convergence to a unique fixed point. Numerical simulations
validate that our developed algorithm converges within a few steps and
significantly outperforms benchmark strategies across various data rate
requirements. The results also indicate that the minimum transmit power, as a
function of the interval between the waveguides, exhibits an approximately
oscillatory decay with a negative trend.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 09:01:02 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 04:47:33 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Fu",
"Yaru",
""
],
[
"He",
"Fuchao",
""
],
[
"Shi",
"Zheng",
""
],
[
"Zhang",
"Haijun",
""
]
] |
2503.20349 | Weiyi You | Weiyi You, Mingyang Zhang, Leheng Zhang, Xingyu Zhou, Kexuan Shi,
Shuhang Gu | Consistency Trajectory Matching for One-Step Generative Super-Resolution | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current diffusion-based super-resolution (SR) approaches achieve commendable
performance at the cost of high inference overhead. Therefore, distillation
techniques are utilized to accelerate the multi-step teacher model into
one-step student model. Nevertheless, these methods significantly raise
training costs and constrain the performance of the student model by the
teacher model. To overcome these tough challenges, we propose Consistency
Trajectory Matching for Super-Resolution (CTMSR), a distillation-free strategy
that is able to generate photo-realistic SR results in one step. Concretely, we
first formulate a Probability Flow Ordinary Differential Equation (PF-ODE)
trajectory to establish a deterministic mapping from low-resolution (LR) images
with noise to high-resolution (HR) images. Then we apply the Consistency
Training (CT) strategy to directly learn the mapping in one step, eliminating
the necessity of pre-trained diffusion model. To further enhance the
performance and better leverage the ground-truth during the training process,
we aim to align the distribution of SR results more closely with that of the
natural images. To this end, we propose to minimize the discrepancy between
their respective PF-ODE trajectories from the LR image distribution by our
meticulously designed Distribution Trajectory Matching (DTM) loss, resulting in
improved realism of our recovered HR images. Comprehensive experimental results
demonstrate that the proposed methods can attain comparable or even superior
capabilities on both synthetic and real datasets while maintaining minimal
inference latency.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 09:20:42 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:59:15 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"You",
"Weiyi",
""
],
[
"Zhang",
"Mingyang",
""
],
[
"Zhang",
"Leheng",
""
],
[
"Zhou",
"Xingyu",
""
],
[
"Shi",
"Kexuan",
""
],
[
"Gu",
"Shuhang",
""
]
] |
2503.20468 | Theo Gheorghiu | T. Gheorghiu, F. Militello, J. Juul Rasmussen | Determination of non-local characteristics of density transport in 2D
simulations of the SOL | 14 pages, 8 figures | null | null | null | physics.plasm-ph physics.flu-dyn | http://creativecommons.org/licenses/by-sa/4.0/ | By use of Lagrangian tracers propagated on 2D simulations of Scrape-Off Layer
(SOL) turbulence, we are able to determine the non-local fractional-advection,
fractional-diffusion equation (FADE) coefficients for a number of equilibrium
cases. Solutions of the resultant FADEs shows good agreement with the simulated
mean density profiles. We detail how the FADE is derived: the stochastic flux
equation is introduced, and it is shown how it is used to find general forms of
Fick's first and second laws, dependent on the jump function. We show for
spatially homogeneous jump functions which belong to the Levy-alpha Stable
distribution that transport may be approximated by a non-local FADE with four
parameters. This work demonstrates the sound basis for FADEs to act as reduced
models of transport in systems dominated by coherent structures; so justifies
the development of a first-principles approach to calculating FADE parameters.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 11:51:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 09:13:06 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Gheorghiu",
"T.",
""
],
[
"Militello",
"F.",
""
],
[
"Rasmussen",
"J. Juul",
""
]
] |
2503.20498 | Ruslan Shaydulin | Minzhao Liu, Ruslan Shaydulin, Pradeep Niroula, Matthew DeCross,
Shih-Han Hung, Wen Yu Kon, Enrique Cervero-Mart\'in, Kaushik Chakraborty,
Omar Amer, Scott Aaronson, Atithi Acharya, Yuri Alexeev, K. Jordan Berg,
Shouvanik Chakrabarti, Florian J. Curchod, Joan M. Dreiling, Neal Erickson,
Cameron Foltz, Michael Foss-Feig, David Hayes, Travis S. Humble, Niraj Kumar,
Jeffrey Larson, Danylo Lykov, Michael Mills, Steven A. Moses, Brian
Neyenhuis, Shaltiel Eloul, Peter Siegfried, James Walker, Charles Lim, Marco
Pistoia | Certified randomness using a trapped-ion quantum processor | null | Nature (2025) | 10.1038/s41586-025-08737-1 | null | quant-ph cs.CR cs.ET | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While quantum computers have the potential to perform a wide range of
practically important tasks beyond the capabilities of classical computers,
realizing this potential remains a challenge. One such task is to use an
untrusted remote device to generate random bits that can be certified to
contain a certain amount of entropy. Certified randomness has many applications
but is fundamentally impossible to achieve solely by classical computation. In
this work, we demonstrate the generation of certifiably random bits using the
56-qubit Quantinuum H2-1 trapped-ion quantum computer accessed over the
internet. Our protocol leverages the classical hardness of recent random
circuit sampling demonstrations: a client generates quantum "challenge"
circuits using a small randomness seed, sends them to an untrusted quantum
server to execute, and verifies the server's results. We analyze the security
of our protocol against a restricted class of realistic near-term adversaries.
Using classical verification with measured combined sustained performance of
$1.1\times10^{18}$ floating-point operations per second across multiple
supercomputers, we certify $71,313$ bits of entropy under this restricted
adversary and additional assumptions. Our results demonstrate a step towards
the practical applicability of today's quantum computers.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 12:38:22 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Liu",
"Minzhao",
""
],
[
"Shaydulin",
"Ruslan",
""
],
[
"Niroula",
"Pradeep",
""
],
[
"DeCross",
"Matthew",
""
],
[
"Hung",
"Shih-Han",
""
],
[
"Kon",
"Wen Yu",
""
],
[
"Cervero-Martín",
"Enrique",
""
],
[
"Chakraborty",
"Kaushik",
""
],
[
"Amer",
"Omar",
""
],
[
"Aaronson",
"Scott",
""
],
[
"Acharya",
"Atithi",
""
],
[
"Alexeev",
"Yuri",
""
],
[
"Berg",
"K. Jordan",
""
],
[
"Chakrabarti",
"Shouvanik",
""
],
[
"Curchod",
"Florian J.",
""
],
[
"Dreiling",
"Joan M.",
""
],
[
"Erickson",
"Neal",
""
],
[
"Foltz",
"Cameron",
""
],
[
"Foss-Feig",
"Michael",
""
],
[
"Hayes",
"David",
""
],
[
"Humble",
"Travis S.",
""
],
[
"Kumar",
"Niraj",
""
],
[
"Larson",
"Jeffrey",
""
],
[
"Lykov",
"Danylo",
""
],
[
"Mills",
"Michael",
""
],
[
"Moses",
"Steven A.",
""
],
[
"Neyenhuis",
"Brian",
""
],
[
"Eloul",
"Shaltiel",
""
],
[
"Siegfried",
"Peter",
""
],
[
"Walker",
"James",
""
],
[
"Lim",
"Charles",
""
],
[
"Pistoia",
"Marco",
""
]
] |
2503.20519 | Jinnan Chen | Jinnan Chen, Lingting Zhu, Zeyu Hu, Shengju Qian, Yugang Chen, Xin
Wang, Gim Hee Lee | MAR-3D: Progressive Masked Auto-regressor for High-Resolution 3D
Generation | Accepted to CVPR 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in auto-regressive transformers have revolutionized
generative modeling across different domains, from language processing to
visual generation, demonstrating remarkable capabilities. However, applying
these advances to 3D generation presents three key challenges: the unordered
nature of 3D data conflicts with sequential next-token prediction paradigm,
conventional vector quantization approaches incur substantial compression loss
when applied to 3D meshes, and the lack of efficient scaling strategies for
higher resolution latent prediction. To address these challenges, we introduce
MAR-3D, which integrates a pyramid variational autoencoder with a cascaded
masked auto-regressive transformer (Cascaded MAR) for progressive latent
upscaling in the continuous space. Our architecture employs random masking
during training and auto-regressive denoising in random order during inference,
naturally accommodating the unordered property of 3D latent tokens.
Additionally, we propose a cascaded training strategy with condition
augmentation that enables efficiently up-scale the latent token resolution with
fast convergence. Extensive experiments demonstrate that MAR-3D not only
achieves superior performance and generalization capabilities compared to
existing methods but also exhibits enhanced scaling capabilities compared to
joint distribution modeling approaches (e.g., diffusion transformers).
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 13:00:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:39:55 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Jinnan",
""
],
[
"Zhu",
"Lingting",
""
],
[
"Hu",
"Zeyu",
""
],
[
"Qian",
"Shengju",
""
],
[
"Chen",
"Yugang",
""
],
[
"Wang",
"Xin",
""
],
[
"Lee",
"Gim Hee",
""
]
] |
2503.20522 | Subramanya Bhat | Subramanya Bhat K. N., Amita Das, Bhooshan Paradkar, V Ravishankar | Experimental Signatures for Identifying Distinct Origins of Color Field
Generation | 4 pages, 3 figures | null | null | null | physics.plasm-ph cond-mat.other hep-ph hep-th nucl-th | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Signatures for non-abelian dynamics have long been central to QCD and QGP.
Equally important are they in spin systems and laser-plasma interactions, where
they emerge as effective interactions. Distinguishing experimentally gauge
inequivalent sources (and hence potentials) that produce the same field tensor
is one major task in this endeavour. As a step in this direction, this paper
investigates how physically distinct sources which produce the same color
electric field (uniform and constant) may be distinguished experimentally in a
gauge-invariant manner. We first study the motion of a test particle in such
fields and show that the resultant trajectories are counterintuitive. We then
examine the radiation emitted - both gluonic and photonic and show that each
source (independent non-abelian configuration) leaves a unique signature in the
energy spectra, laying the ground for application to specific physical systems.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 13:10:59 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:08:25 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"N.",
"Subramanya Bhat K.",
""
],
[
"Das",
"Amita",
""
],
[
"Paradkar",
"Bhooshan",
""
],
[
"Ravishankar",
"V",
""
]
] |
2503.20578 | Mia Mohammad Imran | Alif Al Hasan, Subarna Saha, Mia Mohammad Imran, Tarannum Shaila Zaman | LLPut: Investigating Large Language Models for Bug Report-Based Input
Generation | null | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Failure-inducing inputs play a crucial role in diagnosing and analyzing
software bugs. Bug reports typically contain these inputs, which developers
extract to facilitate debugging. Since bug reports are written in natural
language, prior research has leveraged various Natural Language Processing
(NLP) techniques for automated input extraction. With the advent of Large
Language Models (LLMs), an important research question arises: how effectively
can generative LLMs extract failure-inducing inputs from bug reports? In this
paper, we propose LLPut, a technique to empirically evaluate the performance of
three open-source generative LLMs -- LLaMA, Qwen, and Qwen-Coder -- in
extracting relevant inputs from bug reports. We conduct an experimental
evaluation on a dataset of 206 bug reports to assess the accuracy and
effectiveness of these models. Our findings provide insights into the
capabilities and limitations of generative LLMs in automated bug diagnosis.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 14:25:01 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 10:35:05 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Hasan",
"Alif Al",
""
],
[
"Saha",
"Subarna",
""
],
[
"Imran",
"Mia Mohammad",
""
],
[
"Zaman",
"Tarannum Shaila",
""
]
] |
2503.20605 | Elisa Bellantoni | Elisa Bellantoni, Fabio Guglietta, Francesca Pelusi, Mathieu Desbrun,
Kiwon Um, Mihalis Nicolaou, Nikos Savva, Mauro Sbragaglia | Immersed boundary - lattice Boltzmann method for wetting problems | 14 pages, 9 figures | null | null | null | physics.flu-dyn physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | We develop a mesoscale computational model to describe the interaction of a
droplet with a solid. The model is based on the hybrid combination of the
immersed boundary and the lattice Boltzmann computational schemes: the former
is used to model the non-ideal sharp interface of the droplet coupled with the
inner and outer fluids, simulated with the lattice Boltzmann scheme. We further
introduce an interaction force to model the wetting interactions of the droplet
with the solid: this interaction force is designed with the key computational
advantage of providing a regularization of the interface profile close to the
contact line, avoiding abrupt curvature changes that could otherwise cause
numerical instabilities. The proposed model substantially improves earlier
immersed boundary - lattice Boltzmann models for wetting in that it allows a
description of an ample variety of wetting interactions, ranging from
hydrophobic to hydrophilic cases, without the need for any pre-calibration
study on model parameters to be used. Model validations against theoretical
results for droplet shape at equilibrium and scaling laws for droplet spreading
dynamics are addressed.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 14:52:04 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 15:10:34 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Bellantoni",
"Elisa",
""
],
[
"Guglietta",
"Fabio",
""
],
[
"Pelusi",
"Francesca",
""
],
[
"Desbrun",
"Mathieu",
""
],
[
"Um",
"Kiwon",
""
],
[
"Nicolaou",
"Mihalis",
""
],
[
"Savva",
"Nikos",
""
],
[
"Sbragaglia",
"Mauro",
""
]
] |
2503.20639 | Jeffery Painter | Jeffery L Painter, Gregory E Powell and Andrew Bate | PVLens: Enhancing Pharmacovigilance Through Automated Label Extraction | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Reliable drug safety reference databases are essential for pharmacovigilance,
yet existing resources like SIDER are outdated and static. We introduce PVLens,
an automated system that extracts labeled safety information from FDA
Structured Product Labels (SPLs) and maps terms to MedDRA. PVLens integrates
automation with expert oversight through a web-based review tool. In validation
against 97 drug labels, PVLens achieved an F1 score of 0.882, with high recall
(0.983) and moderate precision (0.799). By offering a scalable, more accurate
and continuously updated alternative to SIDER, PVLens enhances real-time
pharamcovigilance with improved accuracy and contemporaneous insights.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 15:33:26 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:19:38 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Painter",
"Jeffery L",
""
],
[
"Powell",
"Gregory E",
""
],
[
"Bate",
"Andrew",
""
]
] |
2503.20646 | Alexandra Watkins Ph.D | Alexandra Watkins, Ritam Ghosh, Evan Chow, Nilanjan Sarkar | Immersive and Wearable Thermal Rendering for Augmented Reality | null | null | null | null | cs.HC cs.RO cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | In augmented reality (AR), where digital content is overlaid onto the real
world, realistic thermal feedback has been shown to enhance immersion. Yet
current thermal feedback devices, heavily influenced by the needs of virtual
reality, often hinder physical interactions and are ineffective for immersion
in AR. To bridge this gap, we have identified three design considerations
relevant for AR thermal feedback: indirect feedback to maintain dexterity,
thermal passthrough to preserve real-world temperature perception, and
spatiotemporal rendering for dynamic sensations. We then created a unique and
innovative thermal feedback device that satisfies these criteria. Human subject
experiments assessing perceptual sensitivity, object temperature matching,
spatial pattern recognition, and moving thermal stimuli demonstrated the impact
of our design, enabling realistic temperature discrimination, virtual object
perception, and enhanced immersion. These findings demonstrate that carefully
designed thermal feedback systems can bridge the sensory gap between physical
and virtual interactions, enhancing AR realism and usability.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 15:40:10 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:39:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Watkins",
"Alexandra",
""
],
[
"Ghosh",
"Ritam",
""
],
[
"Chow",
"Evan",
""
],
[
"Sarkar",
"Nilanjan",
""
]
] |
2503.20652 | Theo Di Piazza | Theo Di Piazza, Carole Lazarus, Olivier Nempont, Loic Boussel | Imitating Radiological Scrolling: A Global-Local Attention Model for 3D
Chest CT Volumes Multi-Label Anomaly Classification | 13 pages, 4 figures. Accepted for MIDL 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The rapid increase in the number of Computed Tomography (CT) scan
examinations has created an urgent need for automated tools, such as organ
segmentation, anomaly classification, and report generation, to assist
radiologists with their growing workload. Multi-label classification of
Three-Dimensional (3D) CT scans is a challenging task due to the volumetric
nature of the data and the variety of anomalies to be detected. Existing deep
learning methods based on Convolutional Neural Networks (CNNs) struggle to
capture long-range dependencies effectively, while Vision Transformers require
extensive pre-training, posing challenges for practical use. Additionally,
these existing methods do not explicitly model the radiologist's navigational
behavior while scrolling through CT scan slices, which requires both global
context understanding and local detail awareness. In this study, we present
CT-Scroll, a novel global-local attention model specifically designed to
emulate the scrolling behavior of radiologists during the analysis of 3D CT
scans. Our approach is evaluated on two public datasets, demonstrating its
efficacy through comprehensive experiments and an ablation study that
highlights the contribution of each model component.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 15:47:50 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:46:42 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Di Piazza",
"Theo",
""
],
[
"Lazarus",
"Carole",
""
],
[
"Nempont",
"Olivier",
""
],
[
"Boussel",
"Loic",
""
]
] |
2503.20660 | Hozefa Jesawada | Hozefa Jesawada, Antonio Acernese, Giovanni Russo, Carmen Del Vecchio | DR-PETS: Learning-Based Control With Planning in Adversarial
Environments | null | null | null | null | cs.LG math.OC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Ensuring robustness against epistemic, possibly adversarial, perturbations is
essential for reliable real-world decision-making. While the Probabilistic
Ensembles with Trajectory Sampling (PETS) algorithm inherently handles
uncertainty via ensemble-based probabilistic models, it lacks guarantees
against structured adversarial or worst-case uncertainty distributions. To
address this, we propose DR-PETS, a distributionally robust extension of PETS
that certifies robustness against adversarial perturbations. We formalize
uncertainty via a p-Wasserstein ambiguity set, enabling worst-case-aware
planning through a min-max optimization framework. While PETS passively
accounts for stochasticity, DR-PETS actively optimizes robustness via a
tractable convex approximation integrated into PETS planning loop. Experiments
on pendulum stabilization and cart-pole balancing show that DR-PETS certifies
robustness against adversarial parameter perturbations, achieving consistent
performance in worst-case scenarios where PETS deteriorates.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 15:55:44 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:07:27 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Jesawada",
"Hozefa",
""
],
[
"Acernese",
"Antonio",
""
],
[
"Russo",
"Giovanni",
""
],
[
"Del Vecchio",
"Carmen",
""
]
] |
2503.20673 | Yinan Sun | Yinan Sun, Xiongkuo Min, Zicheng Zhang, Yixuan Gao, Yuqin Cao,
Guangtao Zhai | Mitigating Low-Level Visual Hallucinations Requires Self-Awareness:
Database, Model and Training Strategy | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid development of multimodal large language models has resulted in
remarkable advancements in visual perception and understanding, consolidating
several tasks into a single visual question-answering framework. However, these
models are prone to hallucinations, which limit their reliability as artificial
intelligence systems. While this issue is extensively researched in natural
language processing and image captioning, there remains a lack of investigation
of hallucinations in Low-level Visual Perception and Understanding (HLPU),
especially in the context of image quality assessment tasks. We consider that
these hallucinations arise from an absence of clear self-awareness within the
models. To address this issue, we first introduce the HLPU instruction
database, the first instruction database specifically focused on hallucinations
in low-level vision tasks. This database contains approximately 200K
question-answer pairs and comprises four subsets, each covering different types
of instructions. Subsequently, we propose the Self-Awareness Failure
Elimination (SAFEQA) model, which utilizes image features, salient region
features and quality features to improve the perception and comprehension
abilities of the model in low-level vision tasks. Furthermore, we propose the
Enhancing Self-Awareness Preference Optimization (ESA-PO) framework to increase
the model's awareness of knowledge boundaries, thereby mitigating the incidence
of hallucination. Finally, we conduct comprehensive experiments on low-level
vision tasks, with the results demonstrating that our proposed method
significantly enhances self-awareness of the model in these tasks and reduces
hallucinations. Notably, our proposed method improves both accuracy and
self-awareness of the proposed model and outperforms close-source models in
terms of various evaluation metrics.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 16:05:01 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 02:04:02 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Sun",
"Yinan",
""
],
[
"Min",
"Xiongkuo",
""
],
[
"Zhang",
"Zicheng",
""
],
[
"Gao",
"Yixuan",
""
],
[
"Cao",
"Yuqin",
""
],
[
"Zhai",
"Guangtao",
""
]
] |
2503.20677 | Alessio Cela | Alessio Cela and Carl Lian | Curves on Hirzebruch Surfaces and Semistability | Comments welcome | null | null | null | math.AG | http://creativecommons.org/licenses/by/4.0/ | When $a\ge2$, we show that a general pointed curve never interpolates through
the expected number of points in the Hirzebruch surface $\mathcal{H}_a$, with
one exception. In the exceptional case, the number of such interpolating maps
is determined by the geometric Tevelev degrees of $\mathbb{P}^1$, which have
been previously computed.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 16:10:38 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 07:52:29 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Cela",
"Alessio",
""
],
[
"Lian",
"Carl",
""
]
] |
2503.20685 | Yuhao Huang | Yuhao Huang, Ao Chang, Haoran Dou, Xing Tao, Xinrui Zhou, Yan Cao,
Ruobing Huang, Alejandro F Frangi, Lingyun Bao, Xin Yang, Dong Ni | Flip Learning: Weakly Supervised Erase to Segment Nodules in Breast
Ultrasound | Accepted by Medical Image Analysis. 24 pages, 13 figures, 20 tabels | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate segmentation of nodules in both 2D breast ultrasound (BUS) and 3D
automated breast ultrasound (ABUS) is crucial for clinical diagnosis and
treatment planning. Therefore, developing an automated system for nodule
segmentation can enhance user independence and expedite clinical analysis.
Unlike fully-supervised learning, weakly-supervised segmentation (WSS) can
streamline the laborious and intricate annotation process. However, current WSS
methods face challenges in achieving precise nodule segmentation, as many of
them depend on inaccurate activation maps or inefficient pseudo-mask generation
algorithms. In this study, we introduce a novel multi-agent reinforcement
learning-based WSS framework called Flip Learning, which relies solely on 2D/3D
boxes for accurate segmentation. Specifically, multiple agents are employed to
erase the target from the box to facilitate classification tag flipping, with
the erased region serving as the predicted segmentation mask. The key
contributions of this research are as follows: (1) Adoption of a
superpixel/supervoxel-based approach to encode the standardized environment,
capturing boundary priors and expediting the learning process. (2) Introduction
of three meticulously designed rewards, comprising a classification score
reward and two intensity distribution rewards, to steer the agents' erasing
process precisely, thereby avoiding both under- and over-segmentation. (3)
Implementation of a progressive curriculum learning strategy to enable agents
to interact with the environment in a progressively challenging manner, thereby
enhancing learning efficiency. Extensively validated on the large in-house BUS
and ABUS datasets, our Flip Learning method outperforms state-of-the-art WSS
methods and foundation models, and achieves comparable performance as
fully-supervised learning algorithms.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 16:20:02 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:16:16 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Huang",
"Yuhao",
""
],
[
"Chang",
"Ao",
""
],
[
"Dou",
"Haoran",
""
],
[
"Tao",
"Xing",
""
],
[
"Zhou",
"Xinrui",
""
],
[
"Cao",
"Yan",
""
],
[
"Huang",
"Ruobing",
""
],
[
"Frangi",
"Alejandro F",
""
],
[
"Bao",
"Lingyun",
""
],
[
"Yang",
"Xin",
""
],
[
"Ni",
"Dong",
""
]
] |
2503.20711 | Giovanni Compiani | Giovanni Compiani, Ilya Morozov, Stephan Seiler | Demand Estimation with Text and Image Data | null | null | null | null | econ.GN cs.CV cs.LG q-fin.EC | http://creativecommons.org/licenses/by/4.0/ | We propose a demand estimation method that leverages unstructured text and
image data to infer substitution patterns. Using pre-trained deep learning
models, we extract embeddings from product images and textual descriptions and
incorporate them into a random coefficients logit model. This approach enables
researchers to estimate demand even when they lack data on product attributes
or when consumers value hard-to-quantify attributes, such as visual design or
functional benefits. Using data from a choice experiment, we show that our
approach outperforms standard attribute-based models in counterfactual
predictions of consumers' second choices. We also apply it across 40 product
categories on Amazon and consistently find that text and image data help
identify close substitutes within each category.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 16:47:14 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:28:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Compiani",
"Giovanni",
""
],
[
"Morozov",
"Ilya",
""
],
[
"Seiler",
"Stephan",
""
]
] |
2503.20749 | Yuxuan Lu | Yuxuan Lu, Jing Huang, Yan Han, Bennet Bei, Yaochen Xie, Dakuo Wang,
Jessie Wang, Qi He | Beyond Believability: Accurate Human Behavior Simulation with Fine-Tuned
LLMs | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent research shows that LLMs can simulate ``believable'' human behaviors
to power LLM agents via prompt-only methods. In this work, we focus on
evaluating and improving LLM's objective ``accuracy'' rather than the
subjective ``believability'' in the web action generation task, leveraging a
large-scale, real-world dataset collected from online shopping human actions.
We present the first comprehensive quantitative evaluation of state-of-the-art
LLMs (e.g., DeepSeek-R1, Llama, and Claude) on the task of web action
generation. Our results show that fine-tuning LLMs on real-world behavioral
data substantially improves their ability to generate actions compared to
prompt-only methods. Furthermore, incorporating synthesized reasoning traces
into model training leads to additional performance gains, demonstrating the
value of explicit rationale in behavior modeling. This work establishes a new
benchmark for evaluating LLMs in behavior simulation and offers actionable
insights into how real-world action data and reasoning augmentation can enhance
the fidelity of LLM agents.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 17:33:27 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 02:42:03 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Lu",
"Yuxuan",
""
],
[
"Huang",
"Jing",
""
],
[
"Han",
"Yan",
""
],
[
"Bei",
"Bennet",
""
],
[
"Xie",
"Yaochen",
""
],
[
"Wang",
"Dakuo",
""
],
[
"Wang",
"Jessie",
""
],
[
"He",
"Qi",
""
]
] |
2503.20752 | Huajie Tan | Huajie Tan, Yuheng Ji, Xiaoshuai Hao, Minglan Lin, Pengwei Wang,
Zhongyuan Wang, Shanghang Zhang | Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning | 35 pages, 22 figures | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual reasoning abilities play a crucial role in understanding complex
multimodal data, advancing both domain-specific applications and artificial
general intelligence (AGI). Existing methods improve VLM reasoning via
Chain-of-Thought (CoT) supervised fine-tuning, using meticulously annotated
training data to enhance visual reasoning capabilities. However, this training
paradigm may lead to overfitting and cognitive rigidity, restricting the
model's ability to transfer visual reasoning skills across domains and limiting
its real-world applicability. To address these limitations, we propose
Reason-RFT, a novel reinforcement fine-tuning framework that significantly
enhances generalization capabilities in visual reasoning tasks. Reason-RFT
introduces a two-phase training framework for visual reasoning: (1) Supervised
Fine-Tuning (SFT) with curated Chain-of-Thought (CoT) data activates the
reasoning potential of Vision-Language Models (VLMs), followed by (2) Group
Relative Policy Optimization (GRPO)-based reinforcement learning that generates
multiple reasoning-response pairs, significantly enhancing generalization in
visual reasoning tasks. To evaluate Reason-RFT's visual reasoning capabilities,
we reconstructed a comprehensive dataset spanning visual counting, structure
perception, and spatial transformation. Experimental results demonstrate
Reasoning-RFT's three key advantages: (1) Performance Enhancement: achieving
state-of-the-art results across multiple tasks, outperforming most mainstream
open-source and proprietary models; (2) Generalization Superiority:
consistently maintaining robust performance across diverse tasks and domains,
outperforming alternative training paradigms; (3) Data Efficiency: excelling in
few-shot learning scenarios while surpassing full-dataset SFT baselines.
Project website: https://tanhuajie.github.io/ReasonRFT
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 17:38:06 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 03:13:00 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Tan",
"Huajie",
""
],
[
"Ji",
"Yuheng",
""
],
[
"Hao",
"Xiaoshuai",
""
],
[
"Lin",
"Minglan",
""
],
[
"Wang",
"Pengwei",
""
],
[
"Wang",
"Zhongyuan",
""
],
[
"Zhang",
"Shanghang",
""
]
] |
2503.20768 | Qizhen Zhang | Haotian Yang, Zhuoran Wang, Benson Chou, Sophie Xu, Hao Wang, Jingxian
Wang, Qizhen Zhang | An Empirical Study of the Impact of Federated Learning on Machine
Learning Model Accuracy | null | null | null | null | cs.LG cs.DC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Federated Learning (FL) enables distributed ML model training on private user
data at the global scale. Despite the potential of FL demonstrated in many
domains, an in-depth view of its impact on model accuracy remains unclear. In
this paper, we investigate, systematically, how this learning paradigm can
affect the accuracy of state-of-the-art ML models for a variety of ML tasks. We
present an empirical study that involves various data types: text, image,
audio, and video, and FL configuration knobs: data distribution, FL scale,
client sampling, and local and global computations. Our experiments are
conducted in a unified FL framework to achieve high fidelity, with substantial
human efforts and resource investments. Based on the results, we perform a
quantitative analysis of the impact of FL, and highlight challenging scenarios
where applying FL degrades the accuracy of the model drastically and identify
cases where the impact is negligible. The detailed and extensive findings can
benefit practical deployments and future development of FL.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 17:52:30 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 02:16:06 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Yang",
"Haotian",
""
],
[
"Wang",
"Zhuoran",
""
],
[
"Chou",
"Benson",
""
],
[
"Xu",
"Sophie",
""
],
[
"Wang",
"Hao",
""
],
[
"Wang",
"Jingxian",
""
],
[
"Zhang",
"Qizhen",
""
]
] |
2503.20769 | Andrei Zeleneev | Ben Deaner, Chen-Wei Hsiang, Andrei Zeleneev | Inferring Treatment Effects in Large Panels by Uncovering Latent
Similarities | null | null | null | null | econ.EM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The presence of unobserved confounders is one of the main challenges in
identifying treatment effects. In this paper, we propose a new approach to
causal inference using panel data with large large $N$ and $T$. Our approach
imputes the untreated potential outcomes for treated units using the outcomes
for untreated individuals with similar values of the latent confounders. In
order to find units with similar latent characteristics, we utilize long
pre-treatment histories of the outcomes. Our analysis is based on a
nonparametric, nonlinear, and nonseparable factor model for untreated potential
outcomes and treatments. The model satisfies minimal smoothness requirements.
We impute both missing counterfactual outcomes and propensity scores using
kernel smoothing based on the constructed measure of latent similarity between
units, and demonstrate that our estimates can achieve the optimal nonparametric
rate of convergence up to log terms. Using these estimates, we construct a
doubly robust estimator of the period-specifc average treatment effect on the
treated (ATT), and provide conditions, under which this estimator is
$\sqrt{N}$-consistent, and asymptotically normal and unbiased. Our simulation
study demonstrates that our method provides accurate inference for a wide range
of data generating processes.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 17:53:06 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:53:05 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Deaner",
"Ben",
""
],
[
"Hsiang",
"Chen-Wei",
""
],
[
"Zeleneev",
"Andrei",
""
]
] |
2503.20787 | Cheng Wang | Cheng Wang, Chuwen Wang, Shirong Zeng, Changjun Jiang | Advanced Digital Simulation for Financial Market Dynamics: A Case of
Commodity Futures | null | null | null | null | q-fin.TR cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | After decades of evolution, the financial system has increasingly deviated
from an idealized framework based on theorems. It necessitates accurate
projections of complex market dynamics and human behavioral patterns. With the
development of data science and machine intelligence, researchers are trying to
digitalize and automate market prediction. However, existing methodologies
struggle to represent the diversity of individuals and are regardless of the
domino effects of interactions on market dynamics, leading to the poor
performance facing abnormal market conditions where non-quantitative
information dominates the market. To alleviate these disadvantages requires the
introduction of knowledge about how non-quantitative information, like news and
policy, affects market dynamics. This study investigates overcoming these
challenges through rehearsing potential market trends based on the financial
large language model agents whose behaviors are aligned with their cognition
and analyses in markets. We propose a hierarchical knowledge architecture for
financial large language model agents, integrating fine-tuned language models
and specialized generators optimized for trading scenarios. For financial
market, we develop an advanced interactive behavioral simulation system that
enables users to configure agents and automate market simulations. In this
work, we take commodity futures as an example to research the effectiveness of
our methodologies. Our real-world case simulation succeeds in rehearsing
abnormal market dynamics under geopolitical events and reaches an average
accuracy of 3.4% across various points in time after the event on predicting
futures price. Experimental results demonstrate our method effectively
leverages diverse information to simulate behaviors and their impact on market
dynamics through systematic interaction.
| [
{
"version": "v1",
"created": "Tue, 18 Feb 2025 12:40:04 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wang",
"Cheng",
""
],
[
"Wang",
"Chuwen",
""
],
[
"Zeng",
"Shirong",
""
],
[
"Jiang",
"Changjun",
""
]
] |
2503.20788 | Sarah Perez | Sarah Perez, Florian Doster, Julien Maes, Hannah Menke, Ahmed ElSheikh
and Andreas Busch | When Cubic Law and Darcy Fail: Correcting Model Misspecification in
Fracture Conductivities | null | null | null | null | physics.geo-ph physics.flu-dyn | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Uncertainties in the subsurface challenge reliable leakage risk assessments
and long-term integrity of CO2 storage. Fault zones, characterised by
multi-scale heterogeneities, are critical pathways for leakage and suffer from
significant data uncertainties due to unresolved features. Understanding the
multi-scale uncertainties in estimating fracture network conductivity is
essential for mitigating leakage risks and ensuring reliable modelling of
upscaled fault leakage rates. However, current models fail to capture all fault
zone heterogeneity, particularly fracture surface roughness effects on fluid
migration. Simplified assumptions, such as the Cubic Law based on mechanical
aperture measurements, lead to erroneous models known as model
misspecifications and introduce modelling uncertainty in leakage predictions.
Here, we develop an AI-driven framework to address data uncertainties and
correct model misspecifications in estimating fracture conductivities. By
automatically integrating small-scale uncertainties, including fracture surface
roughness, and leveraging their interactions across scales, we improve upon
traditional empirical corrections. We combine physics-based constraints with
adaptive, data-driven and geometric corrections to infer the hydraulic aperture
governing fluid flow in fractures with varying roughness. This approach
generates reliable local permeability maps that account for roughness effects
and discrepancies between mechanical and hydraulic apertures, accurately
reflecting overall fracture conductivity. By propagating uncertainties from
individual fractures to network scales, our approach will support robust
calibration of conductivity ranges for fault leakage sensitivity analyses, not
only resolving uncertainties in fracture-scale modelling but also enabling
efficient integration into larger-scale simulations to enhance predictions for
subsurface CO2 storage.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 19:32:44 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Perez",
"Sarah",
""
],
[
"Doster",
"Florian",
""
],
[
"Maes",
"Julien",
""
],
[
"Menke",
"Hannah",
""
],
[
"ElSheikh",
"Ahmed",
""
],
[
"Busch",
"Andreas",
""
]
] |
2503.20789 | Sowad Rahman | Sowad Rahman | Neuro-Informed Adaptive Learning (NIAL) Algorithm: A Hybrid Deep
Learning Approach for ECG Signal Classification | 1 figure ,2 pages | null | null | null | eess.SP cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The detection of cardiac abnormalities using electrocardiogram (ECG) signals
is crucial for early diagnosis and intervention in cardiovascular diseases.
Traditional deep learning models often lack adaptability to varying signal
patterns. This study introduces the Neuro-Informed Adaptive Learning (NIAL)
algorithm, a hybrid approach integrating convolutional neural networks (CNNs)
and transformer-based attention mechanisms to enhance ECG signal
classification. The algorithm dynamically adjusts learning rates based on
real-time validation performance, ensuring efficient convergence. Using the
MIT-BIH Arrhythmia and PTB Diagnostic ECG datasets, our model achieves high
classification accuracy, outperforming conventional approaches. These findings
highlight the potential of NIAL in real-time cardiovascular monitoring
applications.
| [
{
"version": "v1",
"created": "Wed, 12 Mar 2025 14:37:53 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Rahman",
"Sowad",
""
]
] |
2503.20790 | Hsiang-Ting Chen | Hsiang-Ting Chen, Yuan Zhang, Gustavo Carneiro, Rajvinder Singh | Toward a Human-Centered AI-assisted Colonoscopy System in Australia | 4 pages, accepted by CHI '25 workshop Envisioning the Future of
Interactive Health | null | null | null | cs.HC cs.AI | http://creativecommons.org/licenses/by/4.0/ | While AI-assisted colonoscopy promises improved colorectal cancer screening,
its success relies on effective integration into clinical practice, not just
algorithmic accuracy. This paper, based on an Australian field study
(observations and gastroenterologist interviews), highlights a critical
disconnect: current development prioritizes machine learning model performance,
overlooking essential aspects of user interface design, workflow integration,
and overall user experience. Industry interactions reveal a similar emphasis on
data and algorithms. To realize AI's full potential, the HCI community must
champion user-centered design, ensuring these systems are usable, support
endoscopist expertise, and enhance patient outcomes.
| [
{
"version": "v1",
"created": "Sat, 15 Mar 2025 23:36:48 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Hsiang-Ting",
""
],
[
"Zhang",
"Yuan",
""
],
[
"Carneiro",
"Gustavo",
""
],
[
"Singh",
"Rajvinder",
""
]
] |
2503.20791 | Zifan Liu | John Murzaku, Zifan Liu, Vaishnavi Muppala, Md Mehrab Tanjim, Xiang
Chen, Yunyao Li | ECLAIR: Enhanced Clarification for Interactive Responses in an
Enterprise AI Assistant | 3 pages, 1 figure | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) have shown remarkable progress in understanding
and generating natural language across various applications. However, they
often struggle with resolving ambiguities in real-world, enterprise-level
interactions, where context and domain-specific knowledge play a crucial role.
In this demonstration, we introduce ECLAIR (Enhanced CLArification for
Interactive Responses), a multi-agent framework for interactive disambiguation.
ECLAIR enhances ambiguous user query clarification through an interactive
process where custom agents are defined, ambiguity reasoning is conducted by
the agents, clarification questions are generated, and user feedback is
leveraged to refine the final response. When tested on real-world customer
data, ECLAIR demonstrates significant improvements in clarification question
generation compared to standard few-shot methods.
| [
{
"version": "v1",
"created": "Wed, 19 Mar 2025 23:13:34 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Murzaku",
"John",
""
],
[
"Liu",
"Zifan",
""
],
[
"Muppala",
"Vaishnavi",
""
],
[
"Tanjim",
"Md Mehrab",
""
],
[
"Chen",
"Xiang",
""
],
[
"Li",
"Yunyao",
""
]
] |
2503.20792 | Muhammad Azam | M. Awais and M. Azam | Anisotropic Compact Star with Vaidya-Tikekar Potential in $f(Q)$ Gravity | 26 pages, 19 figures | null | null | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The goal of this work is to develop physical models for spherically symmetric
systems in the realm of $f(Q)$ gravity. The field equations are set up for
anisotropic fluid and formulate these equations using physical ansatz of
Vaidya-Tikekar solution. The intrinsic constants of the model be ascertained by
matching the interior solution to the external Schwarzschild solution. The use
of accessibility conditions helps to understand the physical viability of the
obtained model. LMC X-4 pulsar is used as a toy model to test the viability of
the model and Mathematica software is used for deeper insights and
visualization of the physical parameters against fixed values of constants.
| [
{
"version": "v1",
"created": "Thu, 20 Mar 2025 10:54:47 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Awais",
"M.",
""
],
[
"Azam",
"M.",
""
]
] |
2503.20793 | Tapio Pitk\"aranta Mr | Tapio Pitk\"aranta, Eero Hyv\"onen | Semantic Web -- A Forgotten Wave of Artificial Intelligence? | 21 pages, 9 figures | null | null | null | cs.SI cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The history of Artificial Intelligence is a narrative of waves - rising
optimism followed by crashing disappointments. AI winters, such as the early
2000s, are often remembered as barren periods of innovation. This paper argues
that such a perspective overlooks a crucial wave of AI that seems to be
forgotten: the rise of the Semantic Web, which is based on knowledge
representation, logic, and reasoning, and its interplay with intelligent
Software Agents. Fast forward to today, and ChatGPT has reignited AI
enthusiasm, built on deep learning and advanced neural models. However, before
Large Language Models dominated the conversation, another ambitious vision
emerged - one where AI-driven Software Agents autonomously served Web users
based on a structured, machine-interpretable Web. The Semantic Web aimed to
transform the World Wide Web into an ecosystem where AI could reason,
understand, and act. Between 2000 and 2010, this vision sparked a significant
research boom, only to fade into obscurity as AI's mainstream narrative shifted
elsewhere. Today, as LLMs edge toward autonomous execution, we revisit this
overlooked wave. By analyzing its academic impact through bibliometric data, we
highlight the Semantic Web's role in AI history and its untapped potential for
modern Software Agent development. Recognizing this forgotten chapter not only
deepens our understanding of AI's cyclical evolution but also offers key
insights for integrating emerging technologies.
| [
{
"version": "v1",
"created": "Thu, 20 Mar 2025 12:55:48 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Pitkäranta",
"Tapio",
""
],
[
"Hyvönen",
"Eero",
""
]
] |
2503.20794 | Veysel Kocaman Vk | Veysel Kocaman, Muhammed Santas, Yigit Gul, Mehmet Butgul, David Talby | Can Zero-Shot Commercial APIs Deliver Regulatory-Grade Clinical Text
DeIdentification? | 14 pages, accepted at Text2Story Workshop at ECIR 2025 | null | null | null | cs.CL cs.CR cs.IR cs.LG | http://creativecommons.org/licenses/by/4.0/ | We systematically assess the performance of three leading API-based
de-identification systems - Azure Health Data Services, AWS Comprehend Medical,
and OpenAI GPT-4o - against our de-identification systems on a ground truth
dataset of 48 clinical documents annotated by medical experts. Our analysis,
conducted at both entity-level and token-level, demonstrates that our solution,
Healthcare NLP, achieves the highest accuracy, with a 96% F1-score in protected
health information (PHI) detection, significantly outperforming Azure (91%),
AWS (83%), and GPT-4o (79%). Beyond accuracy, Healthcare NLP is also the most
cost-effective solution, reducing processing costs by over 80% compared to
Azure and GPT-4o. Its fixed-cost local deployment model avoids the escalating
per-request fees of cloud-based services, making it a scalable and economical
choice. Our results underscore a critical limitation: zero-shot commercial APIs
fail to meet the accuracy, adaptability, and cost-efficiency required for
regulatory-grade clinical de-identification. Healthcare NLP's superior
performance, customization capabilities, and economic advantages position it as
the more viable solution for healthcare organizations seeking compliance and
scalability in clinical NLP workflows.
| [
{
"version": "v1",
"created": "Fri, 21 Mar 2025 10:05:04 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kocaman",
"Veysel",
""
],
[
"Santas",
"Muhammed",
""
],
[
"Gul",
"Yigit",
""
],
[
"Butgul",
"Mehmet",
""
],
[
"Talby",
"David",
""
]
] |
2503.20795 | Lorenzo Bertini | Lorenzo Bertini, Joel L. Lebowitz, Mario Pulvirenti | Errico Presutti 1942-2024 | Memorial for E. Presutti | null | null | null | cond-mat.stat-mech math-ph math.MP math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Errico Presutti was a leading figure in mathematical physics and an important
contributor to rigorous results in statistical mechanics. Due to his strong
scientific personality and human qualities, there are many who remember Errico
Presutti as colleague, mentor, and friend.
| [
{
"version": "v1",
"created": "Fri, 21 Mar 2025 13:56:42 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Bertini",
"Lorenzo",
""
],
[
"Lebowitz",
"Joel L.",
""
],
[
"Pulvirenti",
"Mario",
""
]
] |
2503.20796 | Priyanka Kumar | Bryan Lim, Roman Huerta, Alejandro Sotelo, Anthonie Quintela, Priyanka
Kumar | EXPLICATE: Enhancing Phishing Detection through Explainable AI and
LLM-Powered Interpretability | null | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Sophisticated phishing attacks have emerged as a major cybersecurity threat,
becoming more common and difficult to prevent. Though machine learning
techniques have shown promise in detecting phishing attacks, they function
mainly as "black boxes" without revealing their decision-making rationale. This
lack of transparency erodes the trust of users and diminishes their effective
threat response. We present EXPLICATE: a framework that enhances phishing
detection through a three-component architecture: an ML-based classifier using
domain-specific features, a dual-explanation layer combining LIME and SHAP for
complementary feature-level insights, and an LLM enhancement using DeepSeek v3
to translate technical explanations into accessible natural language. Our
experiments show that EXPLICATE attains 98.4 % accuracy on all metrics, which
is on par with existing deep learning techniques but has better explainability.
High-quality explanations are generated by the framework with an accuracy of
94.2 % as well as a consistency of 96.8\% between the LLM output and model
prediction. We create EXPLICATE as a fully usable GUI application and a light
Chrome extension, showing its applicability in many deployment situations. The
research shows that high detection performance can go hand-in-hand with
meaningful explainability in security applications. Most important, it
addresses the critical divide between automated AI and user trust in phishing
detection systems.
| [
{
"version": "v1",
"created": "Sat, 22 Mar 2025 23:37:35 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Lim",
"Bryan",
""
],
[
"Huerta",
"Roman",
""
],
[
"Sotelo",
"Alejandro",
""
],
[
"Quintela",
"Anthonie",
""
],
[
"Kumar",
"Priyanka",
""
]
] |
2503.20797 | Muhammad Haroon | Muhammad Haroon, Magdalena Wojcieszak, Anshuman Chhabra | "Whose Side Are You On?" Estimating Ideology of Political and News
Content Using Large Language Models and Few-shot Demonstration Selection | null | null | null | null | cs.CL cs.CY cs.SI | http://creativecommons.org/licenses/by/4.0/ | The rapid growth of social media platforms has led to concerns about
radicalization, filter bubbles, and content bias. Existing approaches to
classifying ideology are limited in that they require extensive human effort,
the labeling of large datasets, and are not able to adapt to evolving
ideological contexts. This paper explores the potential of Large Language
Models (LLMs) for classifying the political ideology of online content in the
context of the two-party US political spectrum through in-context learning
(ICL). Our extensive experiments involving demonstration selection in
label-balanced fashion, conducted on three datasets comprising news articles
and YouTube videos, reveal that our approach significantly outperforms
zero-shot and traditional supervised methods. Additionally, we evaluate the
influence of metadata (e.g., content source and descriptions) on ideological
classification and discuss its implications. Finally, we show how providing the
source for political and non-political content influences the LLM's
classification.
| [
{
"version": "v1",
"created": "Sun, 23 Mar 2025 02:32:25 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Haroon",
"Muhammad",
""
],
[
"Wojcieszak",
"Magdalena",
""
],
[
"Chhabra",
"Anshuman",
""
]
] |
2503.20798 | Yongcheol Kim | Yongcheol Kim, Chanjae Lee, Young Yoon | Payload-Aware Intrusion Detection with CMAE and Large Language Models | null | null | null | null | cs.CR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Intrusion Detection Systems (IDS) are crucial for identifying malicious
traffic, yet traditional signature-based methods struggle with zero-day attacks
and high false positive rates. AI-driven packet-capture analysis offers a
promising alternative. However, existing approaches rely heavily on flow-based
or statistical features, limiting their ability to detect fine-grained attack
patterns. This study proposes Xavier-CMAE, an enhanced Convolutional Multi-Head
Attention Ensemble (CMAE) model that improves detection accuracy while reducing
computational overhead. By replacing Word2Vec embeddings with a Hex2Int
tokenizer and Xavier initialization, Xavier-CMAE eliminates pre-training,
accelerates training, and achieves 99.971% accuracy with a 0.018% false
positive rate, outperforming Word2Vec-based methods. Additionally, we introduce
LLM-CMAE, which integrates pre-trained Large Language Model (LLM) tokenizers
into CMAE. While LLMs enhance feature extraction, their computational cost
hinders real-time detection. LLM-CMAE balances efficiency and performance,
reaching 99.969% accuracy with a 0.019% false positive rate. This work advances
AI-powered IDS by (1) introducing a payload-based detection framework, (2)
enhancing efficiency with Xavier-CMAE, and (3) integrating LLM tokenizers for
improved real-time detection.
| [
{
"version": "v1",
"created": "Sun, 23 Mar 2025 02:56:32 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kim",
"Yongcheol",
""
],
[
"Lee",
"Chanjae",
""
],
[
"Yoon",
"Young",
""
]
] |
2503.20799 | Boming Zhang | Boming Zhang and Yunguo Jiang | Fractal structure of Christ-Lee model | null | null | null | null | cond-mat.stat-mech nlin.PS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper numerically investigates the dynamical properties of kink and
antikink collisions in the Christ_Lee model in the regime of epsilon
approaching the phi4 theory. With given epsilon and the initial velocity Vin,
we exhibiting the formation of bion, scattering, and n_bounce states.
Additionally, we show the self_similar fractal structures in the plot of
Vout_Vin with given epsilon. Specially, we find the fractal structure in the
plot of Vout versus epsilon, which is not reported previously. We computes the
Box_counting dimension for these fractal structures. We find that the
Box_dimension is positively correlated with epsilon, and approaches to
Hausdorff dimension of the Sierpinski triangle when epsilon is sufficiently
large.
| [
{
"version": "v1",
"created": "Sun, 23 Mar 2025 15:27:46 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Boming",
""
],
[
"Jiang",
"Yunguo",
""
]
] |
2503.20800 | Tao Qi | Qi Tao, Yin Jinhua, Cai Dongqi, Xie Yueqi, Wang Huili, Hu Zhiyang,
Yang Peiru, Nan Guoshun, Zhou Zhili, Wang Shangguang, Lyu Lingjuan, Huang
Yongfeng, Lane Nicholas | Evidencing Unauthorized Training Data from AI Generated Content using
Information Isotopes | null | null | null | null | cs.CR cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | In light of scaling laws, many AI institutions are intensifying efforts to
construct advanced AIs on extensive collections of high-quality human data.
However, in a rush to stay competitive, some institutions may inadvertently or
even deliberately include unauthorized data (like privacy- or intellectual
property-sensitive content) for AI training, which infringes on the rights of
data owners. Compounding this issue, these advanced AI services are typically
built on opaque cloud platforms, which restricts access to internal information
during AI training and inference, leaving only the generated outputs available
for forensics. Thus, despite the introduction of legal frameworks by various
countries to safeguard data rights, uncovering evidence of data misuse in
modern opaque AI applications remains a significant challenge. In this paper,
inspired by the ability of isotopes to trace elements within chemical
reactions, we introduce the concept of information isotopes and elucidate their
properties in tracing training data within opaque AI systems. Furthermore, we
propose an information isotope tracing method designed to identify and provide
evidence of unauthorized data usage by detecting the presence of target
information isotopes in AI generations. We conduct experiments on ten AI models
(including GPT-4o, Claude-3.5, and DeepSeek) and four benchmark datasets in
critical domains (medical data, copyrighted books, and news). Results show that
our method can distinguish training datasets from non-training datasets with
99\% accuracy and significant evidence (p-value$<0.001$) by examining a data
entry equivalent in length to a research paper. The findings show the potential
of our work as an inclusive tool for empowering individuals, including those
without expertise in AI, to safeguard their data rights in the rapidly evolving
era of AI advancements and applications.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 07:35:59 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Tao",
"Qi",
""
],
[
"Jinhua",
"Yin",
""
],
[
"Dongqi",
"Cai",
""
],
[
"Yueqi",
"Xie",
""
],
[
"Huili",
"Wang",
""
],
[
"Zhiyang",
"Hu",
""
],
[
"Peiru",
"Yang",
""
],
[
"Guoshun",
"Nan",
""
],
[
"Zhili",
"Zhou",
""
],
[
"Shangguang",
"Wang",
""
],
[
"Lingjuan",
"Lyu",
""
],
[
"Yongfeng",
"Huang",
""
],
[
"Nicholas",
"Lane",
""
]
] |
2503.20801 | Yuntao Shou | Tao Meng, Shuo Shan, Hongen Shao, Yuntao Shou, Wei Ai, and Keqin Li | SE-GNN: Seed Expanded-Aware Graph Neural Network with Iterative
Optimization for Semi-supervised Entity Alignment | 15 pages | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Entity alignment aims to use pre-aligned seed pairs to find other equivalent
entities from different knowledge graphs (KGs) and is widely used in graph
fusion-related fields. However, as the scale of KGs increases, manually
annotating pre-aligned seed pairs becomes difficult. Existing research utilizes
entity embeddings obtained by aggregating single structural information to
identify potential seed pairs, thus reducing the reliance on pre-aligned seed
pairs. However, due to the structural heterogeneity of KGs, the quality of
potential seed pairs obtained using only a single structural information is not
ideal. In addition, although existing research improves the quality of
potential seed pairs through semi-supervised iteration, they underestimate the
impact of embedding distortion produced by noisy seed pairs on the alignment
effect. In order to solve the above problems, we propose a seed expanded-aware
graph neural network with iterative optimization for semi-supervised entity
alignment, named SE-GNN. First, we utilize the semantic attributes and
structural features of entities, combined with a conditional filtering
mechanism, to obtain high-quality initial potential seed pairs. Next, we
designed a local and global awareness mechanism. It introduces initial
potential seed pairs and combines local and global information to obtain a more
comprehensive entity embedding representation, which alleviates the impact of
KGs structural heterogeneity and lays the foundation for the optimization of
initial potential seed pairs. Then, we designed the threshold nearest neighbor
embedding correction strategy. It combines the similarity threshold and the
bidirectional nearest neighbor method as a filtering mechanism to select
iterative potential seed pairs and also uses an embedding correction strategy
to eliminate the embedding distortion.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 13:28:49 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Meng",
"Tao",
""
],
[
"Shan",
"Shuo",
""
],
[
"Shao",
"Hongen",
""
],
[
"Shou",
"Yuntao",
""
],
[
"Ai",
"Wei",
""
],
[
"Li",
"Keqin",
""
]
] |
2503.20802 | Shuhao Zhang | Shuhao Zhang, Bo Cheng, Jiale Han, Yuli Chen, Zhixuan Wu, Changbao Li,
Pingli Gu | CEFW: A Comprehensive Evaluation Framework for Watermark in Large
Language Models | null | null | null | null | cs.CR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Text watermarking provides an effective solution for identifying synthetic
text generated by large language models. However, existing techniques often
focus on satisfying specific criteria while ignoring other key aspects, lacking
a unified evaluation. To fill this gap, we propose the Comprehensive Evaluation
Framework for Watermark (CEFW), a unified framework that comprehensively
evaluates watermarking methods across five key dimensions: ease of detection,
fidelity of text quality, minimal embedding cost, robustness to adversarial
attacks, and imperceptibility to prevent imitation or forgery. By assessing
watermarks according to all these key criteria, CEFW offers a thorough
evaluation of their practicality and effectiveness. Moreover, we introduce a
simple and effective watermarking method called Balanced Watermark (BW), which
guarantees robustness and imperceptibility through balancing the way watermark
information is added. Extensive experiments show that BW outperforms existing
methods in overall performance across all evaluation dimensions. We release our
code to the community for future research.
https://github.com/DrankXs/BalancedWatermark.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 13:50:32 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Shuhao",
""
],
[
"Cheng",
"Bo",
""
],
[
"Han",
"Jiale",
""
],
[
"Chen",
"Yuli",
""
],
[
"Wu",
"Zhixuan",
""
],
[
"Li",
"Changbao",
""
],
[
"Gu",
"Pingli",
""
]
] |
2503.20803 | Bamidele Ajayi | Bamidele Ajayi, Basel Barakat and Ken McGarry | Leveraging VAE-Derived Latent Spaces for Enhanced Malware Detection with
Machine Learning Classifiers | null | null | null | null | cs.CR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper assesses the performance of five machine learning classifiers:
Decision Tree, Naive Bayes, LightGBM, Logistic Regression, and Random Forest
using latent representations learned by a Variational Autoencoder from malware
datasets. Results from the experiments conducted on different training-test
splits with different random seeds reveal that all the models perform well in
detecting malware with ensemble methods (LightGBM and Random Forest) performing
slightly better than the rest. In addition, the use of latent features reduces
the computational cost of the model and the need for extensive hyperparameter
tuning for improved efficiency of the model for deployment. Statistical tests
show that these improvements are significant, and thus, the practical relevance
of integrating latent space representation with traditional classifiers for
effective malware detection in cybersecurity is established.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 14:44:55 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ajayi",
"Bamidele",
""
],
[
"Barakat",
"Basel",
""
],
[
"McGarry",
"Ken",
""
]
] |
2503.20804 | Zelai Xu | Le Qiu, Zelai Xu, Qixin Tan, Wenhao Tang, Chao Yu, and Yu Wang | AED: Automatic Discovery of Effective and Diverse Vulnerabilities for
Autonomous Driving Policy with Large Language Models | null | null | null | null | cs.CR cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Assessing the safety of autonomous driving policy is of great importance, and
reinforcement learning (RL) has emerged as a powerful method for discovering
critical vulnerabilities in driving policies. However, existing RL-based
approaches often struggle to identify vulnerabilities that are both
effective-meaning the autonomous vehicle is genuinely responsible for the
accidents-and diverse-meaning they span various failure types. To address these
challenges, we propose AED, a framework that uses large language models (LLMs)
to automatically discover effective and diverse vulnerabilities in autonomous
driving policies. We first utilize an LLM to automatically design reward
functions for RL training. Then we let the LLM consider a diverse set of
accident types and train adversarial policies for different accident types in
parallel. Finally, we use preference-based learning to filter ineffective
accidents and enhance the effectiveness of each vulnerability. Experiments
across multiple simulated traffic scenarios and tested policies show that AED
uncovers a broader range of vulnerabilities and achieves higher attack success
rates compared with expert-designed rewards, thereby reducing the need for
manual reward engineering and improving the diversity and effectiveness of
vulnerability discovery.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 14:59:17 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Qiu",
"Le",
""
],
[
"Xu",
"Zelai",
""
],
[
"Tan",
"Qixin",
""
],
[
"Tang",
"Wenhao",
""
],
[
"Yu",
"Chao",
""
],
[
"Wang",
"Yu",
""
]
] |
2503.20805 | Alessio Catanzaro | Alessio Catanzaro, Subodh Patil and Diego Garlaschelli | A General Solution for Network Models with Pairwise Edge Coupling | null | null | null | null | cond-mat.stat-mech physics.data-an | http://creativecommons.org/licenses/by/4.0/ | Network Models with couplings between link pairs are the simplest models for
a class of networks with Higher Order interactions. In this paper we give an
analytic, general solution to this family of Random Graph Models extending
previous results obtained for specific interaction structures. We use the
Hubbard-Stratonovich transform to show that such higher order models can be
decoupled into families of effectively separable random graph models, which
might be enhanced with auxiliary, possibly multi-valued, parameters identifying
the thermodynamical phase of the system. Moreover, given the diagonalizability
of couplings between links, we carry out the full computation of the partition
functions and discuss why in some cases they can effectively reduce to the one
for the Mean-Field case.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 16:53:40 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Catanzaro",
"Alessio",
""
],
[
"Patil",
"Subodh",
""
],
[
"Garlaschelli",
"Diego",
""
]
] |
2503.20806 | Qi Zhang | Shutonu Mitra, Tomas Neguyen, Qi Zhang, Hyungmin Kim, Hossein Salemi,
Chen-Wei Chang, Fengxiu Zhang, Michin Hong, Chang-Tien Lu, Hemant Purohit,
Jin-Hee Cho | SCVI: Bridging Social and Cyber Dimensions for Comprehensive
Vulnerability Assessment | null | null | null | null | cs.CR cs.CY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The rise of cyber threats on social media platforms necessitates advanced
metrics to assess and mitigate social cyber vulnerabilities. This paper
presents the Social Cyber Vulnerability Index (SCVI), a novel framework
integrating individual-level factors (e.g., awareness, behavioral traits,
psychological attributes) and attack-level characteristics (e.g., frequency,
consequence, sophistication) for comprehensive socio-cyber vulnerability
assessment. SCVI is validated using survey data (iPoll) and textual data
(Reddit scam reports), demonstrating adaptability across modalities while
revealing demographic disparities and regional vulnerabilities. Comparative
analyses with the Common Vulnerability Scoring System (CVSS) and the Social
Vulnerability Index (SVI) show the superior ability of SCVI to capture nuanced
socio-technical risks. Monte Carlo-based weight variability analysis confirms
SCVI is robust and highlights its utility in identifying high-risk groups. By
addressing gaps in traditional metrics, SCVI offers actionable insights for
policymakers and practitioners, advancing inclusive strategies to mitigate
emerging threats such as AI-powered phishing and deepfake scams.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 19:10:34 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Mitra",
"Shutonu",
""
],
[
"Neguyen",
"Tomas",
""
],
[
"Zhang",
"Qi",
""
],
[
"Kim",
"Hyungmin",
""
],
[
"Salemi",
"Hossein",
""
],
[
"Chang",
"Chen-Wei",
""
],
[
"Zhang",
"Fengxiu",
""
],
[
"Hong",
"Michin",
""
],
[
"Lu",
"Chang-Tien",
""
],
[
"Purohit",
"Hemant",
""
],
[
"Cho",
"Jin-Hee",
""
]
] |
2503.20807 | Pin-Yu Chen | Pin-Yu Chen and Han Shen and Payel Das and Tianyi Chen | Fundamental Safety-Capability Trade-offs in Fine-tuning Large Language
Models | The first two authors contribute equally to this work and are listed
in alphabetical order | null | null | null | stat.ML cs.AI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fine-tuning Large Language Models (LLMs) on some task-specific datasets has
been a primary use of LLMs. However, it has been empirically observed that this
approach to enhancing capability inevitably compromises safety, a phenomenon
also known as the safety-capability trade-off in LLM fine-tuning. This paper
presents a theoretical framework for understanding the interplay between safety
and capability in two primary safety-aware LLM fine-tuning strategies,
providing new insights into the effects of data similarity, context overlap,
and alignment loss landscape. Our theoretical results characterize the
fundamental limits of the safety-capability trade-off in LLM fine-tuning, which
are also validated by numerical experiments.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2025 20:41:57 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Pin-Yu",
""
],
[
"Shen",
"Han",
""
],
[
"Das",
"Payel",
""
],
[
"Chen",
"Tianyi",
""
]
] |
2503.20808 | Xiaoming Qi | Xiaoming Qi and Jingyang Zhang and Huazhu Fu and Guanyu Yang and Shuo
Li and Yueming Jin | Dynamic Allocation Hypernetwork with Adaptive Model Recalibration for
Federated Continual Learning | null | Information Processing in Medical Imaging(IPMI)2025 | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Federated continual learning (FCL) offers an emerging pattern to facilitate
the applicability of federated learning (FL) in real-world scenarios, where
tasks evolve dynamically and asynchronously across clients, especially in
medical scenario. Existing server-side FCL methods in nature domain construct a
continually learnable server model by client aggregation on all-involved tasks.
However, they are challenged by: (1) Catastrophic forgetting for previously
learned tasks, leading to error accumulation in server model, making it
difficult to sustain comprehensive knowledge across all tasks. (2) Biased
optimization due to asynchronous tasks handled across different clients,
leading to the collision of optimization targets of different clients at the
same time steps. In this work, we take the first step to propose a novel
server-side FCL pattern in medical domain, Dynamic Allocation Hypernetwork with
adaptive model recalibration (FedDAH). It is to facilitate collaborative
learning under the distinct and dynamic task streams across clients. To
alleviate the catastrophic forgetting, we propose a dynamic allocation
hypernetwork (DAHyper) where a continually updated hypernetwork is designed to
manage the mapping between task identities and their associated model
parameters, enabling the dynamic allocation of the model across clients. For
the biased optimization, we introduce a novel adaptive model recalibration
(AMR) to incorporate the candidate changes of historical models into current
server updates, and assign weights to identical tasks across different time
steps based on the similarity for continual optimization. Extensive experiments
on the AMOS dataset demonstrate the superiority of our FedDAH to other FCL
methods on sites with different task streams. The code is
available:https://github.com/jinlab-imvr/FedDAH.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 00:17:47 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Qi",
"Xiaoming",
""
],
[
"Zhang",
"Jingyang",
""
],
[
"Fu",
"Huazhu",
""
],
[
"Yang",
"Guanyu",
""
],
[
"Li",
"Shuo",
""
],
[
"Jin",
"Yueming",
""
]
] |
2503.20809 | Huaiqian Li | Huaiqian Li, Bingyao Wu | Limiting Behaviors of Besov Seminorms for Dunkl Operators | 29 pp, 1 figuer; submitted | null | null | null | math.FA math.AP math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As $s\rightarrow0^+$, we establish limiting formulas of Besov seminorms and
nonlocal perimeters associated with the Dunkl operator, a (nonlocal)
differential-difference operator parameterized by multiplicity functions and
finite reflection groups. Our results are further developments of both the
Maz'ya--Shaposhnikova limiting formula for the Gagliardo seminorm and the
asymptotic behavior of the (relative) fractional $s$-perimeter. The main
contribution is twofold. On the one hand, to establish our dimension-free
Maz'ya--Shaposhnikova limiting formula, we develop a simplified approach which
do not depend on the density property of the corresponding Besov space and
turns out to be quite robust. On the other hand, to derive the limiting formula
of our nonlocal perimeter, we do not demand additional regularity on the
(topological) boundary of the domain, and to obtain the converse assertion, our
assumption on the boundary regularity of the domain, which allows for fractals,
is much weaker than those in existing literatures.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 05:35:09 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Li",
"Huaiqian",
""
],
[
"Wu",
"Bingyao",
""
]
] |
2503.20810 | Robert Alicki | Robert Alicki | Comment on " Second Law of Thermodynamics without Einstein Relation'',
arXiv:2405.17142 | 3 pages, comment on arXiv:2405.17142 | null | null | null | cond-mat.stat-mech quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is argued the the idea of a single temperature-like variable, introduced
in [1], which enters a generalized second law for Markovian open system in
non-equilibrium environment is not sufficient for a consistent and useful
thermodynamic formalism. The origin of the Second Law and ``local
temperatures'' in Markovian theory of Quantum Open Systems is revisited and
illustrated by simple examples.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 08:56:25 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Alicki",
"Robert",
""
]
] |
2503.20811 | Steven Louis | Steven Louis, Hannah Bradley, Vasyl Tyberkevych | Equivalent Electric Model of a Macrospin | 7 pages, 2 figures, preprint only | null | null | null | cond-mat.mes-hall cond-mat.other | http://creativecommons.org/licenses/by/4.0/ | Dynamics of a ferromagnetic macrospin (e.g., a free layer of a magnetic
tunnel junction (MTJ)) can be described in terms of equivalent capacitor charge
$Q$ and inductor flux $\Phi$, in a manner similar to a standard electric LC
circuit, but with strongly nonlinear and coupled capacitance and inductance.
This description allows for the inclusion of Gilbert damping and spin transfer
torques and yields a relatively simple equivalent electric circuit, which can
be easily modeled in LTspice or other electrical engineering software. It
allows one to easily simulate advanced electrical circuits containing MTJs and
conventional electronic components in standard simulation software.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 09:45:42 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Louis",
"Steven",
""
],
[
"Bradley",
"Hannah",
""
],
[
"Tyberkevych",
"Vasyl",
""
]
] |
2503.20812 | Mesfin Asfaw Taye Dr. | Mesfin Taye | Thermodynamic Relations in Time-Dependent Temperatures and Time-Varying
Forces: A Study of Entropy Production and Extraction Rates Governed by the
Fokker-Planck Equation in Active Brownian Systems | 14 pages, 7 figures | null | null | null | cond-mat.stat-mech | http://creativecommons.org/publicdomain/zero/1.0/ | Most previous works have focused on systems exposed to spatially varying
forces and temperatures without looking at the role of time-dependent
temperature or force fluctuations in active matter, molecular motors, and
Brownian heat engines. However time varying temperature and forces regulate the
efficiency and dissipation of molecular motors like kinesin, dynein, and
myosin. Cooling of the system may slow ATP hydrolysis, reducing the transport
efficiency while increasing dissipation. The performance of a Brownian heat
engine can be affected by the cooling conditions. On other hand it is vital to
consider the time-dependent force in ATP depletion, weakening optical tweezers,
and energy loss in damped systems since this directly affects the motor
dynamics and non-equilibrium transport. Thus, this time-dependent temperature
or force considerably affects the transport efficiency, energy dissipation, and
entropy production in non-equilibrium systems. In this work we examines the
impact of time-varying temperature and force on the thermodynamic features of
the system. By deriving analytical expressions for entropy production, entropy
extraction, and dissipation rates, we extend the existing theoretical
frameworks by considering, a force or temperature that varies exponentially,
linearly, and quadratically.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 12:14:43 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Taye",
"Mesfin",
""
]
] |
2503.20813 | Steven Louis | Steven Louis, Hannah Bradley, Artem Litvinenko, Vasyl Tyberkevych | A Physics-Based Circuit Model for Magnetic Tunnel Junctions | 5 pages, 4 figures | null | null | null | cond-mat.mes-hall cond-mat.other | http://creativecommons.org/licenses/by/4.0/ | This work presents an equivalent circuit model for Magnetic Tunnel Junctions
(MTJs) that accurately captures their magnetization dynamics and electrical
behavior. Implemented in LTspice, the model is validated against direct
numerical solutions of the Landau-Lifshitz-Gilbert-Slonczewski (LLGS) equation.
It effectively simulates essential spintronic phenomena, including
ferromagnetic resonance, field- and spin-torque-induced switching, and
spin-torque-induced oscillations. Simulation results demonstrate strong
agreement between LTspice and LLGS solutions, confirming the model accuracy and
utility for efficient circuit-level analysis of spintronic devices. The ability
to incorporate time-dependent magnetic fields and voltage inputs makes the
proposed model suitable for diverse applications such as neuromorphic
computing, microwave signal processing, and spintronic memory technologies. By
providing a computationally efficient yet physically accurate circuit
representation, this work enables seamless integration of MTJs into larger
electronic systems, potentially accelerating the development of advanced
spintronic circuit architectures.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 12:58:01 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Louis",
"Steven",
""
],
[
"Bradley",
"Hannah",
""
],
[
"Litvinenko",
"Artem",
""
],
[
"Tyberkevych",
"Vasyl",
""
]
] |
2503.20814 | Nasr Ahmed | Gunjan Varshney, Anirudh Pradhan and Nasr Ahmed | A novel cosmic framework of interdependent dark matter and Holographic
Dark Energy within the Bianchi type-V universe | null | null | null | null | gr-qc | http://creativecommons.org/licenses/by/4.0/ | We study the anisotropic and homogeneous Bianchi type-V Universe with
holographic dark energy (HDE) and interacting dark matter (DM). The solution
for the field equations have been obtained for a certain form of the
deceleration parameter. As for $\Lambda$CDM, we show that the coincidence
problem disappears for a specific choice of the dark matter-holographic dark
energy interaction. In this study, the observational data combination of OHD
and JLA (Yu et al., Astrophys. J. 856 (2018) 3), where $q_0$ = $-0.52$ and
$H_0$ = $69.2$, have been taken into consideration. We also found that the
anisotropy of the expansion achieves isotropy after some finite time. We have
also explained the physical and geometrical aspects of the model. The physical
acceptability and stability of the model have been examined.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 16:38:34 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Varshney",
"Gunjan",
""
],
[
"Pradhan",
"Anirudh",
""
],
[
"Ahmed",
"Nasr",
""
]
] |
2503.20815 | Lipei Zhang | Lipei Zhang, Rui Sun, Zhongying Deng, Yanqi Cheng, Carola-Bibiane
Sch\"onlieb, Angelica I Aviles-Rivero | D2SA: Dual-Stage Distribution and Slice Adaptation for Efficient
Test-Time Adaptation in MRI Reconstruction | 9 pages, 10 pages (supplementary) | null | null | null | eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Variations in Magnetic resonance imaging (MRI) scanners and acquisition
protocols cause distribution shifts that degrade reconstruction performance on
unseen data. Test-time adaptation (TTA) offers a promising solution to address
this discrepancies. However, previous single-shot TTA approaches are
inefficient due to repeated training and suboptimal distributional models.
Self-supervised learning methods are also limited by scarce date scenarios. To
address these challenges, we propose a novel Dual-Stage Distribution and Slice
Adaptation (D2SA) via MRI implicit neural representation (MR-INR) to improve
MRI reconstruction performance and efficiency, which features two stages. In
the first stage, an MR-INR branch performs patient-wise distribution adaptation
by learning shared representations across slices and modelling patient-specific
shifts with mean and variance adjustments. In the second stage, single-slice
adaptation refines the output from frozen convolutional layers with a learnable
anisotropic diffusion module, preventing over-smoothing and reducing
computation. Experiments across four MRI distribution shifts demonstrate that
our method can integrate well with various self-supervised learning (SSL)
framework, improving performance and accelerating convergence under diverse
conditions.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 18:06:21 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Lipei",
""
],
[
"Sun",
"Rui",
""
],
[
"Deng",
"Zhongying",
""
],
[
"Cheng",
"Yanqi",
""
],
[
"Schönlieb",
"Carola-Bibiane",
""
],
[
"Aviles-Rivero",
"Angelica I",
""
]
] |
2503.20816 | Uri Zvi | Uri Zvi, Shivam Mundhra, David Ovetsky, Qing Chen, Aidan R. Jones,
Stella Wang, Maria Roman, Michele Ferro, Kunle Odunsi, Marina C. Garassino,
Michael E. Flatte', Melody Swartz, Denis R. Candido, Aaron Esser-Kahn, and
Peter C. Maurer | Probing cellular activity via charge-sensitive quantum nanoprobes | 15 pages of main text and references, including 4 main figures. 8
pages of extended data (7 figures and 1 table). 6 pages of SI. Total: 29
pages | null | null | null | physics.app-ph cond-mat.mes-hall physics.bio-ph physics.optics quant-ph | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Nitrogen-vacancy (NV) based quantum sensors hold great potential for
real-time single-cell sensing with far-reaching applications in fundamental
biology and medical diagnostics. Although highly sensitive, the mapping of
quantum measurements onto cellular physiological states has remained an
exceptional challenge. Here we introduce a novel quantum sensing modality
capable of detecting changes in cellular activity. Our approach is based on the
detection of environment-induced charge depletion within an individual particle
that, owing to a previously unaccounted transverse dipole term, induces
systematic shifts in the zero-field splitting (ZFS). Importantly, these
charge-induced shifts serve as a reliable indicator for lipopolysaccharide
(LPS)-mediated inflammatory response in macrophages. Furthermore, we
demonstrate that surface modification of our diamond nanoprobes effectively
suppresses these environment-induced ZFS shifts, providing an important tool
for differentiating electrostatic shifts caused by the environment from other
unrelated effects, such as temperature variations. Notably, this surface
modification also leads to significant reductions in particle-induced toxicity
and inflammation. Our findings shed light on systematic drifts and sensitivity
limits of NV spectroscopy in a biological environment with ramification on the
critical discussion surrounding single-cell thermogenesis. Notably, this work
establishes the foundation for a novel sensing modality capable of probing
complex cellular processes through straightforward physical measurements.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 18:11:43 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zvi",
"Uri",
""
],
[
"Mundhra",
"Shivam",
""
],
[
"Ovetsky",
"David",
""
],
[
"Chen",
"Qing",
""
],
[
"Jones",
"Aidan R.",
""
],
[
"Wang",
"Stella",
""
],
[
"Roman",
"Maria",
""
],
[
"Ferro",
"Michele",
""
],
[
"Odunsi",
"Kunle",
""
],
[
"Garassino",
"Marina C.",
""
],
[
"Flatte'",
"Michael E.",
""
],
[
"Swartz",
"Melody",
""
],
[
"Candido",
"Denis R.",
""
],
[
"Esser-Kahn",
"Aaron",
""
],
[
"Maurer",
"Peter C.",
""
]
] |
2503.20817 | Qiang Wang | Zhenya Zang, David A Dorward, Katherine E Quiohilag, Andrew DJ Wood,
James R Hopgood, Ahsan R Akram, Qiang Wang | Label-free pathological subtyping of non-small cell lung cancer using
deep classification and virtual immunohistochemical staining | 18 pages, 5 figures, 1 table | null | null | null | q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | The differentiation between pathological subtypes of non-small cell lung
cancer (NSCLC) is an essential step in guiding treatment options and prognosis.
However, current clinical practice relies on multi-step staining and labelling
processes that are time-intensive and costly, requiring highly specialised
expertise. In this study, we propose a label-free methodology that facilitates
autofluorescence imaging of unstained NSCLC samples and deep learning (DL)
techniques to distinguish between non-cancerous tissue, adenocarcinoma (AC),
squamous cell carcinoma (SqCC), and other subtypes (OS). We conducted DL-based
classification and generated virtual immunohistochemical (IHC) stains,
including thyroid transcription factor-1 (TTF-1) for AC and p40 for SqCC, and
evaluated these methods using two types of autofluorescence imaging: intensity
imaging and lifetime imaging. The results demonstrate the exceptional ability
of this approach for NSCLC subtype differentiation, achieving an area under the
curve above 0.981 and 0.996 for binary- and multi-class classification.
Furthermore, this approach produces clinical-grade virtual IHC staining which
was blind-evaluated by three experienced thoracic pathologists. Our label-free
NSCLC subtyping approach enables rapid and accurate diagnosis without
conventional tissue processing and staining. Both strategies can significantly
accelerate diagnostic workflows and support efficient lung cancer diagnosis,
without compromising clinical decision-making.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 19:12:56 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zang",
"Zhenya",
""
],
[
"Dorward",
"David A",
""
],
[
"Quiohilag",
"Katherine E",
""
],
[
"Wood",
"Andrew DJ",
""
],
[
"Hopgood",
"James R",
""
],
[
"Akram",
"Ahsan R",
""
],
[
"Wang",
"Qiang",
""
]
] |
2503.20818 | Lucas Williams | Lucas Williams | The Pontryagin-Thom theorem for families of framed manifolds | null | null | null | null | math.AT | http://creativecommons.org/licenses/by/4.0/ | The Pontryagin-Thom theorem gives an isomorphism from the cobordism group of
framed $n$-manifolds to the $n$th stable homotopy group of the sphere spectrum.
In this paper, we prove the generalization of the Pontryagin-Thom theorem for
families of framed equivariant manifolds parameterized over a compact base
space.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 20:21:16 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Williams",
"Lucas",
""
]
] |
2503.20819 | Paraskevi Valergaki | Paraskevi Valergaki, Antonis Argyros, Giorgos Giannakakis, Anastasios
Roussos | Reflections on Diversity: A Real-time Virtual Mirror for Inclusive 3D
Face Transformations | null | null | null | null | cs.GR eess.IV | http://creativecommons.org/licenses/by/4.0/ | Real-time 3D face manipulation has significant applications in virtual
reality, social media and human-computer interaction. This paper introduces a
novel system, which we call Mirror of Diversity (MOD), that combines Generative
Adversarial Networks (GANs) for texture manipulation and 3D Morphable Models
(3DMMs) for facial geometry to achieve realistic face transformations that
reflect various demographic characteristics, emphasizing the beauty of
diversity and the universality of human features. As participants sit in front
of a computer monitor with a camera positioned above, their facial
characteristics are captured in real time and can further alter their digital
face reconstruction with transformations reflecting different demographic
characteristics, such as gender and ethnicity (e.g., a person from Africa,
Asia, Europe). Another feature of our system, which we call Collective Face,
generates an averaged face representation from multiple participants' facial
data. A comprehensive evaluation protocol is implemented to assess the realism
and demographic accuracy of the transformations. Qualitative feedback is
gathered through participant questionnaires, which include comparisons of MOD
transformations with similar filters on platforms like Snapchat and TikTok.
Additionally, quantitative analysis is conducted using a pretrained
Convolutional Neural Network that predicts gender and ethnicity, to validate
the accuracy of demographic transformations.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 20:31:35 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Valergaki",
"Paraskevi",
""
],
[
"Argyros",
"Antonis",
""
],
[
"Giannakakis",
"Giorgos",
""
],
[
"Roussos",
"Anastasios",
""
]
] |
2503.20820 | Tianze Chen | Tianze Chen, Ricardo Frumento, Giulia Pagnanelli, Gianmarco Cei, Villa
Keth, Shahadding Gafarov, Jian Gong, Zihe Ye, Marco Baracca, Salvatore
D'Avella, Matteo Bianchi and Yu Sun | Benchmarking Multi-Object Grasping | This paper contains 11 pages and 5 figures. This paper is under
review of a robotics journal | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In this work, we describe a multi-object grasping benchmark to evaluate the
grasping and manipulation capabilities of robotic systems in both pile and
surface scenarios. The benchmark introduces three robot multi-object grasping
benchmarking protocols designed to challenge different aspects of robotic
manipulation. These protocols are: 1) the Only-Pick-Once protocol, which
assesses the robot's ability to efficiently pick multiple objects in a single
attempt; 2) the Accurate pick-trnsferring protocol, which evaluates the robot's
capacity to selectively grasp and transport a specific number of objects from a
cluttered environment; and 3) the Pick-transferring-all protocol, which
challenges the robot to clear an entire scene by sequentially grasping and
transferring all available objects. These protocols are intended to be adopted
by the broader robotics research community, providing a standardized method to
assess and compare robotic systems' performance in multi-object grasping tasks.
We establish baselines for these protocols using standard planning and
perception algorithms on a Barrett hand, Robotiq parallel jar gripper, and the
Pisa/IIT Softhand-2, which is a soft underactuated robotic hand. We discuss the
results in relation to human performance in similar tasks we well.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 21:52:55 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Tianze",
""
],
[
"Frumento",
"Ricardo",
""
],
[
"Pagnanelli",
"Giulia",
""
],
[
"Cei",
"Gianmarco",
""
],
[
"Keth",
"Villa",
""
],
[
"Gafarov",
"Shahadding",
""
],
[
"Gong",
"Jian",
""
],
[
"Ye",
"Zihe",
""
],
[
"Baracca",
"Marco",
""
],
[
"D'Avella",
"Salvatore",
""
],
[
"Bianchi",
"Matteo",
""
],
[
"Sun",
"Yu",
""
]
] |
2503.20821 | Rajvardhan Oak | Rajvardhan Oak, Zubair Shafiq | "Hello, is this Anna?": A First Look at Pig-Butchering Scams | null | null | null | null | cs.CR cs.CY | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Pig-butchering scams, or Sha Zhu Pan, have emerged as a complex form of
cyber-enabled financial fraud that combines elements of romance, investment
fraud, and advanced social engineering tactics to systematically exploit
victims. In this paper, we present the first qualitative analysis of
pig-butchering scams, informed by in-depth semi-structured interviews with N=26
victims. We capture nuanced, first-hand accounts from victims across multiple
regions, providing insight into the lifecycle of pig-butchering scams and the
complex emotional and financial manipulation involved. We systematically
analyze each phase of the scam, revealing that perpetrators employ tactics such
as staged trust-building, fraudulent financial platforms, fabricated investment
returns, and repeated high-pressure tactics, all designed to exploit victims'
trust and financial resources over extended periods. Our findings reveal an
organized scam lifecycle characterized by emotional manipulation, staged
financial exploitation, and persistent re-engagement efforts that amplify
victim losses. We also find complex psychological and financial impacts on
victims, including heightened vulnerability to secondary scams. Finally, we
propose actionable intervention points for social media and financial platforms
to curb the prevalence of these scams and highlight the need for
non-stigmatizing terminology to encourage victims to report and seek
assistance.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 23:15:48 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Oak",
"Rajvardhan",
""
],
[
"Shafiq",
"Zubair",
""
]
] |
2503.20822 | Qi Zhao | Qi Zhao, Xingyu Ni, Ziyu Wang, Feng Cheng, Ziyan Yang, Lu Jiang, Bohan
Wang | Synthetic Video Enhances Physical Fidelity in Video Synthesis | null | null | null | null | eess.IV cs.AI cs.GR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We investigate how to enhance the physical fidelity of video generation
models by leveraging synthetic videos derived from computer graphics pipelines.
These rendered videos respect real-world physics, such as maintaining 3D
consistency, and serve as a valuable resource that can potentially improve
video generation models. To harness this potential, we propose a solution that
curates and integrates synthetic data while introducing a method to transfer
its physical realism to the model, significantly reducing unwanted artifacts.
Through experiments on three representative tasks emphasizing physical
consistency, we demonstrate its efficacy in enhancing physical fidelity. While
our model still lacks a deep understanding of physics, our work offers one of
the first empirical demonstrations that synthetic video enhances physical
fidelity in video synthesis. Website: https://kevinz8866.github.io/simulation/
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 00:45:07 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhao",
"Qi",
""
],
[
"Ni",
"Xingyu",
""
],
[
"Wang",
"Ziyu",
""
],
[
"Cheng",
"Feng",
""
],
[
"Yang",
"Ziyan",
""
],
[
"Jiang",
"Lu",
""
],
[
"Wang",
"Bohan",
""
]
] |
2503.20823 | Joon Hyun Jeong | Joonhyun Jeong, Seyun Bae, Yeonsung Jung, Jaeryong Hwang, Eunho Yang | Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with
Out-of-Distribution Strategy | Accepted at CVPR2025 | null | null | null | cs.CR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Despite the remarkable versatility of Large Language Models (LLMs) and
Multimodal LLMs (MLLMs) to generalize across both language and vision tasks,
LLMs and MLLMs have shown vulnerability to jailbreaking, generating textual
outputs that undermine safety, ethical, and bias standards when exposed to
harmful or sensitive inputs. With the recent advancement of safety alignment
via preference-tuning from human feedback, LLMs and MLLMs have been equipped
with safety guardrails to yield safe, ethical, and fair responses with regard
to harmful inputs. However, despite the significance of safety alignment,
research on the vulnerabilities remains largely underexplored. In this paper,
we investigate the unexplored vulnerability of the safety alignment, examining
its ability to consistently provide safety guarantees for
out-of-distribution(OOD)-ifying harmful inputs that may fall outside the
aligned data distribution. Our key observation is that OOD-ifying the vanilla
harmful inputs highly increases the uncertainty of the model to discern the
malicious intent within the input, leading to a higher chance of being
jailbroken. Exploiting this vulnerability, we propose JOOD, a new Jailbreak
framework via OOD-ifying inputs beyond the safety alignment. We explore various
off-the-shelf visual and textual transformation techniques for OOD-ifying the
harmful inputs. Notably, we observe that even simple mixing-based techniques
such as image mixup prove highly effective in increasing the uncertainty of the
model, thereby facilitating the bypass of the safety alignment. Experiments
across diverse jailbreak scenarios demonstrate that JOOD effectively jailbreaks
recent proprietary LLMs and MLLMs such as GPT-4 and o1 with high attack success
rate, which previous attack approaches have consistently struggled to
jailbreak. Code is available at https://github.com/naver-ai/JOOD.
| [
{
"version": "v1",
"created": "Wed, 26 Mar 2025 01:25:24 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Jeong",
"Joonhyun",
""
],
[
"Bae",
"Seyun",
""
],
[
"Jung",
"Yeonsung",
""
],
[
"Hwang",
"Jaeryong",
""
],
[
"Yang",
"Eunho",
""
]
] |
Subsets and Splits