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
2025-04-25 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.09577 | Michael Wozniak | Michael P. Wozniak | Unification of Consensus-Based Multi-Objective Optimization and
Multi-Robot Path Planning | null | null | null | null | cs.MA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Multi-agent systems seeking consensus may also have other objective functions
to optimize, requiring the research of multi-objective optimization in
consensus. Several recent publications have explored this domain using various
methods such as weighted-sum optimization and penalization methods. This paper
reviews the state of the art for consensus-based multi-objective optimization,
poses a multi-agent lunar rover exploration problem seeking consensus and
maximization of explored area, and achieves optimal edge weights and steering
angles by applying SQP algorithms.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 13:56:54 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wozniak",
"Michael P.",
""
]
] |
2504.09578 | Lucas C\'eleri | T. H. Moreira and L. C. C\'eleri | Graviton induced decoherence of a composite particle | 18 pages, 1 figures. Chapter of the Proceedings for the III
International Workshop on Quantum Nonstationary Systems (eds. Alexandre
Dodonov and Lucas Chibebe C\'eleri) | null | null | null | quant-ph gr-qc | http://creativecommons.org/licenses/by/4.0/ | We consider a composite particle, whose internal degrees of freedom are
described by quantum mechanics, interacting with the quantum gravitational
field in the linear approximation. Dechorence induced by the quantum
fluctuations of the gravitational waves are detailed discussed.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 13:57:17 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Moreira",
"T. H.",
""
],
[
"Céleri",
"L. C.",
""
]
] |
2504.09581 | Lucas C\'eleri | M. Basso, J. Maziero and L. C. C\'eleri | Relativistic approach to thermodynamic irreversibility | 14 pages, 3 figures. Chapter of the Proceedings for the III
International Workshop on Quantum Nonstationary Systems (eds. Alexandre
Dodonov and Lucas Chibebe C\'eleri) | null | null | null | quant-ph gr-qc | http://creativecommons.org/licenses/by/4.0/ | We consider a localized quantum system living in a curved spacetimes. By
translating into this scenario the paradgmatic two-point measument scheme in
quantum statistical mechanics we are able to prove a relativistic version of
the quantum flutuation theorem. One of the consequences of such a theorem is
that entropy production is deeply observer dependent. The main ideas of this
chapter are published in [M. L. W. Basso, J. Maziero and L. C. C\'eleri. The
irreversibility of relativistic time-dilation. Class. Quantum Grav.
\textbf{40}, 195001 (2023); M. L. W. Basso, J. Maziero and L. C. C\'eleri.
Quantum detailed fluctuation theorem in curved spacetimes: the observer
dependent nature of entropy production. Phys. Rev. Lett. \textbf{134}, 050406
(2025)].
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:04:22 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Basso",
"M.",
""
],
[
"Maziero",
"J.",
""
],
[
"Céleri",
"L. C.",
""
]
] |
2504.09582 | Christos Theodoropoulos | Christos Theodoropoulos, Andrei Catalin Coman, James Henderson,
Marie-Francine Moens | Reduction of Supervision for Biomedical Knowledge Discovery | Published as part of the PhD dissertation: Theodoropoulos, Christos,
Marie-Francine Moens, and Matthew Blaschko. "Deep Learning Models for the
Extraction of Knowledge from Text." (2025) | null | null | null | cs.AI cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Knowledge discovery is hindered by the increasing volume of publications and
the scarcity of extensive annotated data. To tackle the challenge of
information overload, it is essential to employ automated methods for knowledge
extraction and processing. Finding the right balance between the level of
supervision and the effectiveness of models poses a significant challenge.
While supervised techniques generally result in better performance, they have
the major drawback of demanding labeled data. This requirement is
labor-intensive and time-consuming and hinders scalability when exploring new
domains. In this context, our study addresses the challenge of identifying
semantic relationships between biomedical entities (e.g., diseases, proteins)
in unstructured text while minimizing dependency on supervision. We introduce a
suite of unsupervised algorithms based on dependency trees and attention
mechanisms and employ a range of pointwise binary classification methods.
Transitioning from weakly supervised to fully unsupervised settings, we assess
the methods' ability to learn from data with noisy labels. The evaluation on
biomedical benchmark datasets explores the effectiveness of the methods. Our
approach tackles a central issue in knowledge discovery: balancing performance
with minimal supervision. By gradually decreasing supervision, we assess the
robustness of pointwise binary classification techniques in handling noisy
labels, revealing their capability to shift from weakly supervised to entirely
unsupervised scenarios. Comprehensive benchmarking offers insights into the
effectiveness of these techniques, suggesting an encouraging direction toward
adaptable knowledge discovery systems, representing progress in creating
data-efficient methodologies for extracting useful insights when annotated data
is limited.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:05:40 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Theodoropoulos",
"Christos",
""
],
[
"Coman",
"Andrei Catalin",
""
],
[
"Henderson",
"James",
""
],
[
"Moens",
"Marie-Francine",
""
]
] |
2504.09583 | Fei Lin | Fei Lin, Yonglin Tian, Tengchao Zhang, Jun Huang, Sangtian Guan,
Fei-Yue Wang | AirVista-II: An Agentic System for Embodied UAVs Toward Dynamic Scene
Semantic Understanding | null | null | null | null | cs.RO cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unmanned Aerial Vehicles (UAVs) are increasingly important in dynamic
environments such as logistics transportation and disaster response. However,
current tasks often rely on human operators to monitor aerial videos and make
operational decisions. This mode of human-machine collaboration suffers from
significant limitations in efficiency and adaptability. In this paper, we
present AirVista-II -- an end-to-end agentic system for embodied UAVs, designed
to enable general-purpose semantic understanding and reasoning in dynamic
scenes. The system integrates agent-based task identification and scheduling,
multimodal perception mechanisms, and differentiated keyframe extraction
strategies tailored for various temporal scenarios, enabling the efficient
capture of critical scene information. Experimental results demonstrate that
the proposed system achieves high-quality semantic understanding across diverse
UAV-based dynamic scenarios under a zero-shot setting.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:06:50 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Lin",
"Fei",
""
],
[
"Tian",
"Yonglin",
""
],
[
"Zhang",
"Tengchao",
""
],
[
"Huang",
"Jun",
""
],
[
"Guan",
"Sangtian",
""
],
[
"Wang",
"Fei-Yue",
""
]
] |
2504.09584 | William Buchanan Prof | V\'ictor Duarte Melo and William J. Buchanan | Eccfrog512ck2: An Enhanced 512-bit Weierstrass Elliptic Curve | null | null | null | null | cs.CR | http://creativecommons.org/licenses/by/4.0/ | Whilst many key exchange and digital signature methods use the NIST P256
(secp256r1) and secp256k1 curves, there is often a demand for increased
security. With these curves, we have a 128-bit security. These security levels
can be increased to 256-bit security with NIST P-521 Curve 448 and
Brainpool-P512. This paper outlines a new curve - Eccfrog512ck2 - and which
provides 256-bit security and enhanced performance over NIST P-521. Along with
this, it has side-channel resistance and is designed to avoid weaknesses such
as related to the MOV attack. It shows that Eccfrog512ck2 can have a 61.5%
speed-up on scalar multiplication and a 33.3% speed-up on point generation over
the NIST P-521 curve.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:07:21 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Melo",
"Víctor Duarte",
""
],
[
"Buchanan",
"William J.",
""
]
] |
2504.09585 | Swanhild Bernstein | Swanhild Bernstein, Amedeo Altavilla and Martha Lina Zimmermann | Conjugate $(1/q, q)$-harmonic Polynomials in $q$-Clifford Analysis | 15 pages | null | null | null | math.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We consider the problem of constructing a conjugate $(1/q, q)$-harmonic
homogeneous polynomial $V_k$ of degree $k$ to a given $(1/q, q)$-harmonic
homogeneous polynomial $U_k$ of degree $k.$ The conjugated harmonic polynomials
$V_k$ and $U_k$ are associated to the $(1/q, q)$-mono\-genic polynomial $F =
U_k + \overline{e}_0V. $ We investigate conjugate $(1/q, q)$-harmonic
homogeneous polynomials in the setting of $q$-Clifford analysis. Starting from
a given $(1/q, q)$-harmonic polynomial $U_k$ of degree $k$, we construct its
conjugate counterpart $V_k$, such that the Clifford-valued polynomial $F = U_k
+ e_0 V_k$ is $(1/q, q)$-monogenic, i.e., a null solution of a generalized
$q$-Dirac operator. Our construction relies on a combination of Jackson-type
integration, Fischer decomposition, and the resolution of a $q$-Poisson
equation. We further establish existence and uniqueness results, and provide
explicit representations for conjugate pairs, particularly when $U_k$ is
real-valued.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:07:41 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Bernstein",
"Swanhild",
""
],
[
"Altavilla",
"Amedeo",
""
],
[
"Zimmermann",
"Martha Lina",
""
]
] |
2504.09586 | Zuoli Tang | Zuoli Tang, Junjie Ou, Kaiqin Hu, Chunwei Wu, Zhaoxin Huan, Chilin Fu,
Xiaolu Zhang, Jun Zhou, Chenliang Li | Short-Path Prompting in LLMs: Analyzing Reasoning Instability and
Solutions for Robust Performance | Under review | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Recent years have witnessed significant progress in large language models'
(LLMs) reasoning, which is largely due to the chain-of-thought (CoT)
approaches, allowing models to generate intermediate reasoning steps before
reaching the final answer. Building on these advances, state-of-the-art LLMs
are instruction-tuned to provide long and detailed CoT pathways when responding
to reasoning-related questions. However, human beings are naturally cognitive
misers and will prompt language models to give rather short responses, thus
raising a significant conflict with CoT reasoning. In this paper, we delve into
how LLMs' reasoning performance changes when users provide short-path prompts.
The results and analysis reveal that language models can reason effectively and
robustly without explicit CoT prompts, while under short-path prompting, LLMs'
reasoning ability drops significantly and becomes unstable, even on
grade-school problems. To address this issue, we propose two approaches: an
instruction-guided approach and a fine-tuning approach, both designed to
effectively manage the conflict. Experimental results show that both methods
achieve high accuracy, providing insights into the trade-off between
instruction adherence and reasoning accuracy in current models.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:12:14 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Tang",
"Zuoli",
""
],
[
"Ou",
"Junjie",
""
],
[
"Hu",
"Kaiqin",
""
],
[
"Wu",
"Chunwei",
""
],
[
"Huan",
"Zhaoxin",
""
],
[
"Fu",
"Chilin",
""
],
[
"Zhang",
"Xiaolu",
""
],
[
"Zhou",
"Jun",
""
],
[
"Li",
"Chenliang",
""
]
] |
2504.09588 | Zhicong Wu | Zhicong Wu, Hongbin Xu, Gang Xu, Ping Nie, Zhixin Yan, Jinkai Zheng,
Liangqiong Qu, Ming Li, Liqiang Nie | TextSplat: Text-Guided Semantic Fusion for Generalizable Gaussian
Splatting | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advancements in Generalizable Gaussian Splatting have enabled robust
3D reconstruction from sparse input views by utilizing feed-forward Gaussian
Splatting models, achieving superior cross-scene generalization. However, while
many methods focus on geometric consistency, they often neglect the potential
of text-driven guidance to enhance semantic understanding, which is crucial for
accurately reconstructing fine-grained details in complex scenes. To address
this limitation, we propose TextSplat--the first text-driven Generalizable
Gaussian Splatting framework. By employing a text-guided fusion of diverse
semantic cues, our framework learns robust cross-modal feature representations
that improve the alignment of geometric and semantic information, producing
high-fidelity 3D reconstructions. Specifically, our framework employs three
parallel modules to obtain complementary representations: the Diffusion Prior
Depth Estimator for accurate depth information, the Semantic Aware Segmentation
Network for detailed semantic information, and the Multi-View Interaction
Network for refined cross-view features. Then, in the Text-Guided Semantic
Fusion Module, these representations are integrated via the text-guided and
attention-based feature aggregation mechanism, resulting in enhanced 3D
Gaussian parameters enriched with detailed semantic cues. Experimental results
on various benchmark datasets demonstrate improved performance compared to
existing methods across multiple evaluation metrics, validating the
effectiveness of our framework. The code will be publicly available.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:14:10 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wu",
"Zhicong",
""
],
[
"Xu",
"Hongbin",
""
],
[
"Xu",
"Gang",
""
],
[
"Nie",
"Ping",
""
],
[
"Yan",
"Zhixin",
""
],
[
"Zheng",
"Jinkai",
""
],
[
"Qu",
"Liangqiong",
""
],
[
"Li",
"Ming",
""
],
[
"Nie",
"Liqiang",
""
]
] |
2504.09589 | Xiaoyao Yu | Xiaoyao Yu, Talal Rahwan, Tao Jia | Knowledge Independence Breeds Disruption but Limits Recognition | 23 pages, 4 figures, 1 table, and Supplementary Materials | null | null | null | physics.soc-ph cs.DL cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recombinant growth theory highlights the pivotal role of cumulative knowledge
in driving innovation. Although interconnected knowledge facilitates smoother
dissemination, its connection to scientific disruption remains poorly
understood. Here, we quantify knowledge dependence based on the degree to which
references within a given paper's bibliography cite one another. Analyzing 53.8
million papers spanning six decades, we observe that papers built on
independent knowledge have decreased over time. However, propensity score
matching and regression analyses reveal that such papers are associated with
greater scientific disruption, as those who cite them are less likely to cite
their references. Moreover, a team's preference for independent knowledge
amplifies its disruptive potential, regardless of team size, geographic
distance, or collaboration freshness. Despite the disruptive nature, papers
built on independent knowledge receive fewer citations and delayed recognition.
Taken together, these findings fill a critical gap in our fundamental
understanding of scientific innovation, revealing a universal law in peer
recognition: Knowledge independence breeds disruption at the cost of impact.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:15:51 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Yu",
"Xiaoyao",
""
],
[
"Rahwan",
"Talal",
""
],
[
"Jia",
"Tao",
""
]
] |
2504.09590 | Borui Wan | Wan Borui and Zhao Juntao and Jiang Chenyu and Guo Chuanxiong and Wu
Chuan | Efficient LLM Serving on Hybrid Real-time and Best-effort Requests | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent breakthroughs in large Language Models (LLMs) have enabled various
generative tasks on a single model. Real-world services (e.g., OpenAI's ChatGPT
[27]) powered by an LLM often concurrently support latency-critical requests
for interactive applications (e.g., question-answering systems, referred to as
real-time or RT requests) and throughput-oriented requests for back-of-house
processing (e.g., documents batch processing [28], referred to best-effort or
BE requests), with complex hybrid inference workloads to the underlying model.
State-of-the-art (SOTA) LLM serving systems dedicate machines to each type of
request, towards either low inference latency or high serving throughput,
respectively. This practice simplifies request scheduling and management but
suffers from poor resource utilization. We propose BROS, a hybrid LLM serving
system that aims to collocate RT/BE requests, meeting RT requests' latency
requirements while maintaining BE requests' throughput. BROS formulates the
problem of hybrid RT/BE request scheduling and solves it with a dynamic
priority-based algorithm. BROS designs a bidirectional KV cache management
mechanism, allowing RT requests to share KV memory with BE requests to remove
the scheduling restrictions caused by insufficient KV memory and improve
utilization. Extensive experiments validate that BROS achieves a good trade-off
when serving hybrid RT and BE requests. It significantly reduces the latency of
RT requests (up to 74.20%), improving their fine-grained service level
objectives (SLOs) attainments (up to 36.38x), with negligible throughput
reduction for BE requests, showing significant advantages over SOTA systems
like vLLM and TGI.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:16:57 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Borui",
"Wan",
""
],
[
"Juntao",
"Zhao",
""
],
[
"Chenyu",
"Jiang",
""
],
[
"Chuanxiong",
"Guo",
""
],
[
"Chuan",
"Wu",
""
]
] |
2504.09591 | Gurkirat Wadhwa | Gurkirat Wadhwa, Veeraruna Kavitha | What should the encroaching supplier do in markets with some loyal
customers? A Stackelberg Game Approach | null | null | null | null | econ.GN q-fin.EC | http://creativecommons.org/licenses/by/4.0/ | Considering a supply chain with partial vertical integration, we attempt to
seek answers to several questions related to the cooperation competition based
friction, abundant in such networks. Such an SC can represent a supplier with
an inhouse production unit that attempts to control an outhouse production unit
via the said friction. The two production units can have different sets of
loyal customer bases and the aim of the manufacturer supplier duo would be to
get the best out of the two customer bases. Our analysis shows that under
certain market conditions, an optimal strategy might be to allow both units to
earn positive profits particularly when they hold similar market power and when
customer loyalty is high. In cases of weaker customer loyalty, however, the
optimal approach may involve pressurizing the outhouse unit to operate at
minimal profits. Even more intriguing is the scenario where the outhouse unit
has a greater market power and customer loyalty remains strong here, it may be
optimal for the inhouse unit to operate at a loss just enough to dismantle the
downstream monopoly.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:18:05 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wadhwa",
"Gurkirat",
""
],
[
"Kavitha",
"Veeraruna",
""
]
] |
2504.09592 | Eugeniy Beliayev Yurievich | E. Yu. Beliayev, Yogendra Kumar Mishra, Horst-Gunter Rubahn, I. A.
Chichibaba, I. G. Mirzoiev, V. A. Horielyi, A. V. Terekhov | Electron Transport in Compacted Powders of VO2 Nanoparticles: Variable
Range Hopping vs Percolation Behavior | 14 pages, 7 figures. Submission intended for Fizika Nizkikh
Temperatur (Low Temperature Physics) | null | null | null | cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Electron transport properties in compacted VO2 nanopowders were studied.
While VO2 usually exhibits a first-order metal-insulator transition (MIT) at
~340K, in our compressed nanopowder samples the MIT was significantly broadened
due to structural disorder, interparticle barriers, and phase coexistence.
Resistivity measurements in the temperature range of 78 - 682 K initially
suggested a variable range hopping (VRH) transport mechanism, but further
analysis indicates that the observed temperature dependence is governed by
percolative conductivity, modified by activation-assisted tunneling effects.
Suppression of the expected resistance jump at the MIT is attributed to dynamic
intergranular barrier restructuring, residual localized states, and percolative
electron transport. These findings highlight the necessity of considering
percolation effects when analyzing transport mechanisms in granular VO2-based
systems.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:18:11 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Beliayev",
"E. Yu.",
""
],
[
"Mishra",
"Yogendra Kumar",
""
],
[
"Rubahn",
"Horst-Gunter",
""
],
[
"Chichibaba",
"I. A.",
""
],
[
"Mirzoiev",
"I. G.",
""
],
[
"Horielyi",
"V. A.",
""
],
[
"Terekhov",
"A. V.",
""
]
] |
2504.09594 | Vadym Adamyan | Adamyan Vadym | Scattering matrices for perturbations of Laplace operator by infinite
sums of zero-range potentials | null | null | null | null | math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper analyzes the scattering matrix for two unbounded self-adjoint
operators: the standard Laplace operator in three-dimensional space and a
second operator that differs from the first by an infinite sum of zero-range
potentials.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:19:33 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Vadym",
"Adamyan",
""
]
] |
2504.09595 | Daowen Qiu | Hao Li, Daowen Qiu | Distributed quantum algorithm for discrete logarithm problem | 21 pages, 4 figures, comments are welcome | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The quantum algorithm with polynomial time for discrete logarithm problem
proposed by Shor is one of the most significant quantum algorithms, but a large
number of qubits may be required in the Noisy Intermediate-scale Quantum (NISQ)
era. The main contributions of this paper are as follows: (1) A distributed
quantum algorithm for discrete logarithm problem is designed, and its space
complexity and success probability exhibit certain advantages to some extent;
(2) The classical error correction technique proposed by Xiao and Qiu et al. is
generalized by extending three bits to more than three bits.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:30:10 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Li",
"Hao",
""
],
[
"Qiu",
"Daowen",
""
]
] |
2504.09596 | Zan Huang | Zan Huang | Revisiting Self-Attentive Sequential Recommendation | 4 pages, 0 figure, technical report based on
https://github.com/pmixer/SASRec.pytorch experiment findings | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Recommender systems are ubiquitous in on-line services to drive businesses.
And many sequential recommender models were deployed in these systems to
enhance personalization. The approach of using the transformer decoder as the
sequential recommender was proposed years ago and is still a strong baseline in
recent works. But this kind of sequential recommender model did not scale up
well, compared to language models. Quite some details in the classical
self-attentive sequential recommender model could be revisited, and some new
experiments may lead to new findings, without changing the general model
structure which was the focus of many previous works. In this paper, we show
the details and propose new experiment methodologies for future research on
sequential recommendation, in hope to motivate further exploration to new
findings in this area.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:30:57 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Huang",
"Zan",
""
]
] |
2504.09598 | Yining Zhao | Yining Zhao, Ali Braytee, Mukesh Prasad | DualPrompt-MedCap: A Dual-Prompt Enhanced Approach for Medical Image
Captioning | 11 pages, 4 figures, 2 tables | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Medical image captioning via vision-language models has shown promising
potential for clinical diagnosis assistance. However, generating contextually
relevant descriptions with accurate modality recognition remains challenging.
We present DualPrompt-MedCap, a novel dual-prompt enhancement framework that
augments Large Vision-Language Models (LVLMs) through two specialized
components: (1) a modality-aware prompt derived from a semi-supervised
classification model pretrained on medical question-answer pairs, and (2) a
question-guided prompt leveraging biomedical language model embeddings. To
address the lack of captioning ground truth, we also propose an evaluation
framework that jointly considers spatial-semantic relevance and medical
narrative quality. Experiments on multiple medical datasets demonstrate that
DualPrompt-MedCap outperforms the baseline BLIP-3 by achieving a 22%
improvement in modality recognition accuracy while generating more
comprehensive and question-aligned descriptions. Our method enables the
generation of clinically accurate reports that can serve as medical experts'
prior knowledge and automatic annotations for downstream vision-language tasks.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:31:55 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhao",
"Yining",
""
],
[
"Braytee",
"Ali",
""
],
[
"Prasad",
"Mukesh",
""
]
] |
2504.09599 | Elia Bru\`e | Elia Bru\`e, Aaron Naber, Daniele Semola | Fukaya-Yamaguchi Conjecture in Dimension Four | null | null | null | null | math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fukaya and Yamaguchi conjectured that if $M^n$ is a manifold with nonnegative
sectional curvature, then the fundamental group is uniformly virtually abelian.
In this short note we observe that the conjecture holds in dimensions up to
four.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:32:01 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Bruè",
"Elia",
""
],
[
"Naber",
"Aaron",
""
],
[
"Semola",
"Daniele",
""
]
] |
2504.09600 | Phoebus Rosakis | Phoebus Rosakis | Uniqueness for Some Mixed Problems of Nonlinear Elastostatics | 1 figure, 16 pages | null | null | null | math.AP math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | We show that certain mixed displacement/traction problems (including live
pressure tractions) of nonlinear elastostatics that are solved by a homogeneous
deformation, admit no other classical equilibrium solution under suitable
constitutive inequalities and domain boundary restrictions. This extends a well
known theorem of Knops and Stuart on the pure displacement problem.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:32:23 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Rosakis",
"Phoebus",
""
]
] |
2504.09601 | Xiaofeng Liu | Jia Wei, Xiaoqi Zhao, Jonghye Woo, Jinsong Ouyang, Georges El Fakhri,
Qingyu Chen, Xiaofeng Liu | Mixture-of-Shape-Experts (MoSE): End-to-End Shape Dictionary Framework
to Prompt SAM for Generalizable Medical Segmentation | Accepted to CVPR 2025 workshop | null | null | null | cs.CV cs.LG cs.MM eess.IV physics.med-ph | http://creativecommons.org/licenses/by/4.0/ | Single domain generalization (SDG) has recently attracted growing attention
in medical image segmentation. One promising strategy for SDG is to leverage
consistent semantic shape priors across different imaging protocols, scanner
vendors, and clinical sites. However, existing dictionary learning methods that
encode shape priors often suffer from limited representational power with a
small set of offline computed shape elements, or overfitting when the
dictionary size grows. Moreover, they are not readily compatible with large
foundation models such as the Segment Anything Model (SAM). In this paper, we
propose a novel Mixture-of-Shape-Experts (MoSE) framework that seamlessly
integrates the idea of mixture-of-experts (MoE) training into dictionary
learning to efficiently capture diverse and robust shape priors. Our method
conceptualizes each dictionary atom as a shape expert, which specializes in
encoding distinct semantic shape information. A gating network dynamically
fuses these shape experts into a robust shape map, with sparse activation
guided by SAM encoding to prevent overfitting. We further provide this shape
map as a prompt to SAM, utilizing the powerful generalization capability of SAM
through bidirectional integration. All modules, including the shape dictionary,
are trained in an end-to-end manner. Extensive experiments on multiple public
datasets demonstrate its effectiveness.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:34:52 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wei",
"Jia",
""
],
[
"Zhao",
"Xiaoqi",
""
],
[
"Woo",
"Jonghye",
""
],
[
"Ouyang",
"Jinsong",
""
],
[
"Fakhri",
"Georges El",
""
],
[
"Chen",
"Qingyu",
""
],
[
"Liu",
"Xiaofeng",
""
]
] |
2504.09603 | Elia Bru\`e | Elia Bru\`e, Aaron Naber, Daniele Semola | Compact Manifolds with Unbounded Nilpotent Fundamental Groups and
Positive Ricci Curvature | null | null | null | null | math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It follows from the work of Kapovitch and Wilking that a closed manifold with
nonnegative Ricci curvature has an almost nilpotent fundamental group. Leftover
questions and conjectures have asked if in this context the fundamental group
is actually uniformly almost abelian. The main goal of this work is to
construct examples $(M^{10}_k, g_k)$ with uniformly positive Ricci curvature
${\rm Ric}_{g_k}\geq 9$ whose fundamental groups cannot be uniformly virtually
abelian.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:39:53 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Bruè",
"Elia",
""
],
[
"Naber",
"Aaron",
""
],
[
"Semola",
"Daniele",
""
]
] |
2504.09604 | Christopher Ackerman | Christopher M. Ackerman and Nina Panickssery | Mitigating Many-Shot Jailbreaking | null | null | null | null | cs.LG cs.CR | http://creativecommons.org/licenses/by/4.0/ | Many-shot jailbreaking (MSJ) is an adversarial technique that exploits the
long context windows of modern LLMs to circumvent model safety training by
including in the prompt many examples of a ``fake'' assistant responding
inappropriately before the final request. With enough examples, the model's
in-context learning abilities override its safety training, and it responds as
if it were the ``fake'' assistant. In this work, we probe the effectiveness of
different fine tuning and input sanitization approaches on mitigating MSJ
attacks, alone and in combination. We find incremental mitigation effectiveness
for each, and we show that the combined techniques significantly reduce the
effectiveness of MSJ attacks, while retaining model performance in benign
in-context learning and conversational tasks. We suggest that our approach
could meaningfully ameliorate this vulnerability if incorporated into model
safety post-training.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:42:03 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Ackerman",
"Christopher M.",
""
],
[
"Panickssery",
"Nina",
""
]
] |
2504.09605 | Daowen Qiu | Hao Li, Daowen Qiu | Improved distributed quantum algorithm for Simon's problem | 10 pages, 3 figures, comments are welcome | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Simon's problem is one of the most important problems demonstrating the power
of quantum computing. Recently, an interesting distributed quantum algorithm
for Simon's problem was proposed, where a key sorting operator requiring a
large number of qubits was employed. In this paper, we design an improved
distributed quantum algorithm for Simon's problem without using sorting
operators, and our algorithm has the advantage of reducing half number of
qubits required for a single computing node. Moreover, our algorithm does not
involve the classical search process.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:50:01 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Li",
"Hao",
""
],
[
"Qiu",
"Daowen",
""
]
] |
2504.09606 | Lexington Whalen | Lexington Whalen, Zhenbang Du, Haoran You, Chaojian Li, Sixu Li,
Yingyan (Celine) Lin | Early-Bird Diffusion: Investigating and Leveraging Timestep-Aware
Early-Bird Tickets in Diffusion Models for Efficient Training | 10 pages, 5 figures. Accepted to the IEEE/CVF Conference on Computer
Vision and Pattern Recognition 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Training diffusion models (DMs) requires substantial computational resources
due to multiple forward and backward passes across numerous timesteps,
motivating research into efficient training techniques. In this paper, we
propose EB-Diff-Train, a new efficient DM training approach that is orthogonal
to other methods of accelerating DM training, by investigating and leveraging
Early-Bird (EB) tickets -- sparse subnetworks that manifest early in the
training process and maintain high generation quality.
We first investigate the existence of traditional EB tickets in DMs, enabling
competitive generation quality without fully training a dense model.
Then, we delve into the concept of diffusion-dedicated EB tickets, drawing on
insights from varying importance of different timestep regions. These tickets
adapt their sparsity levels according to the importance of corresponding
timestep regions, allowing for aggressive sparsity during non-critical regions
while conserving computational resources for crucial timestep regions.
Building on this, we develop an efficient DM training technique that derives
timestep-aware EB tickets, trains them in parallel, and combines them during
inference for image generation. Extensive experiments validate the existence of
both traditional and timestep-aware EB tickets, as well as the effectiveness of
our proposed EB-Diff-Train method. This approach can significantly reduce
training time both spatially and temporally -- achieving 2.9$\times$ to
5.8$\times$ speedups over training unpruned dense models, and up to
10.3$\times$ faster training compared to standard train-prune-finetune
pipelines -- without compromising generative quality.
Our code is available at https://github.com/GATECH-EIC/Early-Bird-Diffusion.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:53:18 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Whalen",
"Lexington",
"",
"Celine"
],
[
"Du",
"Zhenbang",
"",
"Celine"
],
[
"You",
"Haoran",
"",
"Celine"
],
[
"Li",
"Chaojian",
"",
"Celine"
],
[
"Li",
"Sixu",
"",
"Celine"
],
[
"Yingyan",
"",
"",
"Celine"
],
[
"Lin",
"",
""
]
] |
2504.09607 | Shaoheng Zhang | Shaoheng Zhang, Kui Wang, Yun Wang | Point Singularities of Solutions to the Stationary Incompressible MHD
Equations | 17 pages | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the point singularity of very weak solutions
$(\mathbf{u},\mathbf{B})$ to the stationary MHD equations. More precisely,
assume that the solution $(\mathbf{u},\mathbf{B})$ in the punctured ball
$B_2\setminus \{0\}$ satisfies the vanishing condition (4), and that
$|\mathbf{u}(x)|\le \varepsilon |x|^{-1},\ |\mathbf{B}(x)|\le C |x|^{-1}$ with
small $\varepsilon>0$ and general $C>0$. Then, the leading order term of
$\mathbf{u}$ is a Landau solution, while the $(-1)$ order term of $\mathbf{B}$
is $0$. In particular, for axisymmetric solutions $(\mathbf{u}, \mathbf{B})$,
the condition (4) holds provided $\mathbf{B} = B^{\theta}(r,z)
\mathbf{e}_{\theta}$ or the boundary condition (7) is imposed.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:56:22 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhang",
"Shaoheng",
""
],
[
"Wang",
"Kui",
""
],
[
"Wang",
"Yun",
""
]
] |
2504.09608 | Xingke Song | Xingke Song, Xiaoying Yang, Chenglin Yao, Jianfeng Ren, Ruibin Bai,
Xin Chen, Xudong Jiang | ERL-MPP: Evolutionary Reinforcement Learning with Multi-head Puzzle
Perception for Solving Large-scale Jigsaw Puzzles of Eroded Gaps | 9 pages, 5 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Solving jigsaw puzzles has been extensively studied. While most existing
models focus on solving either small-scale puzzles or puzzles with no gap
between fragments, solving large-scale puzzles with gaps presents distinctive
challenges in both image understanding and combinatorial optimization. To
tackle these challenges, we propose a framework of Evolutionary Reinforcement
Learning with Multi-head Puzzle Perception (ERL-MPP) to derive a better set of
swapping actions for solving the puzzles. Specifically, to tackle the
challenges of perceiving the puzzle with gaps, a Multi-head Puzzle Perception
Network (MPPN) with a shared encoder is designed, where multiple puzzlet heads
comprehensively perceive the local assembly status, and a discriminator head
provides a global assessment of the puzzle. To explore the large swapping
action space efficiently, an Evolutionary Reinforcement Learning (EvoRL) agent
is designed, where an actor recommends a set of suitable swapping actions from
a large action space based on the perceived puzzle status, a critic updates the
actor using the estimated rewards and the puzzle status, and an evaluator
coupled with evolutionary strategies evolves the actions aligning with the
historical assembly experience. The proposed ERL-MPP is comprehensively
evaluated on the JPLEG-5 dataset with large gaps and the MIT dataset with
large-scale puzzles. It significantly outperforms all state-of-the-art models
on both datasets.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:56:41 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Song",
"Xingke",
""
],
[
"Yang",
"Xiaoying",
""
],
[
"Yao",
"Chenglin",
""
],
[
"Ren",
"Jianfeng",
""
],
[
"Bai",
"Ruibin",
""
],
[
"Chen",
"Xin",
""
],
[
"Jiang",
"Xudong",
""
]
] |
2504.09609 | Jun-Gill Kang | Dohyeon Lee, Jun-Gill Kang, and Soohee Han | A highly maneuverable flying squirrel drone with agility-improving
foldable wings | Accepted to IEEE Robotics and Automation Letters Youtube :
https://youtu.be/tckIF3KCJig?si=s-N2VRNThV23y9xQ | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | Drones, like most airborne aerial vehicles, face inherent disadvantages in
achieving agile flight due to their limited thrust capabilities. These physical
constraints cannot be fully addressed through advancements in control
algorithms alone. Drawing inspiration from the winged flying squirrel, this
paper proposes a highly maneuverable drone equipped with agility-enhancing
foldable wings. By leveraging collaborative control between the conventional
propeller system and the foldable wings-coordinated through the Thrust-Wing
Coordination Control (TWCC) framework-the controllable acceleration set is
expanded, enabling the generation of abrupt vertical forces that are
unachievable with traditional wingless drones. The complex aerodynamics of the
foldable wings are modeled using a physics-assisted recurrent neural network
(paRNN), which calibrates the angle of attack (AOA) to align with the real
aerodynamic behavior of the wings. The additional air resistance generated by
appropriately deploying these wings significantly improves the tracking
performance of the proposed "flying squirrel" drone. The model is trained on
real flight data and incorporates flat-plate aerodynamic principles.
Experimental results demonstrate that the proposed flying squirrel drone
achieves a 13.1% improvement in tracking performance, as measured by root mean
square error (RMSE), compared to a conventional wingless drone. A demonstration
video is available on YouTube: https://youtu.be/O8nrip18azY.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 14:57:11 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Lee",
"Dohyeon",
""
],
[
"Kang",
"Jun-Gill",
""
],
[
"Han",
"Soohee",
""
]
] |
2504.09611 | Dev Prakash Jha | Dev Prakash Jha and Raju K. George | An Operator-Theoretic Framework for the Optimal Control Problem of
Nonlinear Caputo Fractional Systems | 29 pages,2 figures | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | This paper addresses the optimal control problem for a class of nonlinear
fractional systems involving Caputo derivatives and nonlocal initial
conditions. The system is reformulated as an abstract Hammerstein-type operator
equation, enabling the application of operator-theoretic techniques. Sufficient
conditions are established to guarantee the existence of mild solutions and
optimal control-state pairs. The analysis covers both convex and non-convex
scenarios through various sets of assumptions on the involved operators. An
optimality system is derived for quadratic cost functionals using the G\^ateaux
derivative, and the connection with Pontryagin-type minimum principles is
discussed. Illustrative examples demonstrate the effectiveness of the proposed
theoretical framework.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:04:02 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Jha",
"Dev Prakash",
""
],
[
"George",
"Raju K.",
""
]
] |
2504.09613 | Michael Borinsky | Michael Borinsky and Mathijs Fraaije | Tropical sampling from Feynman measures | 29 pages, comments welcome! | null | null | null | hep-th hep-ph math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | We introduce an algorithm that samples a set of loop momenta distributed as a
given Feynman integrand. The algorithm uses the tropical sampling method and
can be applied to evaluate phase-space-type integrals efficiently. We provide
an implementation, momtrop, and apply it to a series of relevant integrals from
the loop-tree duality framework. Compared to naive sampling methods, we observe
convergence speedups by factors of more than $10^6$.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:09:36 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Borinsky",
"Michael",
""
],
[
"Fraaije",
"Mathijs",
""
]
] |
2504.09614 | Jerome Lecoq | Ido Aizenbud, Nicholas Audette, Ryszard Auksztulewicz, Krzysztof
Basi\'nski, Andr\'e M. Bastos, Michael Berry, Andres Canales-Johnson, Hannah
Choi, Claudia Clopath, Uri Cohen, Rui Ponte Costa, Roberto De Filippo, Roman
Doronin, Steven P. Errington, Jeffrey P. Gavornik, Colleen J. Gillon, Arno
Granier, Jordan P. Hamm, Loreen Hert\"ag, Henry Kennedy, Sandeep Kumar,
Alexander Ladd, Hugo Ladret, J\'er\^ome A. Lecoq, Alexander Maier, Patrick
McCarthy, Jie Mei, Jorge Mejias, Fabian Mikulasch, Noga Mudrik, Farzaneh
Najafi, Kevin Nejad, Hamed Nejat, Karim Oweiss, Mihai A. Petrovici, Viola
Priesemann, Lucas Rudelt, Sarah Ruediger, Simone Russo, Alessandro
Salatiello, Walter Senn, Eli Sennesh, Sepehr Sima, Cem Uran, Anna
Vasilevskaya, Julien Vezoli, Martin Vinck, Jacob A. Westerberg, Katharina
Wilmes, Yihan Sophy Xiong | Neural mechanisms of predictive processing: a collaborative community
experiment through the OpenScope program | null | null | null | null | q-bio.NC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This review synthesizes advances in predictive processing within the sensory
cortex. Predictive processing theorizes that the brain continuously predicts
sensory inputs, refining neuronal responses by highlighting prediction errors.
We identify key computational primitives, such as stimulus adaptation,
dendritic computation, excitatory/inhibitory balance and hierarchical
processing, as central to this framework. Our review highlights convergences,
such as top-down inputs and inhibitory interneurons shaping mismatch signals,
and divergences, including species-specific hierarchies and modality-dependent
layer roles. To address these conflicts, we propose experiments in mice and
primates using in-vivo two-photon imaging and electrophysiological recordings
to test whether temporal, motor, and omission mismatch stimuli engage shared or
distinct mechanisms. The resulting dataset, collected and shared via the
OpenScope program, will enable model validation and community analysis,
fostering iterative refinement and refutability to decode the neural circuits
of predictive processing.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:10:09 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Aizenbud",
"Ido",
""
],
[
"Audette",
"Nicholas",
""
],
[
"Auksztulewicz",
"Ryszard",
""
],
[
"Basiński",
"Krzysztof",
""
],
[
"Bastos",
"André M.",
""
],
[
"Berry",
"Michael",
""
],
[
"Canales-Johnson",
"Andres",
""
],
[
"Choi",
"Hannah",
""
],
[
"Clopath",
"Claudia",
""
],
[
"Cohen",
"Uri",
""
],
[
"Costa",
"Rui Ponte",
""
],
[
"De Filippo",
"Roberto",
""
],
[
"Doronin",
"Roman",
""
],
[
"Errington",
"Steven P.",
""
],
[
"Gavornik",
"Jeffrey P.",
""
],
[
"Gillon",
"Colleen J.",
""
],
[
"Granier",
"Arno",
""
],
[
"Hamm",
"Jordan P.",
""
],
[
"Hertäg",
"Loreen",
""
],
[
"Kennedy",
"Henry",
""
],
[
"Kumar",
"Sandeep",
""
],
[
"Ladd",
"Alexander",
""
],
[
"Ladret",
"Hugo",
""
],
[
"Lecoq",
"Jérôme A.",
""
],
[
"Maier",
"Alexander",
""
],
[
"McCarthy",
"Patrick",
""
],
[
"Mei",
"Jie",
""
],
[
"Mejias",
"Jorge",
""
],
[
"Mikulasch",
"Fabian",
""
],
[
"Mudrik",
"Noga",
""
],
[
"Najafi",
"Farzaneh",
""
],
[
"Nejad",
"Kevin",
""
],
[
"Nejat",
"Hamed",
""
],
[
"Oweiss",
"Karim",
""
],
[
"Petrovici",
"Mihai A.",
""
],
[
"Priesemann",
"Viola",
""
],
[
"Rudelt",
"Lucas",
""
],
[
"Ruediger",
"Sarah",
""
],
[
"Russo",
"Simone",
""
],
[
"Salatiello",
"Alessandro",
""
],
[
"Senn",
"Walter",
""
],
[
"Sennesh",
"Eli",
""
],
[
"Sima",
"Sepehr",
""
],
[
"Uran",
"Cem",
""
],
[
"Vasilevskaya",
"Anna",
""
],
[
"Vezoli",
"Julien",
""
],
[
"Vinck",
"Martin",
""
],
[
"Westerberg",
"Jacob A.",
""
],
[
"Wilmes",
"Katharina",
""
],
[
"Xiong",
"Yihan Sophy",
""
]
] |
2504.09615 | Hong Duc Bui | Hong Duc Bui | Counting Number of Triangulations of Point Sets: Reinterpreting and
Generalizing the Triangulation Polynomials | 14 pages, 8 figures | null | null | null | cs.CG | http://creativecommons.org/licenses/by/4.0/ | We describe a framework that unifies the two types of polynomials introduced
respectively by Bacher and Mouton and by Rutschmann and Wettstein to analyze
the number of triangulations of point sets. Using this insight, we generalize
the triangulation polynomials of chains to a wider class of near-edges,
enabling efficient computation of the number of triangulations of certain
families of point sets. We use the framework to try to improve the result in
Rutschmann and Wettstein without success, suggesting that their result is close
to optimal.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:16:04 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Bui",
"Hong Duc",
""
]
] |
2504.09616 | Alejandro Anderson | Alejandro Anderson and Rami Katz and Francesca Cal\`a Campana and
Giulia Giordano | Failure and Success in Single-Drug Control of Antimicrobial Resistance | 6 pages, 4 Figures, submitted to CDC+L-CSS | null | null | null | q-bio.PE | http://creativecommons.org/licenses/by/4.0/ | We propose a mathematical model of Antimicrobial Resistance in the host to
predict the failure of two antagonists of bacterial growth: the immune response
and a single-antibiotic therapy. After characterising the initial bacterial
load that cannot be cleared by the immune system alone, we define the success
set of initial conditions for which an infection-free equilibrium can be
reached by a viable single-antibiotic therapy, and we provide a rigorously
defined inner approximation of the set. Finally, we propose an optimal control
approach to design and compare successful single-drug therapies.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:20:36 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Anderson",
"Alejandro",
""
],
[
"Katz",
"Rami",
""
],
[
"Campana",
"Francesca Calà",
""
],
[
"Giordano",
"Giulia",
""
]
] |
2504.09617 | Raj Kumar Mistri Dr. | Raj Kumar Mistri, Nitesh Prajapati | Direct and Inverse Problems for Restricted Signed Sumsets -- II | 47 pages | null | null | null | math.NT math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $A=\{a_{1},\ldots,a_{k}\}$ be a nonempty finite subset of an additive
abelian group $G$. For a positive integer $h$, the restricted $h$-fold signed
sumset of $A$, denoted by $h^{\wedge}_{\pm}A$, is defined as
$$h^{\wedge}_{\pm}A = \left\lbrace \sum_{i=1}^{k} \lambda_{i} a_{i}:
\lambda_{i} \in \left\lbrace -1, 0, 1\right\rbrace \ \text{for} \ i= 1, 2,
\ldots, k \ \text{and} \ \sum_{i=1}^{k} \left|\lambda_{i} \right|
=h\right\rbrace. $$ A direct problem for the restricted $h$-fold signed sumset
is to find the optimal size of $h^{\wedge}_{\pm}A$ in terms of $h$ and $|A|$.
An inverse problem for this sumset is to determine the structure of the
underlying set $A$ when the sumset has optimal size. While the signed sumsets
(which is defined differently compared to the restricted signed sumset) in
finite abelian groups has been investigated by Bajnok and Matzke, the
restricted $h$-fold signed sumset $h^{\wedge}_{\pm}A$ is not well studied even
in the additive group of integers $\Bbb Z$. Bhanja, Komatsu and Pandey studied
these problems for the restricted $h$-fold signed sumset for $h=2, 3$, and $k$,
and conjectured some direct and inverse results for $h \geq 4$. In a recent
paper, Mistri and Prajapati proved these conjectures completely for the set of
positive integers. In this paper, we prove these conjectures for the set of
nonnegative integers, which settles all the conjectures completely.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:21:05 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Mistri",
"Raj Kumar",
""
],
[
"Prajapati",
"Nitesh",
""
]
] |
2504.09618 | Zhaoyang Ming | Zhaoyang Ming, Shanpu Shen, Junhui Rao, Zan Li, Jichen Zhang, Chi Yuk
Chiu, Ross Murch | A Hybrid Transmitting and Reflecting Beyond Diagonal Reconfigurable
Intelligent Surface with Independent Beam Control and Power Splitting | 15 pages, 16 figures | null | null | null | eess.SP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | A hybrid transmitting and reflecting beyond diagonal reconfigurable
intelligent surface (BD-RIS) design is proposed. Operating in the same
aperture, frequency band and polarization, the proposed BD-RIS features
independent beam steering control of its reflected and transmitted waves. In
addition it provides a hybrid mode with both reflected and transmitted waves
using tunable power splitting between beams. The BD-RIS comprises two phase
reconfigurable antenna arrays interconnected by an array of tunable two-port
power splitters. The two-port power splitter in each BD-RIS cell is built upon
a varactor in parallel with a bias inductor to exert tunable impedance
variations on transmission lines. Provided with variable reverse DC voltages,
the two-port power splitter can control the power ratio of S11 over S21 from
-20 dB to 20 dB, thus allowing tunable power splitting. Each antenna is 2-bit
phase reconfigurable with 200 MHz bandwidth at 2.4 GHz so that each cell of
BD-RIS can also achieve independent reflection and transmission phase control.
To characterize and optimize the electromagnetic response of the proposed
BD-RIS design, a Th\'evenin equivalent model and corresponding analytical
method is provided. A BD-RIS with 4 by 4 cells was also prototyped and tested.
Experiments show that in reflection and transmission mode, the fabricated
BD-RIS can realize beam steering in reflection and transmission space,
respectively. It is also verified that when operating in hybrid mode, the
BD-RIS enables independent beam steering of the reflected and transmitted
waves. This work helps fill the gap between realizing practical hardware design
and establishing an accurate physical model for the hybrid transmitting and
reflecting BD-RIS, enabling hybrid transmitting and reflecting BD-RIS assisted
wireless communications.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:22:49 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Ming",
"Zhaoyang",
""
],
[
"Shen",
"Shanpu",
""
],
[
"Rao",
"Junhui",
""
],
[
"Li",
"Zan",
""
],
[
"Zhang",
"Jichen",
""
],
[
"Chiu",
"Chi Yuk",
""
],
[
"Murch",
"Ross",
""
]
] |
2504.09619 | Volodya Pastukhov | O. Hryhorchak, G. Panochko, and V. Pastukhov | Evolution of a single spin in ideal Bose gas at finite temperatures | 6 pages, 4 figures; comments and relevant references are welcome | null | null | null | cond-mat.quant-gas quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the finite-temperature dynamics of non-interacting bosons with a
single static spinful impurity immersed. A non-zero contact boson-impurity
pairwise interaction is assumed only for the spin-up impurity state. By tracing
out bosonic degrees of freedom, the exact time evolution of the impurity spin
is calculated for pure and mixed initial ensembles of states. The
time-dependent momentum distribution of bosons initially created in the Bose
condensed state and driven by the interaction with spin is analyzed.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:23:43 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Hryhorchak",
"O.",
""
],
[
"Panochko",
"G.",
""
],
[
"Pastukhov",
"V.",
""
]
] |
2504.09620 | Tadahiro Taniguchi | Yuta Matsui, Ryosuke Yamaki, Ryo Ueda, Seitaro Shinagawa, Tadahiro
Taniguchi | Metropolis-Hastings Captioning Game: Knowledge Fusion of Vision Language
Models via Decentralized Bayesian Inference | null | null | null | null | cs.CL cs.AI cs.CV cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose the Metropolis-Hastings Captioning Game (MHCG), a method to fuse
knowledge of multiple vision-language models (VLMs) by learning from each
other. Although existing methods that combine multiple models suffer from
inference costs and architectural constraints, MHCG avoids these problems by
performing decentralized Bayesian inference through a process resembling a
language game. The knowledge fusion process establishes communication between
two VLM agents alternately captioning images and learning from each other. We
conduct two image-captioning experiments with two VLMs, each pre-trained on a
different dataset. The first experiment demonstrates that MHCG achieves
consistent improvement in reference-free evaluation metrics. The second
experiment investigates how MHCG contributes to sharing VLMs' category-level
vocabulary by observing the occurrence of the vocabulary in the generated
captions.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:28:09 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Matsui",
"Yuta",
""
],
[
"Yamaki",
"Ryosuke",
""
],
[
"Ueda",
"Ryo",
""
],
[
"Shinagawa",
"Seitaro",
""
],
[
"Taniguchi",
"Tadahiro",
""
]
] |
2504.09621 | Jiuchen Chen | Jiuchen Chen, Xinyu Yan, Qizhi Xu, Kaiqi Li | Tokenize Image Patches: Global Context Fusion for Effective Haze Removal
in Large Images | Accepted by CVPR 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Global contextual information and local detail features are essential for
haze removal tasks. Deep learning models perform well on small, low-resolution
images, but they encounter difficulties with large, high-resolution ones due to
GPU memory limitations. As a compromise, they often resort to image slicing or
downsampling. The former diminishes global information, while the latter
discards high-frequency details. To address these challenges, we propose
DehazeXL, a haze removal method that effectively balances global context and
local feature extraction, enabling end-to-end modeling of large images on
mainstream GPU hardware. Additionally, to evaluate the efficiency of global
context utilization in haze removal performance, we design a visual attribution
method tailored to the characteristics of haze removal tasks. Finally,
recognizing the lack of benchmark datasets for haze removal in large images, we
have developed an ultra-high-resolution haze removal dataset (8KDehaze) to
support model training and testing. It includes 10000 pairs of clear and hazy
remote sensing images, each sized at 8192 $\times$ 8192 pixels. Extensive
experiments demonstrate that DehazeXL can infer images up to 10240 $\times$
10240 pixels with only 21 GB of memory, achieving state-of-the-art results
among all evaluated methods. The source code and experimental dataset are
available at https://github.com/CastleChen339/DehazeXL.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:41:25 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Chen",
"Jiuchen",
""
],
[
"Yan",
"Xinyu",
""
],
[
"Xu",
"Qizhi",
""
],
[
"Li",
"Kaiqi",
""
]
] |
2504.09622 | Yamir Moreno | Xiangrong Wang, Dan Lu, Zongze Wu, Weina Xu, Hongru Hou, Yanqing Hu,
Yamir Moreno | Predicting the critical behavior of complex dynamic systems via learning
the governing mechanisms | 22 pages including figures. Submitted for publication | null | null | null | physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | Critical points separate distinct dynamical regimes of complex systems, often
delimiting functional or macroscopic phases in which the system operates.
However, the long-term prediction of critical regimes and behaviors is
challenging given the narrow set of parameters from which they emerge. Here, we
propose a framework to learn the rules that govern the dynamic processes of a
system. The learned governing rules further refine and guide the representative
learning of neural networks from a series of dynamic graphs. This combination
enables knowledge-based prediction for the critical behaviors of dynamical
networked systems. We evaluate the performance of our framework in predicting
two typical critical behaviors in spreading dynamics on various synthetic and
real-world networks. Our results show that governing rules can be learned
effectively and significantly improve prediction accuracy. Our framework
demonstrates a scenario for facilitating the representability of deep neural
networks through learning the underlying mechanism, which aims to steer
applications for predicting complex behavior that learnable physical rules can
drive.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:41:29 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wang",
"Xiangrong",
""
],
[
"Lu",
"Dan",
""
],
[
"Wu",
"Zongze",
""
],
[
"Xu",
"Weina",
""
],
[
"Hou",
"Hongru",
""
],
[
"Hu",
"Yanqing",
""
],
[
"Moreno",
"Yamir",
""
]
] |
2504.09623 | Dulanga Weerakoon | Atharv Mahesh Mane, Dulanga Weerakoon, Vigneshwaran Subbaraju, Sougata
Sen, Sanjay E. Sarma, Archan Misra | Ges3ViG: Incorporating Pointing Gestures into Language-Based 3D Visual
Grounding for Embodied Reference Understanding | Accepted to the IEEE/CVF Conference on Computer Vision and Pattern
Recognition (CVPR) 2025 | null | null | null | cs.CV cs.AI cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | 3-Dimensional Embodied Reference Understanding (3D-ERU) combines a language
description and an accompanying pointing gesture to identify the most relevant
target object in a 3D scene. Although prior work has explored pure
language-based 3D grounding, there has been limited exploration of 3D-ERU,
which also incorporates human pointing gestures. To address this gap, we
introduce a data augmentation framework-Imputer, and use it to curate a new
benchmark dataset-ImputeRefer for 3D-ERU, by incorporating human pointing
gestures into existing 3D scene datasets that only contain language
instructions. We also propose Ges3ViG, a novel model for 3D-ERU that achieves
~30% improvement in accuracy as compared to other 3D-ERU models and ~9%
compared to other purely language-based 3D grounding models. Our code and
dataset are available at https://github.com/AtharvMane/Ges3ViG.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:43:06 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Mane",
"Atharv Mahesh",
""
],
[
"Weerakoon",
"Dulanga",
""
],
[
"Subbaraju",
"Vigneshwaran",
""
],
[
"Sen",
"Sougata",
""
],
[
"Sarma",
"Sanjay E.",
""
],
[
"Misra",
"Archan",
""
]
] |
2504.09624 | Sai Wang | Zhi-Chao Zhao, Dong-Mei Xia, Sai Wang | Constraints on the lepton asymmetry from DESI DR2 BAO data | 11 pages, 2 figures, 1 table | null | null | null | astro-ph.CO gr-qc hep-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | It is important to explore the potential existence of lepton asymmetry in the
neutrino sector. Conducting a joint analysis of DESI DR2 BAO data and
\emph{Planck} 2018 CMB data, we obtain the upper limits on the neutrino
degeneracy parameter, i.e., $\xi_{3}<0.56$ for the normal mass hierarchy while
$\xi_{3}<0.62$ for the inverted mass hierarchy, at 95\% confidence level.
Considering the influence of the dynamical dark energy, we find that these
upper limits remain to be robust. This work may provide helpful implications
for model buildings of the matter-antimatter asymmetry in the universe.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:46:04 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhao",
"Zhi-Chao",
""
],
[
"Xia",
"Dong-Mei",
""
],
[
"Wang",
"Sai",
""
]
] |
2504.09625 | Hiran Shanake Perera | Hiran Shanake Perera, Rozainee Khairudin, Khazriyati Salehuddin | The Effects of Socioeconomic Status and Depression on The Neural
Correlates of Error Monitoring. An Event-Related Potential Study | 34 pages, 3 figures, 4 tables, version 1 | null | null | null | q-bio.NC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Existing evidence suggests that neural responses to errors were exaggerated
in individuals at risk of depression and anxiety. This phenomenon has led to
the possibility that the error-related negativity (ERN), a well-known neural
correlate of error monitoring could be used as a diagnostic tool for several
psychological disorders. However, conflicting evidence between psychopathology
and the ERN suggests that this phenomenon is modulated by variables are yet to
be identified. Socioeconomic status (SES) could potentially play a role in the
relationship between the ERN and psychopathological disorders, given that SES
is known to be associated with depression and anxiety. In the current study, we
first tested whether SES was related to ERN amplitude. Second, we examined
whether the relationship between the ERN and depression was explained by
differences in SES. We measured error-related negativity (ERN) from a sample of
adult participants from low to high socioeconomic backgrounds while controlling
their depression scores. Results show that SES correlated with variations in
ERN amplitude. Specifically, we found that low-SES individuals had a larger ERN
than wealthier individuals. In addition, the relationship between depression
and the ERN was fully accounted for by variations in SES. Overall, our results
indicate that SES predicts neural responses to errors. Findings also indicate
that the link between depression and ERN may be the result of SES variations.
Future research examining the links between psychopathology and error
monitoring should control SES differences, and caution is needed if they are to
be used as a diagnostic tool in low-income communities.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:48:33 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Perera",
"Hiran Shanake",
""
],
[
"Khairudin",
"Rozainee",
""
],
[
"Salehuddin",
"Khazriyati",
""
]
] |
2504.09626 | Matthew Harrison-Trainor | Rachael Alvir, Barbara Csima, and Matthew Harrison-Trainor | On the computability of optimal Scott sentences | null | null | null | null | math.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Given a countable mathematical structure, its Scott sentence is a sentence of
the infinitary logic $\mathcal{L}_{\omega_1 \omega}$ that characterizes it
among all countable structures. We can measure the complexity of a structure by
the least complexity of a Scott sentence for that structure. It is known that
there can be a difference between the least complexity of a Scott sentence and
the least complexity of a computable Scott sentence; for example, Alvir,
Knight, and McCoy showed that there is a computable structure with a $\Pi_2$
Scott sentence but no computable $\Pi_2$ Scott sentence. It is well known that
a structure with a $\Pi_2$ Scott sentence must have a computable $\Pi_4$ Scott
sentence. We show that this is best possible: there is a computable structure
with a $\Pi_2$ Scott sentence but no computable $\Sigma_4$ Scott sentence. We
also show that there is no reasonable characterization of the computable
structures with a computable $\Pi_n$ Scott sentence by showing that the index
set of such structures is $\Pi^1_1$-$m$-complete.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:50:53 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Alvir",
"Rachael",
""
],
[
"Csima",
"Barbara",
""
],
[
"Harrison-Trainor",
"Matthew",
""
]
] |
2504.09627 | Junjie Zhang | Junjie Zhang, Beichen Zhang, Wenqi Sun, Hongyu Lu, Wayne Xin Zhao, Yu
Chen, Ji-Rong Wen | Slow Thinking for Sequential Recommendation | null | null | null | null | cs.IR cs.AI | http://creativecommons.org/licenses/by/4.0/ | To develop effective sequential recommender systems, numerous methods have
been proposed to model historical user behaviors. Despite the effectiveness,
these methods share the same fast thinking paradigm. That is, for making
recommendations, these methods typically encodes user historical interactions
to obtain user representations and directly match these representations with
candidate item representations. However, due to the limited capacity of
traditional lightweight recommendation models, this one-step inference paradigm
often leads to suboptimal performance. To tackle this issue, we present a novel
slow thinking recommendation model, named STREAM-Rec. Our approach is capable
of analyzing historical user behavior, generating a multi-step, deliberative
reasoning process, and ultimately delivering personalized recommendations. In
particular, we focus on two key challenges: (1) identifying the suitable
reasoning patterns in recommender systems, and (2) exploring how to effectively
stimulate the reasoning capabilities of traditional recommenders. To this end,
we introduce a three-stage training framework. In the first stage, the model is
pretrained on large-scale user behavior data to learn behavior patterns and
capture long-range dependencies. In the second stage, we design an iterative
inference algorithm to annotate suitable reasoning traces by progressively
refining the model predictions. This annotated data is then used to fine-tune
the model. Finally, in the third stage, we apply reinforcement learning to
further enhance the model generalization ability. Extensive experiments
validate the effectiveness of our proposed method.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:53:30 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhang",
"Junjie",
""
],
[
"Zhang",
"Beichen",
""
],
[
"Sun",
"Wenqi",
""
],
[
"Lu",
"Hongyu",
""
],
[
"Zhao",
"Wayne Xin",
""
],
[
"Chen",
"Yu",
""
],
[
"Wen",
"Ji-Rong",
""
]
] |
2504.09628 | Xin Zhang | Xin Zhang, Wensheng Lin, Lixin Li, Zhu Han and Tad Matsumoto | Outage Probability Analysis for OTFS with Finite Blocklength | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Orthogonal time frequency space (OTFS) modulation is widely acknowledged as a
prospective waveform for future wireless communication networks.To provide
insights for the practical system design, this paper analyzes the outage
probability of OTFS modulation with finite blocklength.To begin with, we
present the system model and formulate the analysis of outage probability for
OTFS with finite blocklength as an equivalent problem of calculating the outage
probability with finite blocklength over parallel additive white Gaussian noise
(AWGN) channels.Subsequently, we apply the equivalent noise approach to derive
a lower bound on the outage probability of OTFS with finite blocklength under
both average power allocation and water-filling power allocation strategies,
respectively.Finally, the lower bounds of the outage probability are determined
using the Monte-Carlo method for the two power allocation strategies.The impact
of the number of resolvable paths and coding rates on the outage probability is
analyzed, and the simulation results are compared with the theoretical lower
bounds.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:53:46 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhang",
"Xin",
""
],
[
"Lin",
"Wensheng",
""
],
[
"Li",
"Lixin",
""
],
[
"Han",
"Zhu",
""
],
[
"Matsumoto",
"Tad",
""
]
] |
2504.09629 | Yuma Ichikawa | Yamato Arai, Yuma Ichikawa | Quantization Error Propagation: Revisiting Layer-Wise Post-Training
Quantization | 16 pages, 1 figure | null | null | null | cs.LG stat.AP stat.ME stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Layer-wise post-training quantization has emerged as a widely used technique
for compressing large language models (LLMs) without retraining. However,
recent progress in this line of research is saturating, underscoring the need
to revisit its core limitation and explore further improvements. This study
identifies a critical bottleneck in existing layer-wise PTQ methods: the
accumulation of quantization errors across layers significantly degrades
performance, particularly in low-bit regimes. To address this, we propose
Quantization Error Propagation (QEP), a lightweight and general framework that
enhances layer-wise PTQ by explicitly propagating the quantization error which
enable compensating for accumulated quantization errors. Additionally, we
introduce a tunable propagation mechanism that allows for control over both
propagation strength and computational overhead, making the framework adaptable
to various architectures and resource constraints. Empirical evaluation on
LLaMA2 models (7B, 13B, 70B) demonstrate that incorporating QEP into standard
layer-wise PTQ pipelines outperforms standard PTQ methods. Notably, QEP yields
substantial performance improvements under extreme low-bit quantization
settings.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:56:00 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Arai",
"Yamato",
""
],
[
"Ichikawa",
"Yuma",
""
]
] |
2504.09630 | Yuanlin Gong | Yuanlin Gong, Hongliang Tian, Lei Wu and Bin Zhu | Probing Millicharged Dark Matter with Magnetometer Coupled to Circuit | 7 pages, 2 figures | null | null | CPTNP-2025-002 | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a novel approach to detect millicharged dark matter (mDM) by using
a high-sensitivity magnetometer coupled with the resonant and broadband readout
circuits. In the external magnetic field, the interaction between mDM and the
photon field introduces an effective current corresponding to the mDMs
annihilation into photons that produces a faint oscillating magnetic field
signal, with a frequency uniquely determined by twice the mDM mass. By
calculating the expected signal for two experimental configurations -- toroidal
and solenoidal magnetic fields -- we show the potential to explore the
uncharted regions of mDM parameter space. Our analysis establishes
unprecedented constraints on the mDM coupling constant across the mass range
$1\times 10^{-12}~\mathrm{eV}$ to $6 \times 10^{-8}~\mathrm{eV}$, surpassing
existing experimental limits by more than ten orders of magnitude.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 15:57:38 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Gong",
"Yuanlin",
""
],
[
"Tian",
"Hongliang",
""
],
[
"Wu",
"Lei",
""
],
[
"Zhu",
"Bin",
""
]
] |
2504.09631 | Victor Gabriel Morele Duarte | V. G. M. Duarte, A. J. Chaves, and N. M. R. Peres | Organic-Inorganic Polaritonics: Linking Frenkel and Wannier-Mott
Excitons | 6 pages, 3 figures | null | null | null | cond-mat.mes-hall | http://creativecommons.org/licenses/by/4.0/ | In recent years, organic materials have emerged as promising candidates for a
variety of light-harvesting applications ranging from the infrared to the
visible regions of the electromagnetic spectrum. Their enhanced excitonic
binding energies and large transition dipole moments enable strong coupling
with light, with some systems already reaching the ultrastrong coupling regime.
In contrast, a wide range of two-dimensional (2D) materials has been
extensively explored in the literature, exhibiting high exciton stability and
strong electron-hole coupling due to reduced screening effects. In this Letter,
we present a microscopic model describing the interaction of 2D materials and
organic molecular aggregates in an optical cavity. We predict the formation of
a hybrid Wannier-Mott-Frenkel exciton-polariton with an enhanced Rabi
splitting, exceeding that of the pure organic cavity by several tens of meV. To
elucidate this phenomenon, we examine a cavity with 2D tungsten sulfide and a
cyanine dye, where this enhancement corresponds to a $5\%$ increase relative to
the organic cavity. The complementary characteristics of Wannier-Mott and
Frenkel excitons enable the formation of tunable polariton states that merge
into a single hybrid state as a function of detuning, allowing for dual Rabi
splitting mechanisms. This provides a promising platform for exploring quantum
optical phenomena in both the strong and ultrastrong coupling regimes.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:06:10 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Duarte",
"V. G. M.",
""
],
[
"Chaves",
"A. J.",
""
],
[
"Peres",
"N. M. R.",
""
]
] |
2504.09632 | Sarif Khan | Laura Covi, Shyamashish Dey, Sarif Khan and Santosh Kumar Rai | Multicomponent Dark Matter with Collider Implications | 42 pages | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | The present work aims to study an extension of the Standard Model (SM) that
addresses the prominent SM shortcomings, i.e., can explain the neutrino mass,
the dark matter (DM) content, and the matter-antimatter asymmetry of the
Universe. The model introduces the possibility of a multicomponent DM scenario
leading to distinctive signals at colliders. The SM is extended by a ``dark"
$SU(2)_{D}$ gauge symmetry and new fermions and scalar doublets, charged only
under $SU(2)_{D}$, that provide candidates for a multicomponent DM. Previously,
we have considered in this model the asymmetric DM scenario, while in the
present work, we explore the symmetric DM case. We focus on the parameter
region where the dark fermion DM annihilates dominantly into the additional
dark gauge bosons and the ``inert" doublet DM annihilates to SM states via the
SM Higgs resonance. This particular choice of doublet mass ensures that the
heavier BSM Higgs always has one decay mode open to DM leading to the
possibility of detecting such particles at the LHC, in the missing energy plus
dijet ($\cancel{E}_{T}+2j$) final states. We also discuss the prospects for
detecting DM through direct and indirect detection experiments and via
Long-Lived-Particle searches. Finally, we show that, as typical of other WIMP
models, for low DM mass, signals can be expected in future collider
experiments, but for the higher mass range above $500$ GeV we have to rely
solely on direct detection experiments. Both types of experiments will be
essential to fully cover the allowed parameter space.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:06:35 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Covi",
"Laura",
""
],
[
"Dey",
"Shyamashish",
""
],
[
"Khan",
"Sarif",
""
],
[
"Rai",
"Santosh Kumar",
""
]
] |
2504.09633 | Guy Blachar | Guy Blachar, Be'eri Greenfeld | The speed of random walks on semigroups | 22 pages, 1 figure | null | null | null | math.GR math.CO math.PR | http://creativecommons.org/licenses/by/4.0/ | We construct, for each real number $0\leq \alpha \leq 1$, a random walk on a
finitely generated semigroup whose speed exponent is $\alpha$. We further show
that the speed function of a random walk on a finitely generated semigroup can
be arbitrarily slow, yet tending to infinity. These phenomena demonstrate a
sharp contrast from the group-theoretic setting. On the other hand, we show
that the distance of a random walk on a finitely generated semigroup from its
starting position is infinitely often larger than a non-constant universal
lower bound, excluding a certain degenerate case.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:10:42 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Blachar",
"Guy",
""
],
[
"Greenfeld",
"Be'eri",
""
]
] |
2504.09635 | Md. Noor-E-Alam | Sahil Shikalgar and Md. Noor-E-Alam | A Two-Stage Interpretable Matching Framework for Causal Inference | null | null | null | null | cs.AI stat.ME | http://creativecommons.org/licenses/by/4.0/ | Matching in causal inference from observational data aims to construct
treatment and control groups with similar distributions of covariates, thereby
reducing confounding and ensuring an unbiased estimation of treatment effects.
This matched sample closely mimics a randomized controlled trial (RCT), thus
improving the quality of causal estimates. We introduce a novel Two-stage
Interpretable Matching (TIM) framework for transparent and interpretable
covariate matching. In the first stage, we perform exact matching across all
available covariates. For treatment and control units without an exact match in
the first stage, we proceed to the second stage. Here, we iteratively refine
the matching process by removing the least significant confounder in each
iteration and attempting exact matching on the remaining covariates. We learn a
distance metric for the dropped covariates to quantify closeness to the
treatment unit(s) within the corresponding strata. We used these high- quality
matches to estimate the conditional average treatment effects (CATEs). To
validate TIM, we conducted experiments on synthetic datasets with varying
association structures and correlations. We assessed its performance by
measuring bias in CATE estimation and evaluating multivariate overlap between
treatment and control groups before and after matching. Additionally, we apply
TIM to a real-world healthcare dataset from the Centers for Disease Control and
Prevention (CDC) to estimate the causal effect of high cholesterol on diabetes.
Our results demonstrate that TIM improves CATE estimates, increases
multivariate overlap, and scales effectively to high-dimensional data, making
it a robust tool for causal inference in observational data.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:17:52 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Shikalgar",
"Sahil",
""
],
[
"Noor-E-Alam",
"Md.",
""
]
] |
2504.09636 | Steven Rivetti | Steven Rivetti, Ozlem Tugfe Demir, Emil Bjornson, Mikael Skoglund | Millimeter-Wave Joint Radar and Communications With an RIS-Integrated
Array | 6 pages, 4 figures, submitted to IEEE PIMRC 2025 | null | null | null | eess.SP | http://creativecommons.org/licenses/by-sa/4.0/ | In the context of the joint radar and communications (JRC) framework,
reconfigurable intelligent surfaces (RISs) emerged as a promising technology
for their ability to shape the propagation environment by adjusting their
phase-shift coefficients. However, achieving perfect synchronization and
effective collaboration between access points (APs) and RISs is crucial to
successful operation. This paper investigates the performance of a bistatic JRC
network operating in the millimeter-wave (mmWave) frequency band, where the
receiving AP is equipped with an RIS-integrated array. This system
simultaneously serves multiple UEs while estimating the position of a target
with limited prior knowledge of its position. To achieve this, we optimize both
the power allocation of the transmitted waveform and the RIS phase-shift matrix
to minimize the position error bound (PEB) of the target. At the same time, we
ensure that the UEs achieve an acceptable level of spectral efficiency. The
numerical results show that an RIS-integrated array, even with a small number
of receiving antennas, can achieve high localization accuracy. Additionally,
optimized phase-shifts significantly improve the localization accuracy in
comparison to a random phase-shift configuration.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:20:53 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Rivetti",
"Steven",
""
],
[
"Demir",
"Ozlem Tugfe",
""
],
[
"Bjornson",
"Emil",
""
],
[
"Skoglund",
"Mikael",
""
]
] |
2504.09637 | Liviu Ignat | Liviu I. Ignat and Enrique Zuazua | Optimal convergence rates for the finite element approximation of the
Sobolev constant | 32 pages | null | null | null | math.NA cs.NA math.CA | http://creativecommons.org/licenses/by/4.0/ | We establish optimal convergence rates for the P1 finite element
approximation of the Sobolev constant in arbitrary dimensions N\geq 2 and for
Lebesgue exponents 1<p<N. Our analysis relies on a refined study of the Sobolev
deficit in suitable quasi-norms, which have been introduced and utilized in the
context of finite element approximations of the p- Laplacian. The proof further
involves sharp estimates for the finite element approximation of Sobolev
minimizers.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:22:05 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Ignat",
"Liviu I.",
""
],
[
"Zuazua",
"Enrique",
""
]
] |
2504.09638 | Xunhang Sun | Xunhang Sun, Xiaoyu Cao, Bo Zeng, Miaomiao Li, Xiaohong Guan, Tamer
Ba\c{s}ar | Data-Driven Two-Stage Distributionally Robust Dispatch of Multi-Energy
Microgrid | null | null | null | null | math.OC cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper studies adaptive distributionally robust dispatch (DRD) of the
multi-energy microgrid under supply and demand uncertainties. A Wasserstein
ambiguity set is constructed to support data-driven decision-making. By fully
leveraging the special structure of worst-case expectation from the primal
perspective, a novel and high-efficient decomposition algorithm under the
framework of column-and-constraint generation is customized and developed to
address the computational burden. Numerical studies demonstrate the
effectiveness of our DRD approach, and shed light on the interrelationship of
it with the traditional dispatch approaches through stochastic programming and
robust optimization schemes. Also, comparisons with popular algorithms in the
literature for two-stage distributionally robust optimization verify the
powerful capacity of our algorithm in computing the DRD problem.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:25:34 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Sun",
"Xunhang",
""
],
[
"Cao",
"Xiaoyu",
""
],
[
"Zeng",
"Bo",
""
],
[
"Li",
"Miaomiao",
""
],
[
"Guan",
"Xiaohong",
""
],
[
"Başar",
"Tamer",
""
]
] |
2504.09639 | Yunjie Ji | Haotian Wang, Han Zhao, Shuaiting Chen, Xiaoyu Tian, Sitong Zhao,
Yunjie Ji, Yiping Peng, Xiangang Li | Leveraging Reasoning Model Answers to Enhance Non-Reasoning Model
Capability | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advancements in large language models (LLMs), such as DeepSeek-R1 and
OpenAI-o1, have demonstrated the significant effectiveness of test-time
scaling, achieving substantial performance gains across various benchmarks.
These advanced models utilize deliberate "thinking" steps to systematically
enhance answer quality. In this paper, we propose leveraging these high-quality
outputs generated by reasoning-intensive models to improve less computationally
demanding, non-reasoning models. We explore and compare methodologies for
utilizing the answers produced by reasoning models to train and improve
non-reasoning models. Through straightforward Supervised Fine-Tuning (SFT)
experiments on established benchmarks, we demonstrate consistent improvements
across various benchmarks, underscoring the potential of this approach for
advancing the ability of models to answer questions directly.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:26:56 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wang",
"Haotian",
""
],
[
"Zhao",
"Han",
""
],
[
"Chen",
"Shuaiting",
""
],
[
"Tian",
"Xiaoyu",
""
],
[
"Zhao",
"Sitong",
""
],
[
"Ji",
"Yunjie",
""
],
[
"Peng",
"Yiping",
""
],
[
"Li",
"Xiangang",
""
]
] |
2504.09640 | Jian-Dong Zhang | Yi-Xiang Wei, Jian-dong Zhang, Jianwei Mei | Second-Order Self-Force for Eccentric Extreme-Mass Ratio Inspirals on
Schwarzschild Spacetime | 14 pages | null | null | null | gr-qc | http://creativecommons.org/licenses/by/4.0/ | Extreme Mass Ratio Inspiral (EMRI), which corresponding to a small compact
object inspirals around a Massive Black Hole (MBH) in the center of a galaxy,
is one of the most important kind of source for future space-borne
gravitational-wave (GW) detectors such as TianQin and LISA. By analyzing the
emitted GW signals, we can probe the theory of gravity and the nature of black
holes (BHs) in the strong field region. To achieve these objectives, the second
order self-force effect should be considered in the waveform modelling. Up to
now, the waveform of EMRIs including the second order self-force effect is only
achieved for the circular orbit on Schwarzschild background. In this work, we
generalized the calculation of the second order self-force to the eccentric
orbits on Schwarzschild spacetime. We calculated the puncture field, and give
the form of two timescale expansion for the field equations. The corresponding
numerical calculation and programming can be performed based on these results.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:27:35 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wei",
"Yi-Xiang",
""
],
[
"Zhang",
"Jian-dong",
""
],
[
"Mei",
"Jianwei",
""
]
] |
2504.09641 | Xingjian Zhang | Xingjian Zhang and Siwei Wen and Wenjun Wu and Lei Huang | TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, improving the reasoning ability of large multimodal models (LMMs)
through reinforcement learning has made great progress. However, most existing
works are based on highly reasoning-intensive datasets such as mathematics and
code, and researchers generally choose large-scale models as the foundation. We
argue that exploring small-scale models' reasoning capabilities remains
valuable for researchers with limited computational resources. Moreover,
enabling models to explain their reasoning processes on general
question-answering datasets is equally meaningful. Therefore, we present the
small-scale video reasoning model TinyLLaVA-Video-R1. Based on TinyLLaVA-Video,
a traceably trained video understanding model with no more than 4B parameters,
it not only demonstrates significantly improved reasoning and thinking
capabilities after using reinforcement learning on general Video-QA datasets,
but also exhibits the emergent characteristic of "aha moments". Furthermore, we
share a series of experimental findings, aiming to provide practical insights
for future exploration of video reasoning (thinking) abilities in small-scale
models. It is available at https://github.com/ZhangXJ199/TinyLLaVA-Video-R1.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:32:49 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhang",
"Xingjian",
""
],
[
"Wen",
"Siwei",
""
],
[
"Wu",
"Wenjun",
""
],
[
"Huang",
"Lei",
""
]
] |
2504.09642 | Micha{\l} Kruszewski | Micha{\l} Kruszewski | HBS -- Hardware Build System: A Tcl-based, minimal common abstraction
approach for build system for hardware designs | null | null | null | null | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Build systems become an indispensable part of the software implementation and
deployment process. New programming languages are released with the build
system integrated into the language tools, for example, Go, Rust, or Zig.
However, in the hardware description domain, no official build systems have
been released with the predominant Hardware Description Languages (HDL) such as
VHDL or SystemVerilog. Moreover, hardware design projects are often
multilanguage.
The paper proposes a new build system for the hardware description domain.
The system is called the Hardware Build System (HBS). The main goals of the
system include simplicity, readability, a minimal number of dependencies, and
ease of integration with the existing Electronic Design Automation (EDA) tools.
The system proposes a novel, minimal common abstraction approach, whose
particular implications are described in the article. All the core
functionalities are implemented in Tcl. Only the EDA tool's independent
features, such as dependency graph generation, are implemented in a Python
wrapper.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:34:14 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Kruszewski",
"Michał",
""
]
] |
2504.09643 | Valentin Malykh | Nikita Sorokin and Ivan Sedykh and Valentin Malykh | Iterative Self-Training for Code Generation via Reinforced Re-Ranking | Published at ECIR 2025 | null | 10.1007/978-3-031-88714-7_21 | null | cs.CL cs.IR cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generating high-quality code that solves complex programming tasks is
challenging, especially with current decoder-based models that produce highly
stochastic outputs. In code generation, even minor errors can easily break the
entire solution. Leveraging multiple sampled solutions can significantly
improve the overall output quality.
One effective way to enhance code generation is by pairing a code generation
model with a reranker model, which selects the best solution from the generated
samples. We propose a novel iterative self-training approach for self-training
reranker models using Proximal Policy Optimization (PPO), aimed at improving
both reranking accuracy and the overall code generation process. Unlike
traditional PPO approaches, where the focus is on optimizing a generative model
with a reward model, our approach emphasizes the development of a robust
reward/reranking model. This model improves the quality of generated code
through reranking and addresses problems and errors that the reward model might
overlook during PPO alignment with the reranker. Our method iteratively refines
the training dataset by re-evaluating outputs, identifying high-scoring
negative examples, and incorporating them into the training loop, that boosting
model performance.
Our evaluation on the MultiPL-E dataset demonstrates that our 13.4B parameter
model outperforms a 33B model in code generation quality while being three
times faster. Moreover, it achieves performance comparable to GPT-4 and
surpasses it in one programming language.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:34:17 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Sorokin",
"Nikita",
""
],
[
"Sedykh",
"Ivan",
""
],
[
"Malykh",
"Valentin",
""
]
] |
2504.09644 | Kaiyu Li | Kaiyu Li, Zepeng Xin, Li Pang, Chao Pang, Yupeng Deng, Jing Yao,
Guisong Xia, Deyu Meng, Zhi Wang, Xiangyong Cao | SegEarth-R1: Geospatial Pixel Reasoning via Large Language Model | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Remote sensing has become critical for understanding environmental dynamics,
urban planning, and disaster management. However, traditional remote sensing
workflows often rely on explicit segmentation or detection methods, which
struggle to handle complex, implicit queries that require reasoning over
spatial context, domain knowledge, and implicit user intent. Motivated by this,
we introduce a new task, \ie, geospatial pixel reasoning, which allows implicit
querying and reasoning and generates the mask of the target region. To advance
this task, we construct and release the first large-scale benchmark dataset
called EarthReason, which comprises 5,434 manually annotated image masks with
over 30,000 implicit question-answer pairs. Moreover, we propose SegEarth-R1, a
simple yet effective language-guided segmentation baseline that integrates a
hierarchical visual encoder, a large language model (LLM) for instruction
parsing, and a tailored mask generator for spatial correlation. The design of
SegEarth-R1 incorporates domain-specific adaptations, including aggressive
visual token compression to handle ultra-high-resolution remote sensing images,
a description projection module to fuse language and multi-scale features, and
a streamlined mask prediction pipeline that directly queries description
embeddings. Extensive experiments demonstrate that SegEarth-R1 achieves
state-of-the-art performance on both reasoning and referring segmentation
tasks, significantly outperforming traditional and LLM-based segmentation
methods. Our data and code will be released at
https://github.com/earth-insights/SegEarth-R1.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:36:47 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Li",
"Kaiyu",
""
],
[
"Xin",
"Zepeng",
""
],
[
"Pang",
"Li",
""
],
[
"Pang",
"Chao",
""
],
[
"Deng",
"Yupeng",
""
],
[
"Yao",
"Jing",
""
],
[
"Xia",
"Guisong",
""
],
[
"Meng",
"Deyu",
""
],
[
"Wang",
"Zhi",
""
],
[
"Cao",
"Xiangyong",
""
]
] |
2504.09645 | Aung Kyaw Htet | Aung Kyaw Htet and Mark Dras | Myanmar XNLI: Building a Dataset and Exploring Low-resource Approaches
to Natural Language Inference with Myanmar | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Despite dramatic recent progress in NLP, it is still a major challenge to
apply Large Language Models (LLM) to low-resource languages. This is made
visible in benchmarks such as Cross-Lingual Natural Language Inference (XNLI),
a key task that demonstrates cross-lingual capabilities of NLP systems across a
set of 15 languages. In this paper, we extend the XNLI task for one additional
low-resource language, Myanmar, as a proxy challenge for broader low-resource
languages, and make three core contributions. First, we build a dataset called
Myanmar XNLI (myXNLI) using community crowd-sourced methods, as an extension to
the existing XNLI corpus. This involves a two-stage process of community-based
construction followed by expert verification; through an analysis, we
demonstrate and quantify the value of the expert verification stage in the
context of community-based construction for low-resource languages. We make the
myXNLI dataset available to the community for future research. Second, we carry
out evaluations of recent multilingual language models on the myXNLI benchmark,
as well as explore data-augmentation methods to improve model performance. Our
data-augmentation methods improve model accuracy by up to 2 percentage points
for Myanmar, while uplifting other languages at the same time. Third, we
investigate how well these data-augmentation methods generalise to other
low-resource languages in the XNLI dataset.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:36:59 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Htet",
"Aung Kyaw",
""
],
[
"Dras",
"Mark",
""
]
] |
2504.09646 | Tuhin G M Al Mamun | Tuhin G M Al Mamun | Replacing ARDL? Introducing the NSB-ARDL Model for Structural and
Asymmetric Forecasting | draft | null | null | null | stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces the NSB-ARDL (Nonlinear Structural Break Autoregressive
Distributed Lag) model, a novel econometric framework designed to capture
asymmetric and nonlinear dynamics in macroeconomic time series. Traditional
ARDL models, while widely used for estimating short- and long-run
relationships, rely on assumptions of linearity and symmetry that may overlook
critical structural features in real-world data. The NSB-ARDL model overcomes
these limitations by decomposing explanatory variables into cumulative positive
and negative partial sums, enabling the identification of both short- and
long-term asymmetries.
Monte Carlo simulations show that NSB-ARDL consistently outperforms
conventional ARDL models in terms of forecasting accuracy when asymmetric
responses are present in the data-generating process. An empirical application
to South Korea's CO2 emissions demonstrates the model's practical advantages,
yielding a better in-sample fit and more interpretable long-run coefficients.
These findings highlight the NSB-ARDL model as a structurally robust and
forecasting-efficient alternative for analyzing nonlinear macroeconomic
phenomena.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:40:17 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Mamun",
"Tuhin G M Al",
""
]
] |
2504.09647 | Yun Tang | Yun Tang, Mengbang Zou, Udhaya Chandhar Srinivasan, Obumneme Umealor,
Dennis Kevogo, Benjamin James Scott, Weisi Guo | Building AI Service Repositories for On-Demand Service Orchestration in
6G AI-RAN | 6 pages, three figures, one table, submitted to IEEE GlobeCOM 2025
for possible publication | null | null | null | cs.AI cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Efficient orchestration of AI services in 6G AI-RAN requires well-structured,
ready-to-deploy AI service repositories combined with orchestration methods
adaptive to diverse runtime contexts across radio access, edge, and cloud
layers. Current literature lacks comprehensive frameworks for constructing such
repositories and generally overlooks key practical orchestration factors. This
paper systematically identifies and categorizes critical attributes influencing
AI service orchestration in 6G networks and introduces an open-source,
LLM-assisted toolchain that automates service packaging, deployment, and
runtime profiling. We validate the proposed toolchain through the Cranfield AI
Service repository case study, demonstrating significant automation benefits,
reduced manual coding efforts, and the necessity of infrastructure-specific
profiling, paving the way for more practical orchestration frameworks.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:40:58 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Tang",
"Yun",
""
],
[
"Zou",
"Mengbang",
""
],
[
"Srinivasan",
"Udhaya Chandhar",
""
],
[
"Umealor",
"Obumneme",
""
],
[
"Kevogo",
"Dennis",
""
],
[
"Scott",
"Benjamin James",
""
],
[
"Guo",
"Weisi",
""
]
] |
2504.09648 | Salar Fattahi | Guixian Chen, Jianhao Ma, Salar Fattahi | RANSAC Revisited: An Improved Algorithm for Robust Subspace Recovery
under Adversarial and Noisy Corruptions | null | null | null | null | cs.LG cs.CV stat.CO | http://creativecommons.org/licenses/by/4.0/ | In this paper, we study the problem of robust subspace recovery (RSR) in the
presence of both strong adversarial corruptions and Gaussian noise.
Specifically, given a limited number of noisy samples -- some of which are
tampered by an adaptive and strong adversary -- we aim to recover a
low-dimensional subspace that approximately contains a significant fraction of
the uncorrupted samples, up to an error that scales with the Gaussian noise.
Existing approaches to this problem often suffer from high computational costs
or rely on restrictive distributional assumptions, limiting their applicability
in truly adversarial settings. To address these challenges, we revisit the
classical random sample consensus (RANSAC) algorithm, which offers strong
robustness to adversarial outliers, but sacrifices efficiency and robustness
against Gaussian noise and model misspecification in the process. We propose a
two-stage algorithm, RANSAC+, that precisely pinpoints and remedies the failure
modes of standard RANSAC. Our method is provably robust to both Gaussian and
adversarial corruptions, achieves near-optimal sample complexity without
requiring prior knowledge of the subspace dimension, and is more efficient than
existing RANSAC-type methods.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:47:03 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Chen",
"Guixian",
""
],
[
"Ma",
"Jianhao",
""
],
[
"Fattahi",
"Salar",
""
]
] |
2504.09649 | Marcelo Byrro Ribeiro | Bruno J. Souza, Osvaldo L. Santos-Pereira and Marcelo B. Ribeiro | Galaxy Mergers in a Fractal Cosmology | 25 pages. 2 figures. LaTeX. Accepted for publication in the "European
Physical Journal C" | null | null | null | astro-ph.CO gr-qc | http://creativecommons.org/licenses/by/4.0/ | This work discusses the influence of galaxy mergers in the evolution of a
parabolic Lema\^{\i}tre-Tolman-Bondi (LTB) cosmology with simultaneous big bang
endowed with two consecutive single fractal galaxy distributions systems
possessing fractal dimension $D$. Based on recent empirical findings, it is
assumed that the resulting galaxy mass from mergers can be expressed by a
redshift dependent decaying power law. The proposed cosmological model modifies
the relativistic fractal number counts distribution by including a merger rate
evolution that estimates the model's radial density. Numerical solutions for
the first order small-merger-rate approximation (SMRA) are found and the
results show that a fractal galaxy distribution having $D=1.5$ in the range
$0.1<z<1.0$, and $D=0.5$ for $1<z<6$, as suggested by recent empirical
findings, the SMRA allows consistent description of the model for a merger rate
power law exponent up to $q=0.2$ considering a fractal galaxy distribution
starting from the Local Group. Consistent values were also found up to $q=2.5$
and $z=7$ from a scale smaller than the Local Supercluster. These results show
that galaxy mergers can be successfully incorporated into the dynamics of a
parabolic LTB fractal cosmology.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:50:35 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Souza",
"Bruno J.",
""
],
[
"Santos-Pereira",
"Osvaldo L.",
""
],
[
"Ribeiro",
"Marcelo B.",
""
]
] |
2504.09650 | Stephan Baier | Stephan Baier and Habibur Rahaman | Diophantine approximation with integers represented by binary quadratic
forms | 14 pages | null | null | null | math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For any given positive definite binary quadratic form (PBQF), we prove that
for every irrational number $\alpha$, there exist infinitely many positive
integers $n$ represented by this PBQF and satisfying $||\alpha
n||<n^{-3/7+\varepsilon}$ for any fixed but arbitrarily small $\varepsilon>0$.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:51:31 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Baier",
"Stephan",
""
],
[
"Rahaman",
"Habibur",
""
]
] |
2504.09651 | Avinash Patil | Avinash Patil | GitBugs: Bug Reports for Duplicate Detection, Retrieval Augmented
Generation, Triage, and More | 2 Tables, 9 Figures | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Bug reports provide critical insights into software quality, yet existing
datasets often suffer from limited scope, outdated content, or insufficient
metadata for machine learning. To address these limitations, we present
GitBugs-a comprehen- sive and up-to-date dataset comprising over 150,000 bug
reports from nine actively maintained open-source projects, including Firefox,
Cassandra, and VS Code. GitBugs aggregates data from Github, Bugzilla and Jira
issue trackers, offering standardized categorical fields for classification
tasks and predefined train/test splits for duplicate bug detection. In
addition, it includes ex- ploratory analysis notebooks and detailed
project-level statistics, such as duplicate rates and resolution times. GitBugs
supports various software engineering research tasks, including duplicate
detection, retrieval augmented generation, resolution prediction, automated
triaging, and temporal analysis. The openly licensed dataset provides a
valuable cross-project resource for bench- marking and advancing automated bug
report analysis. Access the data and code at
https://github.com/av9ash/gitbugs/.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:55:28 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Patil",
"Avinash",
""
]
] |
2504.09652 | Faisal Haque Bappy | Tahrim Hossain, Sakib Hassan, Faisal Haque Bappy, Muhammad Nur
Yanhaona, Tarannum Shaila Zaman, and Tariqul Islam | Bridging Immutability with Flexibility: A Scheme for Secure and
Efficient Smart Contract Upgrades | null | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The emergence of blockchain technology has revolutionized contract execution
through the introduction of smart contracts. Ethereum, the leading blockchain
platform, leverages smart contracts to power decentralized applications
(DApps), enabling transparent and self-executing systems across various
domains. While the immutability of smart contracts enhances security and trust,
it also poses significant challenges for updates, defect resolution, and
adaptation to changing requirements. Existing upgrade mechanisms are complex,
resource-intensive, and costly in terms of gas consumption, often compromising
security and limiting practical adoption. To address these challenges, we
propose FlexiContracts+, a novel scheme that reimagines smart contracts by
enabling secure, in-place upgrades on Ethereum while preserving historical data
without relying on multiple contracts or extensive pre-deployment planning.
FlexiContracts+ enhances security, simplifies development, reduces engineering
overhead, and supports adaptable, expandable smart contracts. Comprehensive
testing demonstrates that FlexiContracts+ achieves a practical balance between
immutability and flexibility, advancing the capabilities of smart contract
systems.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 16:59:28 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Hossain",
"Tahrim",
""
],
[
"Hassan",
"Sakib",
""
],
[
"Bappy",
"Faisal Haque",
""
],
[
"Yanhaona",
"Muhammad Nur",
""
],
[
"Zaman",
"Tarannum Shaila",
""
],
[
"Islam",
"Tariqul",
""
]
] |
2504.09653 | Mohsen Farsiani | Mohsen Farsiani and Hossein M. Shodja | Multiphysics analysis of acoustically actuated nanospherical antennas
embedded in polymer/metal medium with magneto-electro-elastic
surface/interface effects | 53 pages, 12 figures | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A precise analytical treatment for predicting the behavior of nano-sized
magneto-electro-elastic (MEE) antennas and resonators under incident acoustic
waves requires consideration of multiphysics surface/interface effects,
including magnetization, polarization, and elasticity. No analytical solutions
to date have incorporated all three phenomena simultaneously. This work
presents a rigorous mathematical analysis of a nano-sized spherically isotropic
embedded MEE spherical shell subjected to acoustic waves. The study
distinguishes between the coupled spectral constitutive relations for the bulk
of the MEE shell and those for its free inner surface and matrix-shell
interface. The surrounding matrix may be isotropic dielectric or metallic.
Conventional electrodynamics theories do not address MEE effects at the surface
or interface. To overcome this, the equivalent impedance matrix (EIM) method
combined with surface/interface elasticity is used to model the MEE behaviors.
For metallic matrices, a plasmonics-based framework with optical properties
described by the plasma model captures metallic behavior. The spectral EIM
method, along with vector and tensor spherical harmonics, solves the coupled
elastodynamics and Maxwell's equations. This approach enables the exploration
of surface/interface characteristic lengths, revealing size-dependent effects
on electromagnetic radiated power and resonance frequency. The findings provide
insights into the behavior of acoustically actuated nanospherical antennas,
sensors, and resonators, with implications for the design of nanoscale devices
in advanced technological applications.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:00:26 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Farsiani",
"Mohsen",
""
],
[
"Shodja",
"Hossein M.",
""
]
] |
2504.09654 | Meng Zhou | Meng Zhou, Shuangge Ma, Mengyun Wu | Integrated Bayesian non-parametric spatial modeling for cross-sample
identification of spatially variable genes | null | null | null | null | stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spatial transcriptomics has revolutionized tissue analysis by simultaneously
mapping gene expression, spatial topography, and histological context across
consecutive tissue sections, enabling systematic investigation of spatial
heterogeneity. The detection of spatially variable (SV) genes-molecular
signatures with position-dependent expression-provides critical insights into
disease mechanisms spanning oncology, neurology, and cardiovascular research.
Current methodologies, however, confront dual constraints: predominant reliance
on predefined spatial pattern templates restricts detection of novel complex
spatial architectures, and inconsistent sample selection strategies compromise
analytical stability and biological interpretability. To overcome these
challenges, we propose a novel Bayesian hierarchical framework incorporating
non-parametric spatial modeling and across-sample integration. It takes
advantage of the non-parametric technique and develops an adaptive spatial
process accommodating complex pattern discovery. A novel cross-sample bi-level
shrinkage prior is further introduced for robust multi-sample SV gene
detection, facilitating more effective information fusion. An efficient
variational inference is developed for posterior inference ensuring
computational scalability. This architecture synergistically addresses spatial
complexity through adaptive pattern learning while maintaining biological
interpretability. Comprehensive simulations and empirical analyzes confirm the
improved performance of the proposed method in resolving complex spatial
expression patterns compared to existing analytical frameworks.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:00:52 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhou",
"Meng",
""
],
[
"Ma",
"Shuangge",
""
],
[
"Wu",
"Mengyun",
""
]
] |
2504.09656 | Xingrui Wang | Xingrui Wang, Jiang Liu, Ze Wang, Xiaodong Yu, Jialian Wu, Ximeng Sun,
Yusheng Su, Alan Yuille, Zicheng Liu, Emad Barsoum | KeyVID: Keyframe-Aware Video Diffusion for Audio-Synchronized Visual
Animation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Generating video from various conditions, such as text, image, and audio,
enables both spatial and temporal control, leading to high-quality generation
results. Videos with dramatic motions often require a higher frame rate to
ensure smooth motion. Currently, most audio-to-visual animation models use
uniformly sampled frames from video clips. However, these uniformly sampled
frames fail to capture significant key moments in dramatic motions at low frame
rates and require significantly more memory when increasing the number of
frames directly. In this paper, we propose KeyVID, a keyframe-aware
audio-to-visual animation framework that significantly improves the generation
quality for key moments in audio signals while maintaining computation
efficiency. Given an image and an audio input, we first localize keyframe time
steps from the audio. Then, we use a keyframe generator to generate the
corresponding visual keyframes. Finally, we generate all intermediate frames
using the motion interpolator. Through extensive experiments, we demonstrate
that KeyVID significantly improves audio-video synchronization and video
quality across multiple datasets, particularly for highly dynamic motions. The
code is released in https://github.com/XingruiWang/KeyVID.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:06:03 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wang",
"Xingrui",
""
],
[
"Liu",
"Jiang",
""
],
[
"Wang",
"Ze",
""
],
[
"Yu",
"Xiaodong",
""
],
[
"Wu",
"Jialian",
""
],
[
"Sun",
"Ximeng",
""
],
[
"Su",
"Yusheng",
""
],
[
"Yuille",
"Alan",
""
],
[
"Liu",
"Zicheng",
""
],
[
"Barsoum",
"Emad",
""
]
] |
2504.09657 | Francesco Popolizio | Francesco Popolizio, Torsten Wik, Chih Feng Lee, Changfu Zou | Nonlinear Online Optimization for Vehicle-Home-Grid Integration
including Household Load Prediction and Battery Degradation | Submitted to the 2025 IEEE Conference on Decision and Control (CDC) | null | null | null | eess.SY cs.SY math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper investigates the economic impact of vehicle-home-grid integration,
by proposing an online energy management algorithm that optimizes energy flows
between an electric vehicle (EV), a household, and the electrical grid. The
algorithm leverages vehicle-to-home (V2H) for self-consumption and
vehicle-to-grid (V2G) for energy trading, adapting to real-time conditions
through a hybrid long short-term memory (LSTM) neural network for accurate
household load prediction, alongside a comprehensive nonlinear battery
degradation model accounting for both cycle and calendar aging. Simulation
results reveal significant economic advantages: compared to smart
unidirectional charging, the proposed method yields an annual economic benefit
of up to EUR 3046.81, despite a modest 1.96% increase in battery degradation.
Even under unfavorable market conditions, where V2G energy selling generates no
revenue, V2H alone ensures yearly savings of EUR 425.48. A systematic
sensitivity analysis investigates how variations in battery capacity, household
load, and price ratios affect economic outcomes, confirming the consistent
benefits of bidirectional energy exchange. These findings highlight the
potential of EVs as active energy nodes, enabling sustainable energy management
and cost-effective battery usage in real-world conditions.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:11:28 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Popolizio",
"Francesco",
""
],
[
"Wik",
"Torsten",
""
],
[
"Lee",
"Chih Feng",
""
],
[
"Zou",
"Changfu",
""
]
] |
2504.09658 | Michael Assaf | Elad Korngut and Michael Assaf | Impact of network assortativity on disease lifetime in the SIS model of
epidemics | 10 pages, 6 figures | null | null | null | cond-mat.stat-mech physics.comp-ph q-bio.PE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | To accurately represent disease spread, epidemiological models must account
for the complex network topology and contact heterogeneity. Traditionally, most
studies have used random heterogeneous networks, which ignore correlations
between the nodes' degrees. Yet, many real-world networks exhibit degree
assortativity - the tendency for nodes with similar degrees to connect. Here we
explore the effect degree assortativity (or disassortativity) has on long-term
dynamics and disease extinction in the realm of the
susceptible-infected-susceptible model on heterogeneous networks. We derive
analytical results for the mean time to extinction (MTE) in assortative
networks with weak heterogeneity, and show that increased assortativity reduces
the MTE and that assortativity and degree heterogeneity are interchangeable
with regard to their impact on the MTE. Our analytical results are verified
using the weighted ensemble numerical method, on both synthetic and real-world
networks. Notably, this method allows us to go beyond the capabilities of
traditional numerical tools, enabling us to study rare events in large
assortative networks, which were previously inaccessible.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:14:27 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Korngut",
"Elad",
""
],
[
"Assaf",
"Michael",
""
]
] |
2504.09659 | Hayk Soghomonyan | Hayk Soghomonyan, Justinas Pupeikis, Benjamin Willenberg, Armin
Stumpp, Lukas Lang, Christopher R. Phillips, Bojan Resan, Ursula Keller | 3D in-situ profiling in a laser micromachining station using dual-comb
LiDAR | The generated pdf contains the main part (7 pages) and an appendix (2
pages). There are 6 figures in the main part and 3 figures (one with two
subfigures) in the appendix | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We demonstrate the integration of coaxial dual-comb LiDAR into a laser
micromachining station, enabling 3D profiling with sub-micron axial precision
without moving the machined piece. This setup facilitates in-situ
nondestructive testing (NDT) and evaluation, reducing the effort and time
required for micromachining process development and control.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:14:47 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Soghomonyan",
"Hayk",
""
],
[
"Pupeikis",
"Justinas",
""
],
[
"Willenberg",
"Benjamin",
""
],
[
"Stumpp",
"Armin",
""
],
[
"Lang",
"Lukas",
""
],
[
"Phillips",
"Christopher R.",
""
],
[
"Resan",
"Bojan",
""
],
[
"Keller",
"Ursula",
""
]
] |
2504.09660 | Fallou Niakh | Fallou Niakh, Alicia Bassi\`ere, Michel Denuit, Christian Robert | Peer-to-Peer Basis Risk Management for Renewable Production Parametric
Insurance | null | null | null | null | stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work presents a framework for peer-to-peer (P2P) basis risk management
applied to solar electricity generation. The approach leverages physically
based simulation models to estimate the day-ahead production forecasts and the
actual realized production at the solar farm level. We quantify the financial
loss from mismatches between forecasted and actual production using the outputs
of these simulations. The framework then implements a parametric insurance
mechanism to mitigate these financial losses and combines it with a P2P market
structure to enhance participant risk sharing. By integrating day-ahead
forecasts and actual production data with physical modeling, this method
provides a comprehensive solution to manage production variability, offering
practical insights for improving financial resilience in renewable energy
systems. The results highlight the potential of combining parametric insurance
with P2P mechanisms to foster reliability and collaboration in renewable energy
markets.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:14:48 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Niakh",
"Fallou",
""
],
[
"Bassière",
"Alicia",
""
],
[
"Denuit",
"Michel",
""
],
[
"Robert",
"Christian",
""
]
] |
2504.09661 | Ilmar Gahramanov | O\u{g}uz Alp A\u{g}{\i}rba\c{s}, An{\i}l Ata, Eren Demirci, Ilmar
Gahramanov, Tu\u{g}ba H{\i}rl{\i}, R. Semih Kanber, Ahmet Berk Kavruk,
Mustafa Mullahasano\u{g}lu, Zehra \"Ozcan, Cansu \"Ozdemir, Irmak
\"Ozg\"u\c{c}, Sinan Ula\c{s} \"Ozt\"urk, Uveys Turhan, Ali Mert T. Yetkin,
Yunus Emre Y{\i}ld{\i}r{\i}m, and Reyhan Yumu\c{s}ak | Ising 100: review of solutions | 158 pages | null | null | null | math-ph cond-mat.stat-mech hep-th math.MP nlin.SI | http://creativecommons.org/licenses/by/4.0/ | We present several known solutions to the two-dimensional Ising model. This
review originated from the ``Ising 100'' seminar series held at
Bo\u{g}azi\c{c}i University, Istanbul, in 2024.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:15:34 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Ağırbaş",
"Oğuz Alp",
""
],
[
"Ata",
"Anıl",
""
],
[
"Demirci",
"Eren",
""
],
[
"Gahramanov",
"Ilmar",
""
],
[
"Hırlı",
"Tuğba",
""
],
[
"Kanber",
"R. Semih",
""
],
[
"Kavruk",
"Ahmet Berk",
""
],
[
"Mullahasanoğlu",
"Mustafa",
""
],
[
"Özcan",
"Zehra",
""
],
[
"Özdemir",
"Cansu",
""
],
[
"Özgüç",
"Irmak",
""
],
[
"Öztürk",
"Sinan Ulaş",
""
],
[
"Turhan",
"Uveys",
""
],
[
"Yetkin",
"Ali Mert T.",
""
],
[
"Yıldırım",
"Yunus Emre",
""
],
[
"Yumuşak",
"Reyhan",
""
]
] |
2504.09662 | Jenny Ma | Jenny Ma, Riya Sahni, Karthik Sreedhar, Lydia B. Chilton | AgentDynEx: Nudging the Mechanics and Dynamics of Multi-Agent
Simulations | null | null | null | null | cs.MA cs.AI cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-agent large language model simulations have the potential to model
complex human behaviors and interactions. If the mechanics are set up properly,
unanticipated and valuable social dynamics can surface. However, it is
challenging to consistently enforce simulation mechanics while still allowing
for notable and emergent dynamics. We present AgentDynEx, an AI system that
helps set up simulations from user-specified mechanics and dynamics. AgentDynEx
uses LLMs to guide users through a Configuration Matrix to identify core
mechanics and define milestones to track dynamics. It also introduces a method
called \textit{nudging}, where the system dynamically reflects on simulation
progress and gently intervenes if it begins to deviate from intended outcomes.
A technical evaluation found that nudging enables simulations to have more
complex mechanics and maintain its notable dynamics compared to simulations
without nudging. We discuss the importance of nudging as a technique for
balancing mechanics and dynamics of multi-agent simulations.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:26:35 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Ma",
"Jenny",
""
],
[
"Sahni",
"Riya",
""
],
[
"Sreedhar",
"Karthik",
""
],
[
"Chilton",
"Lydia B.",
""
]
] |
2504.09663 | Philippe Goulet Coulombe | Philippe Goulet Coulombe | Ordinary Least Squares as an Attention Mechanism | null | null | null | null | cs.LG econ.EM math.ST stat.ML stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | I show that ordinary least squares (OLS) predictions can be rewritten as the
output of a restricted attention module, akin to those forming the backbone of
large language models. This connection offers an alternative perspective on
attention beyond the conventional information retrieval framework, making it
more accessible to researchers and analysts with a background in traditional
statistics. It falls into place when OLS is framed as a similarity-based method
in a transformed regressor space, distinct from the standard view based on
partial correlations. In fact, the OLS solution can be recast as the outcome of
an alternative problem: minimizing squared prediction errors by optimizing the
embedding space in which training and test vectors are compared via inner
products. Rather than estimating coefficients directly, we equivalently learn
optimal encoding and decoding operations for predictors. From this vantage
point, OLS maps naturally onto the query-key-value structure of attention
mechanisms. Building on this foundation, I discuss key elements of
Transformer-style attention and draw connections to classic ideas from time
series econometrics.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:26:44 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Coulombe",
"Philippe Goulet",
""
]
] |
2504.09664 | Anxian Liu | Anxian Liu, Junying Ma, Guang Zhang | Adapting to the Unknown: Robust Meta-Learning for Zero-Shot Financial
Time Series Forecasting | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Financial time series forecasting in the zero-shot setting is essential for
risk management and investment decision-making, particularly during abrupt
market regime shifts or in emerging markets with limited historical data. While
Model-Agnostic Meta-Learning (MAML)-based approaches have shown promise in this
domain, existing meta task construction strategies often lead to suboptimal
performance, especially when dealing with highly turbulent financial time
series. To address this challenge, we propose a novel task construction method
that leverages learned embeddings for more effective meta-learning in the
zero-shot setting. Specifically, we construct two complementary types of
meta-tasks based on the learned embeddings: intra-cluster tasks and
inter-cluster tasks. To capture diverse fine-grained patterns, we apply
stochastic projection matrices to the learned embeddings and use clustering
algorithm to form the tasks. Additionally, to improve generalization
capabilities, we employ hard task mining strategies and leverage inter-cluster
tasks to identify invariant patterns across different time series. Extensive
experiments on the real world financial dataset demonstrate that our method
significantly outperforms existing approaches, showing better generalization
ability in the zero-shot scenario.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:27:07 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Liu",
"Anxian",
""
],
[
"Ma",
"Junying",
""
],
[
"Zhang",
"Guang",
""
]
] |
2504.09665 | Guanming Xiong | Liqiang Wen, Guanming Xiong, Tong Mo, Bing Li, Weiping Li, Wen Zhao | CLEAR-KGQA: Clarification-Enhanced Ambiguity Resolution for Knowledge
Graph Question Answering | This work has been accepted by the IJCNN 2025 main track | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This study addresses the challenge of ambiguity in knowledge graph question
answering (KGQA). While recent KGQA systems have made significant progress,
particularly with the integration of large language models (LLMs), they
typically assume user queries are unambiguous, which is an assumption that
rarely holds in real-world applications. To address these limitations, we
propose a novel framework that dynamically handles both entity ambiguity (e.g.,
distinguishing between entities with similar names) and intent ambiguity (e.g.,
clarifying different interpretations of user queries) through interactive
clarification. Our approach employs a Bayesian inference mechanism to quantify
query ambiguity and guide LLMs in determining when and how to request
clarification from users within a multi-turn dialogue framework. We further
develop a two-agent interaction framework where an LLM-based user simulator
enables iterative refinement of logical forms through simulated user feedback.
Experimental results on the WebQSP and CWQ dataset demonstrate that our method
significantly improves performance by effectively resolving semantic
ambiguities. Additionally, we contribute a refined dataset of disambiguated
queries, derived from interaction histories, to facilitate future research in
this direction.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:34:35 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Wen",
"Liqiang",
""
],
[
"Xiong",
"Guanming",
""
],
[
"Mo",
"Tong",
""
],
[
"Li",
"Bing",
""
],
[
"Li",
"Weiping",
""
],
[
"Zhao",
"Wen",
""
]
] |
2504.09666 | Yao Yuan | Yao Yuan and Pan Gao and Qun Dai and Jie Qin and Wei Xiang | Uncertainty Guided Refinement for Fine-Grained Salient Object Detection | IEEE Transactions on Image Processing 2025 | null | 10.1109/TIP.2025.3557562 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, salient object detection (SOD) methods have achieved impressive
performance. However, salient regions predicted by existing methods usually
contain unsaturated regions and shadows, which limits the model for reliable
fine-grained predictions. To address this, we introduce the uncertainty
guidance learning approach to SOD, intended to enhance the model's perception
of uncertain regions. Specifically, we design a novel Uncertainty Guided
Refinement Attention Network (UGRAN), which incorporates three important
components, i.e., the Multilevel Interaction Attention (MIA) module, the Scale
Spatial-Consistent Attention (SSCA) module, and the Uncertainty Refinement
Attention (URA) module. Unlike conventional methods dedicated to enhancing
features, the proposed MIA facilitates the interaction and perception of
multilevel features, leveraging the complementary characteristics among
multilevel features. Then, through the proposed SSCA, the salient information
across diverse scales within the aggregated features can be integrated more
comprehensively and integrally. In the subsequent steps, we utilize the
uncertainty map generated from the saliency prediction map to enhance the
model's perception capability of uncertain regions, generating a
highly-saturated fine-grained saliency prediction map. Additionally, we devise
an adaptive dynamic partition (ADP) mechanism to minimize the computational
overhead of the URA module and improve the utilization of uncertainty guidance.
Experiments on seven benchmark datasets demonstrate the superiority of the
proposed UGRAN over the state-of-the-art methodologies. Codes will be released
at https://github.com/I2-Multimedia-Lab/UGRAN.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:34:37 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Yuan",
"Yao",
""
],
[
"Gao",
"Pan",
""
],
[
"Dai",
"Qun",
""
],
[
"Qin",
"Jie",
""
],
[
"Xiang",
"Wei",
""
]
] |
2504.09667 | Getuar Rexhepi | Getuar Rexhepi, Hyeon Seok Rou, Giuseppe Thadeu Freitas de Abreu | Quantum Manifold Optimization: A Design Framework for Future
Communications Systems | null | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Inspired by recent developments in various areas of science relevant to
quantum computing, we introduce quantum manifold optimization (QMO) as a
promising framework for solving constrained optimization problems in
next-generation wireless communication systems. We begin by showing how
classical wireless design problems - such as pilot design in cell-free
(CF)-massive MIMO (mMIMO), beamformer optimization in gigantic multiple input
multiple output (MIMO), and reconfigurable intelligent surface (RIS) phase
tuning - naturally reside on structured manifolds like the Stiefel,
Grassmannian, and oblique manifolds, with the latter novelly formulated in this
work. Then, we demonstrate how these problems can be reformulated as
trace-based quantum expectation values over variationally-encoded quantum
states. While theoretical in scope, the work lays a foundation for a new class
of quantum optimization algorithms with broad application to the design of
future beyond-sixth-generation (B6G) systems.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:36:21 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Rexhepi",
"Getuar",
""
],
[
"Rou",
"Hyeon Seok",
""
],
[
"de Abreu",
"Giuseppe Thadeu Freitas",
""
]
] |
2504.09668 | Shubhalaxmi Rath | Shubhalaxmi Rath and Nicol\'as A. Neill | Deciphering the viscous properties and the Bjorken expansion of the QGP
medium at finite angular velocity | 31 pages, 9 figures | null | null | null | hep-ph hep-th nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We have studied the viscous properties as well as the Bjorken expansion of a
rotating QGP medium. In the noncentral events of heavy-ion collisions, the
produced medium can carry a finite angular momentum with a finite range of
angular velocity. This rotation can significantly affect various properties,
including viscous properties and the expansion of the QGP medium. Using a novel
relaxation time approximation for the collision integral in the relativistic
Boltzmann transport equation at finite angular velocity, we have calculated the
shear and bulk viscosities and compared them with their counterparts in the
standard relaxation time approximation within the kinetic theory approach. Our
results show that the angular velocity increases both shear and bulk
viscosities, suggesting an enhanced momentum transfer within the medium and
greater fluctuations in local pressure. This rotational effect on viscosities
is more evident at lower temperatures than at higher temperatures. Our analysis
also shows that, compared to the standard relaxation time approximation, the
shear viscosity is lower while the bulk viscosity is higher in the novel
relaxation time approximation for all temperatures. Additionally, some
observables related to the flow characteristic, fluid behavior and conformal
symmetry of the medium are markedly impacted due to rotation. We have also
studied the hydrodynamic evolution of matter within the Bjorken boost-invariant
scenario and have found that the energy density evolves faster in the presence
of finite rotation than in the nonrotating case. Consequently, rapid rotation
accelerates the cooling process of the QGP medium.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:40:03 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Rath",
"Shubhalaxmi",
""
],
[
"Neill",
"Nicolás A.",
""
]
] |
2504.09669 | Ruilong Zhang | Xiaohui Bei, Yuda Feng, Yang Hu, Shi Li, Ruilong Zhang | Nash Social Welfare with Submodular Valuations: Approximation Algorithms
and Integrality Gaps | null | null | null | null | cs.GT cs.DS | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We study the problem of allocating items to agents such that the (un)weighted
Nash social welfare (NSW) is maximized under submodular valuations. The
best-known results for unweighted and weighted problems are the $(4+\epsilon)$
approximation given by Garg, Husic, Li, Vega, and
Vondrak~\cite{stoc/GargHLVV23} and the $(233+\epsilon)$ approximation given by
Feng, Hu, Li, and Zhang~\cite{stoc/FHLZ25}, respectively.
For the weighted NSW problem, we present a $(5.18+\epsilon)$-approximation
algorithm, significantly improving the previous approximation ratio and
simplifying the analysis. Our algorithm is based on the same configuration LP
in~\cite{stoc/FHLZ25}, but with a modified rounding algorithm. For the
unweighted NSW problem, we show that the local search-based algorithm
in~\cite{stoc/GargHLVV23} is an approximation of $(3.914+\epsilon)$ by more
careful analysis.
On the negative side, we prove that the configuration LP for weighted NSW
with submodular valuations has an integrality gap at least $2^{\ln 2}-\epsilon
\approx 1.617 - \epsilon$, which is slightly larger than the current best-known
$e/(e-1)-\epsilon \approx 1.582-\epsilon$ hardness of
approximation~\cite{talg/GargKK23}. For the additive valuation case, we show an
integrality gap of $(e^{1/e}-\epsilon)$, which proves that the ratio of
$(e^{1/e}+\epsilon)$~\cite{icalp/FengLi24} is tight for algorithms based on the
configuration LP. For unweighted NSW with additive valuations, we show a gap of
$(2^{1/4}-\epsilon) \approx 1.189-\epsilon$, slightly larger than the current
best-known $\sqrt{8/7} \approx 1.069$-hardness for the
problem~\cite{mor/Garg0M24}.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:41:31 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Bei",
"Xiaohui",
""
],
[
"Feng",
"Yuda",
""
],
[
"Hu",
"Yang",
""
],
[
"Li",
"Shi",
""
],
[
"Zhang",
"Ruilong",
""
]
] |
2504.09670 | Yifei Guan | Yifei Guan and Pedram Hassanzadeh | Semi-analytical eddy-viscosity and backscattering closures for 2D
geophysical turbulence | 11 pages, 1 figure, 1 table | null | null | null | physics.flu-dyn physics.geo-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Physics-based closures such as eddy-viscosity and backscattering models are
widely used for large-eddy simulation (LES) of geophysical turbulence for
applications including weather and climate prediction. However, these closures
have parameters that are often chosen empirically. Here, for the first time, we
semi-analytically derive the parameters of the Leith and Smagorinsky
eddy-viscosity closures and the Jansen-Held backscattering closure for 2D
geophysical turbulence. The semi-analytical derivation provides these
parameters up to a constant that can be estimated from the turbulent kinetic
energy spectrum of a few snapshots of direct numerical simulation (DNS) or
other high-fidelity (eddy resolving) simulations, or even obtained from earlier
analytical work based on renormalization group. The semi-analytically estimated
closure parameters agree with those obtained from online (a-posteriori)
learning in several setups of 2D geophysical turbulence in our earlier work.
LES with closures that use these parameters can correctly reproduce the key
statistics of DNS, including those of the extreme events and interscale energy
and enstrophy transfers, and outperform the baselines (dynamic Leith and
Smagorinsky and the latter with standard parameter).
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:42:09 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Guan",
"Yifei",
""
],
[
"Hassanzadeh",
"Pedram",
""
]
] |
2504.09671 | Astitva Srivastava | Pranav Manu, Astitva Srivastava, Amit Raj, Varun Jampani, Avinash
Sharma, P.J. Narayanan | LightHeadEd: Relightable & Editable Head Avatars from a Smartphone | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Creating photorealistic, animatable, and relightable 3D head avatars
traditionally requires expensive Lightstage with multiple calibrated cameras,
making it inaccessible for widespread adoption. To bridge this gap, we present
a novel, cost-effective approach for creating high-quality relightable head
avatars using only a smartphone equipped with polaroid filters. Our approach
involves simultaneously capturing cross-polarized and parallel-polarized video
streams in a dark room with a single point-light source, separating the skin's
diffuse and specular components during dynamic facial performances. We
introduce a hybrid representation that embeds 2D Gaussians in the UV space of a
parametric head model, facilitating efficient real-time rendering while
preserving high-fidelity geometric details. Our learning-based neural
analysis-by-synthesis pipeline decouples pose and expression-dependent
geometrical offsets from appearance, decomposing the surface into albedo,
normal, and specular UV texture maps, along with the environment maps. We
collect a unique dataset of various subjects performing diverse facial
expressions and head movements.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:51:56 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Manu",
"Pranav",
""
],
[
"Srivastava",
"Astitva",
""
],
[
"Raj",
"Amit",
""
],
[
"Jampani",
"Varun",
""
],
[
"Sharma",
"Avinash",
""
],
[
"Narayanan",
"P. J.",
""
]
] |
2504.09672 | Christopher Hull | Chris Hull | Duality Symmetries | Invited contribution to the book Half a Century of Supergravity,
edited by Anna Ceresole and Gianguido Dall'Agata | null | null | Imperial-TP-2025-CH-2 | hep-th | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Duality symmetries in supergravity theories, together with their implications
for string theory, are reviewed.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:54:03 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Hull",
"Chris",
""
]
] |
2504.09673 | Palak Chawla | Palak Chawla | Earthquake Simulation | 4 pages | null | null | null | cs.CE | http://creativecommons.org/licenses/by/4.0/ | This paper presents a seismic activity simulator that models the effects of
fault lines on surface pressure. This project uses C programming to create a
fully interactive learning resource intended to educate users on the mechanics
of earthquakes. The motivation behind this project is to make studying seismic
activity more accessible, engaging and cost effective.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:54:25 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Chawla",
"Palak",
""
]
] |
2504.09674 | Marziyeh Soltani | Marziyeh Soltani, Mahtab Mirmohseni, and Rahim Tafazolli | On Stochastic Performance Analysis of Secure Integrated Sensing and
Communication Networks | null | null | null | null | cs.IT eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper analyzes the stochastic security performance of a multiple-input
multiple-output (MIMO) integrated sensing and communication (ISAC) system in a
downlink scenario. A base station (BS) transmits a multi-functional signal to
simultaneously communicate with a user, sense a target angular location, and
counteract eavesdropping threats. The system includes a passive single-antenna
communication eavesdropper and a multi-antenna sensing eavesdropper attempting
to infer the target location. The BS-user and BS-eavesdroppers channels follow
Rayleigh fading, while the target azimuth angle is uniformly distributed. To
evaluate the performance, we derive exact expressions for the secrecy ergodic
rate and the ergodic Cramer-Rao lower bound (CRB) for target localization at
both the BS and the sensing eavesdropper. This involves computing the
probability density functions (PDFs) of the signal-to-noise ratio (SNR) and
CRB, leveraging the central limit theorem for tractability. Numerical results
validate our findings.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:54:53 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Soltani",
"Marziyeh",
""
],
[
"Mirmohseni",
"Mahtab",
""
],
[
"Tafazolli",
"Rahim",
""
]
] |
2504.09675 | Ivan Beldiev | Ivan Beldiev | Projective Hypersurfaces of High Degree Admitting an Induced Additive
Action | null | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study induced additive actions on projective hypersurfaces, i.e. effective
regular actions of the algebraic group $\mathbb G_a^m$ with an open orbit that
can be extended to a regular action on the ambient projective space. It is
known that the degree of a hypersurface $X\subseteq\mathbb{P}^n$ admitting an
induced additive action cannot be greater than $n$ and there is a unique such
hypersurface of degree $n$. We give a complete classification of hypersurfaces
$X\subseteq \mathbb{P}^n$ admitting an induced additive action of degrees from
$n-1$ to $n-3$.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:59:55 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Beldiev",
"Ivan",
""
]
] |
2504.09676 | Jaya Maithil | Jaya Maithil, Daniel A. Schwartz, Aneta Siemiginowska, Preeti Kharb,
Diana M. Worrall, John F. C. Wardle, Giulia Migliori, Chi C. Cheung, Bradford
Snios, Doug B. Gobeille, Herman L. Marshall, and Mark Birkinshaw | Investigating the emission mechanism in the spatially-resolved jet of
two z $\approx$ 3 radio-loud quasars | Accepted in ApJ. 31 pages, 20 figures. All figures are included in
the source zip file (Download --> Other formats --> Source) | null | null | null | astro-ph.HE astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | This study focuses on high-redshift, z > 3, quasars where resolved X-ray jets
remain underexplored in comparison to nearby sources. Building upon previous
work, we identify and confirm extended kpc-scale jets emission in two quasars
(J1405+0415, z = 3.215 ; J1610+1811, z = 3.122) through meticulous analysis of
Chandra X-ray data. To deepen our understanding, high-resolution radio follow-
up observations were conducted to constrain relativistic parameters, providing
valuable insights into the enthalpy flux of these high-redshift AGN jets. The
investigation specifically aims to test the X-ray emission mechanism in these
quasars by exploring the inverse Compton scattering of cosmic microwave
background (IC/CMB) photons by synchrotron-emitting electrons. Our novel method
uses a prior to make a Bayesian estimate of the unknown angle of the jet to our
line of sight, thus breaking the usual degeneracy of the bulk Lorentz factor
and the Doppler beaming factor.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:02:34 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Maithil",
"Jaya",
""
],
[
"Schwartz",
"Daniel A.",
""
],
[
"Siemiginowska",
"Aneta",
""
],
[
"Kharb",
"Preeti",
""
],
[
"Worrall",
"Diana M.",
""
],
[
"Wardle",
"John F. C.",
""
],
[
"Migliori",
"Giulia",
""
],
[
"Cheung",
"Chi C.",
""
],
[
"Snios",
"Bradford",
""
],
[
"Gobeille",
"Doug B.",
""
],
[
"Marshall",
"Herman L.",
""
],
[
"Birkinshaw",
"Mark",
""
]
] |
2504.09677 | Ling Zhu | Ling Zhu, Ana L. Chies-Santos, Micheli Trindade Moura, Hanjing Shi | Distinguishing the formation paths of massive compact ETGs through their
internal dynamical structures | Accepted for publication in A&A | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Massive early-type galaxies (ETGs) are thought to form in two phases: an
initial phase of rapid star formation and a later phase of mergers. A small
fraction of these galaxies, called red nuggets, formed during the first phase
may have survived frozen until today, experiencing no massive mergers since
z~2. Nearby massive compact ETGs are considered candidates for such relic
galaxies. We study the internal dynamical structures of 15 compact ETGs with
existing integral field unit (IFU) observations and 79 compact ETGs from the
TNG50 simulation. We dynamically decompose each galaxy into a disk, bulge, and
hot inner stellar halo, for both observations and simulations. In TNG50, the
luminosity fraction of the hot inner stellar halo (or the size of the spheroid,
which includes the bulge and halo) strongly correlates with the galaxy merger
history. The true merger-free galaxies show an extremely low fraction of a hot
inner stellar halo (or an extremely compact spheroid). Although such
compactness could result from the tidal stripping of satellites, tidal forces
would also destroy the dynamically cold disk (if one exists) when the halo is
removed. Thus, a galaxy is guaranteed to be merger-free if it has a very low
fraction of the hot inner stellar halo and retains a dynamically cold disk.
Comparing observed galaxies with TNG50, we identify 7 of the 15 compact ETGs,
PGC 11179, UGC 3816, NGC 2767, NGC 1277, PGC 32873, PGC 12562, and PGC 70520,
as true merger-free galaxies. These galaxies have compact, massive bulges,
likely formed through secular heating, as supported by their TNG50 analogues.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:02:58 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhu",
"Ling",
""
],
[
"Chies-Santos",
"Ana L.",
""
],
[
"Moura",
"Micheli Trindade",
""
],
[
"Shi",
"Hanjing",
""
]
] |
2504.09678 | Jhony Fernando Caranguay-Mainguez | Jhony F. Caranguay-Mainguez, Pedro Rizzo, Jos\'e A. V\'elez-Marulanda | Universal deformation rings of a special class of modules over
generalized Brauer tree algebras | 34 pages, 6 figures | null | null | null | math.RT math.RA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $\Bbbk$ be an algebraically closed field and $\Lambda$ a generalized
Brauer tree algebra over $\Bbbk$. We compute the universal deformation rings of
the periodic string modules over $\Lambda$. Moreover, for a specific class of
generalized Brauer tree algebras $\Lambda(n,\overline{m})$, we classify the
universal deformation rings of the modules lying in $\Omega$-stable components
$\mathfrak{C}$ of the stable Auslander-Reiten quiver provided that
$\mathfrak{C}$ contains at least one simple module. Our approach uses several
tools and techniques from the representation theory of Brauer graph algebras.
Notably, we leverage Duffield's work on the Auslander-Reiten theory of these
algebras and Opper-Zvonareva's results on derived equivalences between Brauer
graph algebras.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:04:12 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Caranguay-Mainguez",
"Jhony F.",
""
],
[
"Rizzo",
"Pedro",
""
],
[
"Vélez-Marulanda",
"José A.",
""
]
] |
2504.09679 | Qing Diao | Qing Diao, Hongxin Wang, He Wang, Jun Nian, Peng Xu, Minghui Du | Impact of Massive Black Hole Binaries Source Confusion on Uncertainties
of Parameters Estimation in Space-based Gravitational Wave Detection for the
TaiJi Mission | null | null | null | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We systematically investigates the impact of source confusion on parameter
estimation for massive black hole binaries in the context of the Taiji
space-based gravitational wave mission. Three representative MBHB population
models, such as PopIII, Q3d, and Q3nod, are considered. By performing
high-precision numerical simulations and employing both the Fisher information
matrix and Markov chain Monte Carlo techniques, we quantitatively assess the
uncertainties in parameter estimation. Source confusion primarily arises from
overlapping time-frequency tracks of multiple signals. We find that when the
relative difference in detector-frame chirp masses between two signals is less
than 0.2 percent, the uncertainty in parameter estimation increases
significantly. However, this degradation can be substantially mitigated by
incorporating higher-order modes into the waveform model. Furthermore, using
full Bayesian inference, we demonstrate the clear advantage of higher-order
modes in reducing systematic biases in parameter recovery. These results
provide important guidance for future data analysis strategies in space-based
gravitational wave observations.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:08:16 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Diao",
"Qing",
""
],
[
"Wang",
"Hongxin",
""
],
[
"Wang",
"He",
""
],
[
"Nian",
"Jun",
""
],
[
"Xu",
"Peng",
""
],
[
"Du",
"Minghui",
""
]
] |
2504.09680 | Pascal Van Hentenryck | Sikai Cheng, Amira Hijazi, Jeren Konak, Alan Erera, Pascal Van
Hentenryck | SPOT: Spatio-Temporal Pattern Mining and Optimization for Load
Consolidation in Freight Transportation Networks | null | null | null | null | cs.LG cs.AI math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Freight consolidation has significant potential to reduce transportation
costs and mitigate congestion and pollution. An effective load consolidation
plan relies on carefully chosen consolidation points to ensure alignment with
existing transportation management processes, such as driver scheduling,
personnel planning, and terminal operations. This complexity represents a
significant challenge when searching for optimal consolidation strategies.
Traditional optimization-based methods provide exact solutions, but their
computational complexity makes them impractical for large-scale instances and
they fail to leverage historical data. Machine learning-based approaches
address these issues but often ignore operational constraints, leading to
infeasible consolidation plans.
This work proposes SPOT, an end-to-end approach that integrates the benefits
of machine learning (ML) and optimization for load consolidation. The ML
component plays a key role in the planning phase by identifying the
consolidation points through spatio-temporal clustering and constrained
frequent itemset mining, while the optimization selects the most cost-effective
feasible consolidation routes for a given operational day. Extensive
experiments conducted on industrial load data demonstrate that SPOT
significantly reduces travel distance and transportation costs (by about 50% on
large terminals) compared to the existing industry-standard load planning
strategy and a neighborhood-based heuristic. Moreover, the ML component
provides valuable tactical-level insights by identifying frequently recurring
consolidation opportunities that guide proactive planning. In addition, SPOT is
computationally efficient and can be easily scaled to accommodate large
transportation networks.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:14:38 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Cheng",
"Sikai",
""
],
[
"Hijazi",
"Amira",
""
],
[
"Konak",
"Jeren",
""
],
[
"Erera",
"Alan",
""
],
[
"Van Hentenryck",
"Pascal",
""
]
] |
2504.09681 | Bikash R. Dinda | Bikash R. Dinda, Roy Maartens, Shun Saito, and Chris Clarkson | Improved null tests of $\Lambda$CDM and FLRW in light of DESI DR2 | 20 pages, 11 figures | null | null | null | astro-ph.CO gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The DESI DR2 BAO data exclude the flat $\Lambda$CDM model at more than
2.5$\sigma$, depending on different data combinations when analyzed through the
$w_0w_a$CDM parametrization. This simple parametrization may introduce bias in
the results. We use null tests that probe for deviations from flat $\Lambda$CDM
at late times, independent of any specific dark energy parametrization. We
provide several diagnostics for null tests and discuss their advantages and
disadvantages. In particular, we derive diagnostics that improve on previous
ones, such as the popular $O_{\rm m}$ diagnostic. The diagnostics are derived
from both background and perturbed quantities. Using the combination of DESI
DR2 BAO and supernova data, with or without CMB data, we find that deviations
from flat $\Lambda$CDM are at $\sim1\sigma$ confidence level in most of the
redshift range (more than 1$\sigma$ for a few small redshift intervals in a few
cases). These deviations are minor for other non-DESI SDSS-IV BAO data combined
with Pantheon+, with or without CMB data. Since spatial curvature can
potentially modify the results, we also test for curvature in the general
$\Lambda$CDM model and the general FLRW model. While there is slight evidence
for nonzero cosmic curvature at lower redshifts in a general $\Lambda$CDM
model, there is no statistically significant evidence in a general FLRW model.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:21:38 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Dinda",
"Bikash R.",
""
],
[
"Maartens",
"Roy",
""
],
[
"Saito",
"Shun",
""
],
[
"Clarkson",
"Chris",
""
]
] |
2504.09682 | Opal Issan | Opal Issan, Oleksandr Koshkarov, Federico D. Halpern, Gian Luca
Delzanno, Boris Kramer | Conservative data-driven model order reduction of a fluid-kinetic
spectral solver | null | null | null | null | physics.plasm-ph physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Kinetic simulations are computationally intensive due to six-dimensional
phase space discretization. Many kinetic spectral solvers use the
asymmetrically weighted Hermite expansion due to its conservation and
fluid-kinetic coupling properties, i.e., the lower-order Hermite moments
capture and describe the macroscopic fluid dynamics and higher-order Hermite
moments describe the microscopic kinetic dynamics. We leverage this structure
by developing a parametric data-driven reduced-order model based on the proper
orthogonal decomposition, which projects the higher-order kinetic moments while
retaining the fluid moments intact. This approach can also be understood as
learning a nonlocal closure via a reduced modal decomposition. We demonstrate
analytically and numerically that the method ensures local and global mass,
momentum, and energy conservation. The numerical results show that the proposed
method effectively replicates the high-dimensional spectral simulations at a
fraction of the computational cost and memory, as validated on the weak Landau
damping and two-stream instability benchmark problems.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:29:02 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Issan",
"Opal",
""
],
[
"Koshkarov",
"Oleksandr",
""
],
[
"Halpern",
"Federico D.",
""
],
[
"Delzanno",
"Gian Luca",
""
],
[
"Kramer",
"Boris",
""
]
] |
2504.09684 | Fabio Centofanti | Fabio Centofanti, Antonio Lepore, Biagio Palumbo | An Adaptive Multivariate Functional Control Chart | null | null | null | null | stat.ME stat.AP | http://creativecommons.org/licenses/by/4.0/ | New data acquisition technologies allow one to gather huge amounts of data
that are best represented as functional data. In this setting, profile
monitoring assesses the stability over time of both univariate and multivariate
functional quality characteristics. The detection power of profile monitoring
methods could heavily depend on parameter selection criteria, which usually do
not take into account any information from the out-of-control (OC) state. This
work proposes a new framework, referred to as adaptive multivariate functional
control chart (AMFCC), capable of adapting the monitoring of a multivariate
functional quality characteristic to the unknown OC distribution, by combining
$p$-values of the partial tests corresponding to Hotelling $T^2$-type
statistics calculated at different parameter combinations. Through an extensive
Monte Carlo simulation study, the performance of AMFCC is compared with methods
that have already appeared in the literature. Finally, a case study is
presented in which the proposed framework is used to monitor a resistance spot
welding process in the automotive industry. AMFCC is implemented in the R
package funcharts, available on CRAN.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:35:49 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Centofanti",
"Fabio",
""
],
[
"Lepore",
"Antonio",
""
],
[
"Palumbo",
"Biagio",
""
]
] |
2504.09685 | Wassim Hamidouche | Christophe El Zeinaty, Wassim Hamidouche, Glenn Herrou, Daniel Menard
and Merouane Debbah | Can LLMs Revolutionize the Design of Explainable and Efficient TinyML
Models? | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper introduces a novel framework for designing efficient neural
network architectures specifically tailored to tiny machine learning (TinyML)
platforms. By leveraging large language models (LLMs) for neural architecture
search (NAS), a vision transformer (ViT)-based knowledge distillation (KD)
strategy, and an explainability module, the approach strikes an optimal balance
between accuracy, computational efficiency, and memory usage. The LLM-guided
search explores a hierarchical search space, refining candidate architectures
through Pareto optimization based on accuracy, multiply-accumulate operations
(MACs), and memory metrics. The best-performing architectures are further
fine-tuned using logits-based KD with a pre-trained ViT-B/16 model, which
enhances generalization without increasing model size. Evaluated on the
CIFAR-100 dataset and deployed on an STM32H7 microcontroller (MCU), the three
proposed models, LMaNet-Elite, LMaNet-Core, and QwNet-Core, achieve accuracy
scores of 74.50%, 74.20% and 73.00%, respectively. All three models surpass
current state-of-the-art (SOTA) models, such as MCUNet-in3/in4 (69.62% /
72.86%) and XiNet (72.27%), while maintaining a low computational cost of less
than 100 million MACs and adhering to the stringent 320 KB static random-access
memory (SRAM) constraint. These results demonstrate the efficiency and
performance of the proposed framework for TinyML platforms, underscoring the
potential of combining LLM-driven search, Pareto optimization, KD, and
explainability to develop accurate, efficient, and interpretable models. This
approach opens new possibilities in NAS, enabling the design of efficient
architectures specifically suited for TinyML.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:36:03 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zeinaty",
"Christophe El",
""
],
[
"Hamidouche",
"Wassim",
""
],
[
"Herrou",
"Glenn",
""
],
[
"Menard",
"Daniel",
""
],
[
"Debbah",
"Merouane",
""
]
] |
2504.09686 | Horacio Espinosa | Yue Zhang, Robert J. Appleton, Kui Lin, Megan J. McCarthy, Jeffrey T.
Paci, Subramanian K.R.S. Sankaranarayanan, Alejandro Strachan, Horacio D.
Espinosa | Harnessing Machine Learning for Quantum-Accurate Predictions of
Non-Equilibrium Behavior in 2D Materials | 23 pages, 5 figures, Supplementary Information (5 figures and two
tables) | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurately predicting the non-equilibrium mechanical properties of
two-dimensional (2D) materials is essential for understanding their
deformation, thermo-mechanical properties, and failure mechanisms. In this
study, we parameterize and evaluate two machine learning (ML) interatomic
potentials, SNAP and Allegro, for modeling the non-equilibrium behavior of
monolayer MoSe2. Using a density functional theory (DFT) derived dataset, we
systematically compare their accuracy and transferability against the
physics-based Tersoff force field. Our results show that SNAP and Allegro
significantly outperform Tersoff, achieving near-DFT accuracy while maintaining
computational efficiency. Allegro surpasses SNAP in both accuracy and
efficiency due to its advanced neural network architecture. Both ML potentials
demonstrate strong transferability, accurately predicting out-of-sample
properties such as surface stability, inversion domain formation, and fracture
toughness. Unlike Tersoff, SNAP and Allegro reliably model
temperature-dependent edge stabilities and phase transformation pathways,
aligning closely with DFT benchmarks. Notably, their fracture toughness
predictions closely match experimental measurements, reinforcing their
suitability for large-scale simulations of mechanical failure in 2D materials.
This study establishes ML-based force fields as a powerful alternative to
traditional potentials for modeling non-equilibrium mechanical properties in 2D
materials.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:37:01 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Zhang",
"Yue",
""
],
[
"Appleton",
"Robert J.",
""
],
[
"Lin",
"Kui",
""
],
[
"McCarthy",
"Megan J.",
""
],
[
"Paci",
"Jeffrey T.",
""
],
[
"Sankaranarayanan",
"Subramanian K. R. S.",
""
],
[
"Strachan",
"Alejandro",
""
],
[
"Espinosa",
"Horacio D.",
""
]
] |
2504.09687 | Salman Faroz | Salman Faroz | Domain-Adaptive Continued Pre-Training of Small Language Models | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Continued pre-training of small language models offers a promising path for
domain adaptation with limited computational resources. I've investigated this
approach within educational domains, evaluating it as a resource-efficient
alternative to training models from scratch. Using a 125M parameter model, I
demonstrate significant performance improvements through incremental training
on 400 million tokens, followed by further training to reach 1 billion tokens.
My approach includes comprehensive data preprocessing, memory-optimized
training configurations, and benchmark-based evaluation. Results show notable
gains in knowledge-intensive tasks (MMLU +8.1%) and contextual understanding
(HellaSwag +7.6%), while revealing educational domain specialization
trade-offs. I analyze token efficiency, catastrophic forgetting mitigation
strategies, and scaling patterns. My findings suggest that thoughtful
preprocessing and training methodologies enable meaningful improvements in
language model capabilities even with constrained computational resources,
opening pathways for domain-specific adaptation of smaller language models.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 18:40:32 GMT"
}
] | 2025-04-15T00:00:00 | [
[
"Faroz",
"Salman",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.