id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 6
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-03-29 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2501.01521 | Christophe Royon | C. Royon, J. Penttala | Forward dijet and vector meson production at the LHC | Proceedings of the ICHEP 2024 conference, Prague, Czech Republic,
July 17-24 2024 | null | null | null | hep-ph hep-ex nucl-ex nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We first describe the predictions concerning jet production in the very
forward region, especially in the FOCAL kinematic region of the ALICE
collaboration at CERN as a test of saturation. We also compare vector meson
production cross sections at the LHC in Pb Pb and p Pb collisions with the
Balitsky Fadin Kuraev Lipatov and the Balitsky Kovchegov evolution equations
and data favor the observation of saturation at the LHC.
| [
{
"version": "v1",
"created": "Thu, 2 Jan 2025 20:22:28 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:57:46 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Royon",
"C.",
""
],
[
"Penttala",
"J.",
""
]
] |
2501.01855 | Huaxiang Zhang | Huaxiang Zhang, Kai Liu, Zhongxue Gan, and Guo-Niu Zhu | UAV-DETR: Efficient End-to-End Object Detection for Unmanned Aerial
Vehicle Imagery | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unmanned aerial vehicle object detection (UAV-OD) has been widely used in
various scenarios. However, most existing UAV-OD algorithms rely on manually
designed components, which require extensive tuning. End-to-end models that do
not depend on such manually designed components are mainly designed for natural
images, which are less effective for UAV imagery. To address such challenges,
this paper proposes an efficient detection transformer (DETR) framework
tailored for UAV imagery, i.e., UAV-DETR. The framework includes a multi-scale
feature fusion with frequency enhancement module, which captures both spatial
and frequency information at different scales. In addition, a frequency-focused
down-sampling module is presented to retain critical spatial details during
down-sampling. A semantic alignment and calibration module is developed to
align and fuse features from different fusion paths. Experimental results
demonstrate the effectiveness and generalization of our approach across various
UAV imagery datasets. On the VisDrone dataset, our method improves AP by 3.1\%
and $\text{AP}_{50}$ by 4.2\% over the baseline. Similar enhancements are
observed on the UAVVaste dataset. The project page:
https://github.com/ValiantDiligent/UAV-DETR
| [
{
"version": "v1",
"created": "Fri, 3 Jan 2025 15:11:14 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:17:42 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Huaxiang",
""
],
[
"Liu",
"Kai",
""
],
[
"Gan",
"Zhongxue",
""
],
[
"Zhu",
"Guo-Niu",
""
]
] |
2501.02471 | Yishen Liu | Yishen Liu and Shengda Luo and Zishao Zhong and Tongtong Wu and
Jianguo Zhang and Peiyao Ou and Yong Liang and Liang Liu and Hudan Pan | Hengqin-RA-v1: Advanced Large Language Model for Diagnosis and Treatment
of Rheumatoid Arthritis with Dataset based Traditional Chinese Medicine | 8 pages, 5 figures, AAAI-2025 Workshop | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) primarily trained on English texts, often face
biases and inaccuracies in Chinese contexts. Their limitations are pronounced
in fields like Traditional Chinese Medicine (TCM), where cultural and clinical
subtleties are vital, further hindered by a lack of domain-specific data, such
as rheumatoid arthritis (RA). To address these issues, this paper introduces
Hengqin-RA-v1, the first large language model specifically tailored for TCM
with a focus on diagnosing and treating RA. We also present HQ-GCM-RA-C1, a
comprehensive RA-specific dataset curated from ancient Chinese medical
literature, classical texts, and modern clinical studies. This dataset empowers
Hengqin-RA-v1 to deliver accurate and culturally informed responses,
effectively bridging the gaps left by general-purpose models. Extensive
experiments demonstrate that Hengqin-RA-v1 outperforms state-of-the-art models,
even surpassing the diagnostic accuracy of TCM practitioners in certain cases.
| [
{
"version": "v1",
"created": "Sun, 5 Jan 2025 07:46:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:39:45 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Liu",
"Yishen",
""
],
[
"Luo",
"Shengda",
""
],
[
"Zhong",
"Zishao",
""
],
[
"Wu",
"Tongtong",
""
],
[
"Zhang",
"Jianguo",
""
],
[
"Ou",
"Peiyao",
""
],
[
"Liang",
"Yong",
""
],
[
"Liu",
"Liang",
""
],
[
"Pan",
"Hudan",
""
]
] |
2501.03048 | Qingyuan Zhao | Qingyuan Zhao | On statistical and causal models associated with acyclic directed mixed
graphs | 36 pages. Revised the definition of nested Markov property and the
proof of Theorem 3 | null | null | null | math.ST stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Causal models in statistics are often described using acyclic directed mixed
graphs (ADMGs), which contain directed and bidirected edges and no directed
cycles. This article surveys various interpretations of ADMGs, discusses their
relations in different sub-classes of ADMGs, and argues that one of them -- the
noise expansion (NE) model -- should be used as the default interpretation. Our
endorsement of the NE model is based on two observations. First, in a subclass
of ADMGs called unconfounded graphs (which retain most of the good properties
of directed acyclic graphs and bidirected graphs), the NE model is equivalent
to many other interpretations including the global Markov and nested Markov
models. Second, the NE model for an arbitrary ADMG is exactly the union of that
for all unconfounded expansions of that graph. This property is referred to as
completeness, as it shows that the model does not commit to any specific latent
variable explanation. In proving that the NE model is nested Markov, we also
develop an ADMG-based theory for causality. Finally, we compare the NE model
with the closely related but different interpretation of ADMGs as directed
acyclic graphs (DAGs) with latent variables that is commonly used in the
literature. We argue that the "latent DAG" interpretation is mathematically
unnecessary, makes obscure ontological assumptions, and discourages
practitioners from deliberating over important structural assumptions.
| [
{
"version": "v1",
"created": "Mon, 6 Jan 2025 14:37:10 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:18:58 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhao",
"Qingyuan",
""
]
] |
2501.03514 | Fumihiko Ishiyama | Fumihiko Ishiyama | Refreshing idea on Fourier analysis | 9 pages, 5 figures. Proc. IEEE CSPA 2025 | Proc. 2025 21st IEEE International Colloquium on Signal Processing
& Its Applications (CSPA), pp. 1-4 | 10.1109/CSPA64953.2025.10933386 | null | physics.soc-ph cs.NA math.NA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The "theoretical limit of time-frequency resolution in Fourier analysis" is
thought to originate in certain mathematical and/or physical limitations. This,
however, is not true. The actual origin arises from the numerical (technical)
method deployed to reduce computation time. In addition, there is a gap between
the theoretical equation for Fourier analysis and its numerical implementation.
Knowing the facts brings us practical benefits. In this case, these related to
boundary conditions, and complex integrals. For example, replacing a Fourier
integral with a complex integral brings a hybrid method for the Laplace and
Fourier transforms, and reveals another perspective on time-frequency analysis.
We present such a perspective here with a simple demonstrative analysis.
| [
{
"version": "v1",
"created": "Tue, 7 Jan 2025 04:16:18 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Jan 2025 04:03:24 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ishiyama",
"Fumihiko",
""
]
] |
2501.03550 | Sha Wang | Pan Guo, Yuan Gao, Yongjie Pu, Zhigang Zhao, Zhenhua Cong and Sha Wang | Intelligent Mode-Locked Single-Cavity Dual-Comb Laser Utilizing
Time-Stretch Dispersive Fourier Transform Spectroscopy with Supplemental File | 10 pages, 8 figures | null | null | null | physics.optics | http://creativecommons.org/licenses/by/4.0/ | As dual combs play a significant role in numerous high-precision
measurements, their efficient generation has been widely researched. Although
the single-cavity dual-comb generation can avoid the complex active
stabilization methods, achieving and maintaining stable dual-comb mode locking
within a single cavity remains a critical challenge. To break through this
constraint, a two-part evaluation criterion containing a fitness function and a
CNN-Transformer network is employed to achieve mode locking and classify the
dual-comb mode-locked state. Simulated time-stretch dispersive Fourier
transform (DFT) spectra are used as datasets, which simplifies the optimization
process and does not rely on specific experimental data. A developed
evolutionary algorithm (EA) for paddle-based motorized polarization controllers
(MPCs) is proposed, enabling the intelligent attainment of dual-comb
mode-locked states. A real-time library stores fitness and MPC angles,
facilitating mode-locked state achievement within 2 seconds. Finally, long term
running of dual-comb mode locking is ensured by a random collision algorithm
utilizing an evaluation criterion of weak soliton peaks.
| [
{
"version": "v1",
"created": "Tue, 7 Jan 2025 05:52:35 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:35:53 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Guo",
"Pan",
""
],
[
"Gao",
"Yuan",
""
],
[
"Pu",
"Yongjie",
""
],
[
"Zhao",
"Zhigang",
""
],
[
"Cong",
"Zhenhua",
""
],
[
"Wang",
"Sha",
""
]
] |
2501.03579 | Yonghan Xiao | Yonghan Xiao | Grid homology for singular links in lens space and a resolution cube | 55 pages, 25 figures. Comparing with first two versions, we correct
some subtle mistakes in proof and figures in this new one. Comments are
welcomed | null | null | null | math.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we define grid homologies for singular links in lens spaces
and use them to construct a resolution cube for knot Floer homology of regular
links in lens spaces. The results will first be proved over
$\mathbb{Z}/2\mathbb{Z}$ and then over $\mathbb{Z}$ with the help of sign
assignments. We will also identify the signed grid homology and classical knot
Floer homology over $\mathbb{Z}$ for regular links in lens spaces, illustrating
the fact that our resolution cube is genuinely one for knot Floer homology. The
main advancement in the paper is that we give a complete description of
singular knot theory in lens spaces which was only defined in $S^3$ previously
and we construct a signed combinatorial resolution cube for knot Floer homology
in lens spaces which may be powerful in relating $HFK^\circ$ to other link
homology theories.
| [
{
"version": "v1",
"created": "Tue, 7 Jan 2025 07:07:59 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Jan 2025 13:16:12 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 11:35:13 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Xiao",
"Yonghan",
""
]
] |
2501.04037 | Xiyun Zhang | Xiyun Zhang, Bojun Wang, Hongjie Bi | Fluctuation induced intermittent transitions between distinct rhythms in
balanced excitatory-inhibitory spiking networks | null | Chaos, Solitons & Fractals 196, 116321 (2025) | 10.1016/j.chaos.2025.116321 | null | physics.soc-ph q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Intermittent transitions, associated with critical dynamics and characterized
by power-law distributions, are commonly observed during sleep. These critical
behaviors are evident at the microscopic level through neuronal avalanches and
at the macroscopic level through transitions between sleep stages. To clarify
these empirical observations, models grounded in statistical physics have been
proposed. At the mesoscopic level of cortical activity, critical behavior is
indicated by the intermittent transitions between various cortical rhythms. For
instance, empirical investigations utilizing EEG data from rats have identified
intermittent transitions between $\delta$ and $\theta$ rhythms, with the
duration of $\theta$ rhythm exhibiting a power-law distribution. However, a
dynamic model to account for this phenomenon is currently absent. In this
study, we introduce a network of sparsely coupled excitatory and inhibitory
populations of quadratic integrate-and-fire (QIF) neurons to demonstrate that
intermittent transitions can emerge from the intrinsic fluctuations of a
finite-sized system, particularly when the system is positioned near a Hopf
bifurcation point, which is a critical point. The resulting power-law
distributions and exponents are consistent with empirical observations.
Additionally, we illustrate how modifications in network connectivity can
affect the power-law exponent by influencing the attractivity and oscillation
frequency of the stable limit cycle. Our findings, interpreted through the
fundamental dynamics of neuronal networks, provide a plausible mechanism for
the generation of intermittent transitions between cortical rhythms, in
alignment with the power-law distributions documented in empirical researches.
| [
{
"version": "v1",
"created": "Fri, 3 Jan 2025 10:34:36 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Xiyun",
""
],
[
"Wang",
"Bojun",
""
],
[
"Bi",
"Hongjie",
""
]
] |
2501.04380 | Jing-Ling Chen | Jiang-Lin Zhou, Zou-Chen Fu, Choo Hiap Oh, and Jing-Ling Chen | Electric-Type Stern-Gerlach Effect | Main Text 6 pages + SM 97 pages.8 Figures. A Revised version | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | The Stern-Gerlach (SG) experiment is a fundamental experiment for revealing
the existence of ``spin''. In such an experiment, beams of silver atoms were
sent through inhomogeneous magnetic fields to observe their deflection. Thus,
the conventional SG experiment can be actually viewed as a magnetic-type spin
effect. In this work, we successfully generalize the SG effect from
magnetic-type to electric-type by solving Dirac's equation with a potential
barrier, thus revealing an extraordinary spin effect. Beams of Dirac's
particles can be regarded as matter waves. Based on Dirac's equation, we obtain
the explicit forms of the incident, reflected, and transmitted waves. The
electric-type SG effect shows that the reflected and transmitted waves can have
notable spatial shifts, which depend on the spin direction and the incident
angle of incident wave. The electric-type SG effect has potential applications
to separating Dirac's particles with different spin directions and to
estimating the spin direction of Dirac's particles. Some discussions related to
the interaction between spin and electric field are also made.
| [
{
"version": "v1",
"created": "Wed, 8 Jan 2025 09:45:35 GMT"
},
{
"version": "v2",
"created": "Tue, 18 Mar 2025 08:22:54 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 02:15:35 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhou",
"Jiang-Lin",
""
],
[
"Fu",
"Zou-Chen",
""
],
[
"Oh",
"Choo Hiap",
""
],
[
"Chen",
"Jing-Ling",
""
]
] |
2501.04465 | Adam Rancon | Adam Ran\c{c}on, Bertrand Delamotte, Lovro \v{S}aravanja and Ivan
Balog | Probability distributions of the order parameter of the $O(N)$ model | v1) 20 pages, 11 figures; v2) minor corrections, published version | Phys. Rev. E 111, 034131 (2025) | 10.1103/PhysRevE.111.034131 | null | cond-mat.stat-mech | http://creativecommons.org/licenses/by/4.0/ | We study the probability distribution function (PDF) of the order parameter
of the three-dimensional $O(N)$ model at criticality using the functional
renormalisation group. For this purpose, we generalize the method introduced in
[Balog et al., Phys. Rev. Lett. {\bf 129}, 210602 (2022)] to the $O(N)$ model.
We study the large $N$ limit, as well as the cases $N=2$ and $N=3$ at the level
of the Local Potential Approximation (LPA), and compare our results to Monte
Carlo simulations. We compute the entire family of universal scaling functions,
obtained in the limit where the system size $L$ and the correlation length of
the infinite system $\xi_\infty$ diverge, with the ratio $\zeta=L/\xi_\infty$
constant. We also generalize our results to the approach of criticality from
the low-temperature phase where another infinite family of universal PDF
exists. We find that the LPA describes very well the functional form of the
family of PDFs, once we correct for a global amplitude of the (logarithm of
the) PDF and of $\zeta$.
| [
{
"version": "v1",
"created": "Wed, 8 Jan 2025 12:41:09 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 08:28:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Rançon",
"Adam",
""
],
[
"Delamotte",
"Bertrand",
""
],
[
"Šaravanja",
"Lovro",
""
],
[
"Balog",
"Ivan",
""
]
] |
2501.04765 | Felix Krause | Felix Krause, Timy Phan, Ming Gui, Stefan Andreas Baumann, Vincent Tao
Hu, Bj\"orn Ommer | TREAD: Token Routing for Efficient Architecture-agnostic Diffusion
Training | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Diffusion models have emerged as the mainstream approach for visual
generation. However, these models typically suffer from sample inefficiency and
high training costs. Consequently, methods for efficient finetuning, inference
and personalization were quickly adopted by the community. However, training
these models in the first place remains very costly. While several recent
approaches - including masking, distillation, and architectural modifications -
have been proposed to improve training efficiency, each of these methods comes
with a tradeoff: they achieve enhanced performance at the expense of increased
computational cost or vice versa. In contrast, this work aims to improve
training efficiency as well as generative performance at the same time through
routes that act as a transport mechanism for randomly selected tokens from
early layers to deeper layers of the model. Our method is not limited to the
common transformer-based model - it can also be applied to state-space models
and achieves this without architectural modifications or additional parameters.
Finally, we show that TREAD reduces computational cost and simultaneously
boosts model performance on the standard ImageNet-256 benchmark in
class-conditional synthesis. Both of these benefits multiply to a convergence
speedup of 14x at 400K training iterations compared to DiT and 37x compared to
the best benchmark performance of DiT at 7M training iterations. Furthermore,
we achieve a competitive FID of 2.09 in a guided and 3.93 in an unguided
setting, which improves upon the DiT, without architectural changes.
| [
{
"version": "v1",
"created": "Wed, 8 Jan 2025 18:38:25 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:42:53 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Krause",
"Felix",
""
],
[
"Phan",
"Timy",
""
],
[
"Gui",
"Ming",
""
],
[
"Baumann",
"Stefan Andreas",
""
],
[
"Hu",
"Vincent Tao",
""
],
[
"Ommer",
"Björn",
""
]
] |
2501.04925 | Cullen Abelson | C.S. Abelson, Carles Badenes, Laura Chomiuk, Benjamin F. Williams,
Katelyn Breivik, Llu\'is Galbany, Cristina Jimenez Palau | The Progenitor Systems of Classical Novae in M31 | 14 pages, 6 figures. Accepted to The Astrophysical Journal | null | null | null | astro-ph.SR astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | We present the first characterization of the statistical relationship between
a large sample of novae in M31 and their progenitor stellar populations in the
form of a delay time distribution. To this end, we leverage the spatially
resolved stellar age distribution of the M31 disk derived from deep HST
photometry by the Panchromatic Hubble Andromeda Treasury (PHAT) survey and a
large catalog of novae in M31. Our delay time distribution has two
statistically significant detections: one population of nova progenitors, ages
between 2 and 3.2 Gyr, with an unnormalized rate of ($3.7^{+6.8}_{-3.5} \pm
2.1) \cdot 10^{-9}$ events / $M_{\odot}$, and another of ages between 7.9 Gyr
and the age of the Universe with ($4.8^{+1.0}_{-0.9} \pm 0.2) \cdot 10^{-9}$
events / $M_{\odot}$ (uncertainties are statistical and systematic,
respectively). Together with the upper limits we derive at other time bins,
these detections are consistent with either a constant production efficiency or
a higher production efficiency of novae at earlier delay times.
| [
{
"version": "v1",
"created": "Thu, 9 Jan 2025 02:30:07 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 20:50:49 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Abelson",
"C. S.",
""
],
[
"Badenes",
"Carles",
""
],
[
"Chomiuk",
"Laura",
""
],
[
"Williams",
"Benjamin F.",
""
],
[
"Breivik",
"Katelyn",
""
],
[
"Galbany",
"Lluís",
""
],
[
"Palau",
"Cristina Jimenez",
""
]
] |
2501.04959 | Wonseong Kim Dr. | Wonseong Kim, Christina Niklaus, Choong Lyol Lee, Siegfried Handschuh | DisSim-FinBERT: Text Simplification for Core Message Extraction in
Complex Financial Texts | 28 pages, 5 figures, 2 tables | null | null | null | econ.EM stat.CO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This study proposes DisSim-FinBERT, a novel framework that integrates
Discourse Simplification (DisSim) with Aspect-Based Sentiment Analysis (ABSA)
to enhance sentiment prediction in complex financial texts. By simplifying
intricate documents such as Federal Open Market Committee (FOMC) minutes,
DisSim improves the precision of aspect identification, resulting in sentiment
predictions that align more closely with economic events. The model preserves
the original informational content and captures the inherent volatility of
financial language, offering a more nuanced and accurate interpretation of
long-form financial communications. This approach provides a practical tool for
policymakers and analysts aiming to extract actionable insights from central
bank narratives and other detailed economic documents.
| [
{
"version": "v1",
"created": "Thu, 9 Jan 2025 04:23:59 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 00:05:56 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kim",
"Wonseong",
""
],
[
"Niklaus",
"Christina",
""
],
[
"Lee",
"Choong Lyol",
""
],
[
"Handschuh",
"Siegfried",
""
]
] |
2501.05001 | Guoyang Rong Chris | Guoyang Rong, Ying Chen, Feicheng Ma, Thorsten Koch | 40 Years of Interdisciplinary Research: Phases, Origins, and Key Turning
Points (1981-2020) | 16 pages, 3 figures | null | null | null | stat.AP cs.DL physics.soc-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This study examines the historical evolution of interdisciplinary research
(IDR) over a 40-year period, focusing on its dynamic trends, phases, and key
turning points. We apply time series analysis to identify critical years for
interdisciplinary citations (CYICs) and categorizes IDR into three distinct
phases based on these trends: Period I (1981-2002), marked by sporadic and
limited interdisciplinary activity; Period II (2003-2016), characterized by the
emergence of large-scale IDR led primarily by Medicine, with significant
breakthroughs in cloning and medical technology; and Period III (2017-present),
where IDR became a widely adopted research paradigm. Our findings indicate that
IDR has been predominantly concentrated within the Natural Sciences, with
Medicine consistently at the forefront, and highlights increasing contributions
from Engineering and Environmental disciplines as a new trend. These insights
enhance the understanding of the evolution of IDR, its driving factors, and the
shifts in the focus of interdisciplinary collaborations.
| [
{
"version": "v1",
"created": "Thu, 9 Jan 2025 06:30:12 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 04:47:45 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Rong",
"Guoyang",
""
],
[
"Chen",
"Ying",
""
],
[
"Ma",
"Feicheng",
""
],
[
"Koch",
"Thorsten",
""
]
] |
2501.05037 | Rujie Wu | Rujie Wu, Xiaojian Ma, Hai Ci, Yue Fan, Yuxuan Wang, Haozhe Zhao, Qing
Li, Yizhou Wang | LongViTU: Instruction Tuning for Long-Form Video Understanding | null | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces LongViTU, a large-scale (~121k QA pairs, ~900h videos),
automatically generated dataset for long-form video understanding. We propose a
systematic approach that organizes videos into a hierarchical tree structure
for QA generation and incorporates self-revision mechanisms to ensure
high-quality QA pairs. Each QA pair in LongViTU features: 1) long-term context
(average certificate length of 4.6 minutes); 2) rich knowledge and condensed
reasoning (commonsense, causality, planning, etc.)). We also offer explicit
timestamp annotations of relevant events for each QA pair. We have conducted
extensive human studies on LongViTU, and the results prove the quality of our
dataset. To better evaluate the challenges posed by LongViTU's emphasis on
long-term context and condensed reasoning, we manually curate a subset of
LongViTU into a benchmark. Evaluations using a state-of-the-art open-source
model (LongVU), a proprietary model (Gemini-1.5-Pro), and human annotators
yield GPT-4 scores of 49.9, 52.3, and 81.0, respectively, underscoring the
substantial difficulty presented by LongViTU questions. Performing supervised
fine-tuning (SFT) of LongVU and LLaVA-Video on LongViTU data results in average
performance gains of 2.5% and 3.7%, respectively, across a suite of long video
understanding benchmarks (EgoSchema, VideoMME-Long, MLVU, LVBench).
| [
{
"version": "v1",
"created": "Thu, 9 Jan 2025 07:51:14 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 09:39:11 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wu",
"Rujie",
""
],
[
"Ma",
"Xiaojian",
""
],
[
"Ci",
"Hai",
""
],
[
"Fan",
"Yue",
""
],
[
"Wang",
"Yuxuan",
""
],
[
"Zhao",
"Haozhe",
""
],
[
"Li",
"Qing",
""
],
[
"Wang",
"Yizhou",
""
]
] |
2501.05510 | Yifei Li | Yifei Li, Junbo Niu, Ziyang Miao, Chunjiang Ge, Yuanhang Zhou, Qihao
He, Xiaoyi Dong, Haodong Duan, Shuangrui Ding, Rui Qian, Pan Zhang, Yuhang
Zang, Yuhang Cao, Conghui He, Jiaqi Wang | OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video
Understanding? | CVPR 2025 | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Temporal Awareness, the ability to reason dynamically based on the timestamp
when a question is raised, is the key distinction between offline and online
video LLMs. Unlike offline models, which rely on complete videos for static,
post hoc analysis, online models process video streams incrementally and
dynamically adapt their responses based on the timestamp at which the question
is posed. Despite its significance, temporal awareness has not been adequately
evaluated in existing benchmarks. To fill this gap, we present OVO-Bench
(Online-VideO-Benchmark), a novel video benchmark that emphasizes the
importance of timestamps for advanced online video understanding capability
benchmarking. OVO-Bench evaluates the ability of video LLMs to reason and
respond to events occurring at specific timestamps under three distinct
scenarios: (1) Backward tracing: trace back to past events to answer the
question. (2) Real-time understanding: understand and respond to events as they
unfold at the current timestamp. (3) Forward active responding: delay the
response until sufficient future information becomes available to answer the
question accurately. OVO-Bench comprises 12 tasks, featuring 644 unique videos
and approximately human-curated 2,800 fine-grained meta-annotations with
precise timestamps. We combine automated generation pipelines with human
curation. With these high-quality samples, we further developed an evaluation
pipeline to systematically query video LLMs along the video timeline.
Evaluations of nine Video-LLMs reveal that, despite advancements on traditional
benchmarks, current models struggle with online video understanding, showing a
significant gap compared to human agents. We hope OVO-Bench will drive progress
in video LLMs and inspire future research in online video reasoning. Our
benchmark and code can be accessed at https://github.com/JoeLeelyf/OVO-Bench.
| [
{
"version": "v1",
"created": "Thu, 9 Jan 2025 19:00:01 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:40:09 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Li",
"Yifei",
""
],
[
"Niu",
"Junbo",
""
],
[
"Miao",
"Ziyang",
""
],
[
"Ge",
"Chunjiang",
""
],
[
"Zhou",
"Yuanhang",
""
],
[
"He",
"Qihao",
""
],
[
"Dong",
"Xiaoyi",
""
],
[
"Duan",
"Haodong",
""
],
[
"Ding",
"Shuangrui",
""
],
[
"Qian",
"Rui",
""
],
[
"Zhang",
"Pan",
""
],
[
"Zang",
"Yuhang",
""
],
[
"Cao",
"Yuhang",
""
],
[
"He",
"Conghui",
""
],
[
"Wang",
"Jiaqi",
""
]
] |
2501.07552 | Tarek Hamdi | Nizar Demni and Tarek Hamdi | Spectral distribution of the free Jacobi process with equal rank
projections | null | null | null | null | math.PR math.AP math.CV math.FA math.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The free Jacobi process is the radial part of the compression of the free
unitary Brownian motion by two free orthogonal projections in a non commutative
probability space. In this paper, we derive spectral properties of the free
Jacobi process associated with projections having the same rank $\alpha \in
(0,1)$. To start with, we determine the characteristic curves of the partial
differential equation satisfied by the moment generating function of its
spectral distribution. Doing so leads for any fixed time $t >0$ to an
expression of this function in a neighborhood of the origin, therefore extends
our previous results valid for $\alpha = 1/2$. Moreover, the obtained
characteristic curves are encoded by an $\alpha$-deformation of the
compositional inverse of the $\chi$-transform of the spectral distribution of
the free unitary Brownian motion. In this respect, we study mapping properties
of this deformation and use the saddle point method to prove that the
compositional inverse of a $\alpha$-deformation of the $\chi$-transform of the
free unitary Brownian motion is analytic in the open unit disc (for large
enough time $t$). The last part of the paper is devoted to a dynamical version
of a recent identity pointed out by T. Kunisky in \cite{Kun}. Actually, this
identity relates the stationary distributions of the free Jacobi processes
corresponding to the sets of parameters $(\alpha, \alpha)$ and $(1/2,\alpha)$
respectively and we explain how it follows from the Nica-Speicher semi-group.
Our dynamical version then relates the partial differential equations of the
Cauchy-Stieltjes transforms of the densities of the finite-time spectral
distributions. It also raises the problem of whether a dynamical analogue of
the Nica-Speicher semi-group exists when the compressing projection has rank
$1/2$.
| [
{
"version": "v1",
"created": "Mon, 13 Jan 2025 18:36:42 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 20:42:42 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Demni",
"Nizar",
""
],
[
"Hamdi",
"Tarek",
""
]
] |
2501.07586 | Arnaud Beauville | Arnaud Beauville | Hyperplane sections of cubic threefolds | Added reference to arXiv:2503.17991 by Beorchia and Mir\`o-Roig | null | null | null | math.AG | http://creativecommons.org/licenses/by/4.0/ | Let X be a smooth cubic hypersurface. We prove that a general cubic surface
is isomorphic to a hyperplane section of X .
| [
{
"version": "v1",
"created": "Fri, 3 Jan 2025 07:58:29 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 08:41:09 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Beauville",
"Arnaud",
""
]
] |
2501.07601 | Yi-Ping Chen | Yi-Ping Chen, Vispi Karkaria, Ying-Kuan Tsai, Faith Rolark, Daniel
Quispe, Robert X. Gao, Jian Cao, Wei Chen | Real-Time Decision-Making for Digital Twin in Additive Manufacturing
with Model Predictive Control using Time-Series Deep Neural Networks | null | null | 10.1016/j.jmsy.2025.03.009 | null | cs.LG cs.AI cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | Digital Twin -- a virtual replica of a physical system enabling real-time
monitoring, model updating, prediction, and decision-making -- combined with
recent advances in machine learning, offers new opportunities for proactive
control strategies in autonomous manufacturing. However, achieving real-time
decision-making with Digital Twins requires efficient optimization driven by
accurate predictions of highly nonlinear manufacturing systems. This paper
presents a simultaneous multi-step Model Predictive Control (MPC) framework for
real-time decision-making, using a multivariate deep neural network, named
Time-Series Dense Encoder (TiDE), as the surrogate model. Unlike conventional
MPC models which only provide one-step ahead prediction, TiDE is capable of
predicting future states within the prediction horizon in one shot
(multi-step), significantly accelerating the MPC. Using Directed Energy
Deposition (DED) additive manufacturing as a case study, we demonstrate the
effectiveness of the proposed MPC in achieving melt pool temperature tracking
to ensure part quality, while reducing porosity defects by regulating laser
power to maintain melt pool depth constraints. In this work, we first show that
TiDE is capable of accurately predicting melt pool temperature and depth.
Second, we demonstrate that the proposed MPC achieves precise temperature
tracking while satisfying melt pool depth constraints within a targeted
dilution range (10\%-30\%), reducing potential porosity defects. Compared to
PID controller, the MPC results in smoother and less fluctuating laser power
profiles with competitive or superior melt pool temperature control
performance. This demonstrates the MPC's proactive control capabilities,
leveraging time-series prediction and real-time optimization, positioning it as
a powerful tool for future Digital Twin applications and real-time process
optimization in manufacturing.
| [
{
"version": "v1",
"created": "Fri, 10 Jan 2025 22:31:53 GMT"
},
{
"version": "v2",
"created": "Thu, 30 Jan 2025 14:20:33 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Mar 2025 20:56:35 GMT"
},
{
"version": "v4",
"created": "Fri, 14 Mar 2025 02:33:47 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Yi-Ping",
""
],
[
"Karkaria",
"Vispi",
""
],
[
"Tsai",
"Ying-Kuan",
""
],
[
"Rolark",
"Faith",
""
],
[
"Quispe",
"Daniel",
""
],
[
"Gao",
"Robert X.",
""
],
[
"Cao",
"Jian",
""
],
[
"Chen",
"Wei",
""
]
] |
2501.07820 | Jinger Wu | J.E. Wu and Q.S. Yan | Confront a dilaton model with the LHC measurements | null | null | null | null | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the scalar potential and investigate a couple of scenarios for the
symmetry breaking mechanisms with a dilaton model which is derived from the
geometry. We examine the LHC constraints for the couplings of Higgs-weak vector
bosons and Higgs self-couplings in this model, which identifies the parameter
space where the discovered Higgs boson $m_h=125$ GeV can be dilaton dominant
and the features of Higgs self-couplings are explored. It is found that via the
measurement of Higgs pair production, the High Luminosity LHC running can
either confirm or rule out the dilaton dominance.
| [
{
"version": "v1",
"created": "Tue, 14 Jan 2025 03:51:00 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 15:44:53 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wu",
"J. E.",
""
],
[
"Yan",
"Q. S.",
""
]
] |
2501.07907 | Tseleung So | Tseleung So | Steenrod operations for $4$-dimensional toric orbifolds | null | null | null | null | math.AT | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We prove necessary and sufficient conditions for the existence of non-trivial
Steenrod actions on the mod-$2$ cohomology of 4-dimensional toric orbifolds. As
applications, the stable homotopy type and the gauge groups of a
$4$-dimensional toric orbifold are determined, a partial solution to the
cohomological rigidity problem for $4$-dimensional toric orbifolds is provided,
and, in the smooth case, a combinatorial criterion is established for when the
toric orbifold is spin.
| [
{
"version": "v1",
"created": "Tue, 14 Jan 2025 07:50:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 05:24:55 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"So",
"Tseleung",
""
]
] |
2501.09474 | George Papadopoulos | Georgios Papadopoulos | Geometry and symmetries of Hermitian-Einstein and instanton connection
moduli spaces | 65 pages, minor corrections, more references added | null | null | null | hep-th math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the geometry of the moduli spaces
$\mathscr{M}_{\HE}^*(M^{2n})$ of Hermitian-Einstein irreducible connections on
a vector bundle $E$ over a K\"ahler with torsion (KT) manifold $M^{2n}$ that
admits holomorphic and $\h\nabla$-covariantly constant vector fields, where
$\h\nabla$ is the connection with skew-symmetric torsion $H$. We demonstrate
that such vector fields induce an action on $\mathscr{M}_{\HE}^*(M^{2n})$ that
leaves both the metric and complex structure invariant. Moreover, if an
additional condition is satisfied, the induced vector fields are covariantly
constant with respect to the connection with skew-symmetric torsion
$\h{\mathcal{ D}}$ on $\mathscr{M}_{\HE}^*(M^{2n})$. We demonstrate that in the
presence of such vector fields, the geometry of $\mathscr{M}_{\HE}^*(M^{2n})$
can be modelled on that of holomorphic toric principal bundles with base space
KT manifolds and give some examples.
We also extend our analysis to the moduli spaces
$\mathscr{M}_{\asd}^*(M^{4})$ of instanton connections on vector bundles over
KT, bi-KT (generalised K\"ahler) and hyper-K\"ahler with torsion (HKT)
manifolds $M^4$. We find that the geometry of $\mathscr{M}_{\asd}^*(S^3\times
S^1)$ can be modelled on that of principal bundles with fibre $S^3\times S^1$
over Quaternionic K\"ahler manifolds with torsion (QKT). In addition motivated
by applications to AdS/CFT, we explore the (superconformal) symmetry algebras
of two-dimensional sigma models with target spaces such moduli spaces.
| [
{
"version": "v1",
"created": "Thu, 16 Jan 2025 11:27:06 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Feb 2025 13:43:02 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 10:56:37 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Papadopoulos",
"Georgios",
""
]
] |
2501.09766 | Yirong Zeng | Yirong Zeng, Xiao Ding, Yuxian Wang, Weiwen Liu, Wu Ning, Yutai Hou,
Xu Huang, Bing Qin, Ting Liu | iTool: Boosting Tool Use of Large Language Models via Iterative
Reinforced Fine-Tuning | under review ACL | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Augmenting large language models (LLMs) with external tools is known as a
promising approach to enhancing their capabilities, especially for complex
tasks. Synthesizing tool-use data through real-world simulations is an
effective way to achieve it. Nevertheless, our investigation reveals that (1)
training gains significantly decay as synthetic data increases. The model
struggles to benefit from more synthetic data due to potential data diversity
issues, resulting in poor performance in complex scenarios. Moreover, we find
that (2) this challenge primarily manifests as minor discrepancies between the
model's output and the ground truth response (termed as deficiency), such as
errors in parameter values that require complex reasoning from the context to
resolve. To this end, we propose an iterative reinforced fine-tuning strategy
designed to alleviate these challenges. This strategy involves: (1) enhancing
the diversity of synthetic data through path exploration of Monte Carlo Tree
Search. (2) iteratively identifying deficiency-related data, constructing
fine-grained preference pairs to pinpoint deficiencies, and then applying
preference optimization to optimize these deficiencies. Our experiments show
that models trained using our method achieve about 12\% better performance than
baseline models, outperforming larger open-source and closed-source models.
| [
{
"version": "v1",
"created": "Wed, 15 Jan 2025 04:52:34 GMT"
},
{
"version": "v2",
"created": "Sun, 16 Feb 2025 13:51:09 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 05:05:03 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zeng",
"Yirong",
""
],
[
"Ding",
"Xiao",
""
],
[
"Wang",
"Yuxian",
""
],
[
"Liu",
"Weiwen",
""
],
[
"Ning",
"Wu",
""
],
[
"Hou",
"Yutai",
""
],
[
"Huang",
"Xu",
""
],
[
"Qin",
"Bing",
""
],
[
"Liu",
"Ting",
""
]
] |
2501.10149 | Jan Oliver Schunck | J.O. Schunck, P.S. Miedema, R.Y. Engel, S. Dziarzhytski, G. Brenner,
N. Ekanayake, C.-F. Chang, P. Bougiatioti, F. D\"oring, B. R\"osner, C.
David, C. Sch\"u{\ss}ler-Langeheine and M. Beye | Simultaneous mapping of the ultrafast time and fluence dependence of the
laser-induced insulator-to-metal transition in magnetite | The following article has been submitted to Structural Dynamics.
After it is published, it will be found at
https://publishing.aip.org/resources/librarians/products/journals/ | null | 10.1063/4.0000288 | null | cond-mat.mtrl-sci cond-mat.str-el | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Pump-probe methods are a ubiquitous tool in the field of ultrafast dynamic
measurements. In recent years, X-ray free-electron laser experiments have
gained importance due to their ability to probe with high chemical selectivity
and at atomic length scales. To obtain the complete dynamic information,
measurements are typically repeated many thousands of times with varying delay
and/or fluence settings. This generally necessitates that the sample fully
recovers before the subsequent excitation and that probe pulses and the induced
dynamic evolution are comparable to each other. These conditions present a
significant challenge when the sample fluctuates between different initial
states or when it is susceptible to damage. Also, source fluctuations are
normally intrinsic to free-electron laser pulses. Here, we present a
time-to-space mapping imaging scheme that enables us to record a delay range of
several picoseconds as well as a laser fluence range of about one order of
magnitude in every single shot of the x-ray probe. This approach can circumvent
the aforementioned preconditions. We demonstrate the use of this scheme by
mapping the femto- and picosecond dynamics of the optically induced
insulator-to-metal Verwey transition in a magnetite thin film. We employ
resonant diffraction at the free-electron laser FLASH for probing. The dynamics
of the magnetite thin film are found to follow a biexponential decay in line
with earlier studies on bulk crystals. By extrapolating our results towards the
conditions found at X-ray free-electron lasers with higher energy, we
demonstrate that the presented data could be recorded in a single shot.
| [
{
"version": "v1",
"created": "Fri, 17 Jan 2025 12:23:09 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Schunck",
"J. O.",
""
],
[
"Miedema",
"P. S.",
""
],
[
"Engel",
"R. Y.",
""
],
[
"Dziarzhytski",
"S.",
""
],
[
"Brenner",
"G.",
""
],
[
"Ekanayake",
"N.",
""
],
[
"Chang",
"C. -F.",
""
],
[
"Bougiatioti",
"P.",
""
],
[
"Döring",
"F.",
""
],
[
"Rösner",
"B.",
""
],
[
"David",
"C.",
""
],
[
"Schüßler-Langeheine",
"C.",
""
],
[
"Beye",
"M.",
""
]
] |
2501.10356 | Claire Chen | Claire Chen, Zhongchun Yu, Hojung Choi, Mark Cutkosky, Jeannette Bohg | DexForce: Extracting Force-informed Actions from Kinesthetic
Demonstrations for Dexterous Manipulation | Videos can be found here:
https://clairelc.github.io/dexforce.github.io/ | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Imitation learning requires high-quality demonstrations consisting of
sequences of state-action pairs. For contact-rich dexterous manipulation tasks
that require dexterity, the actions in these state-action pairs must produce
the right forces. Current widely-used methods for collecting dexterous
manipulation demonstrations are difficult to use for demonstrating contact-rich
tasks due to unintuitive human-to-robot motion retargeting and the lack of
direct haptic feedback. Motivated by these concerns, we propose DexForce.
DexForce leverages contact forces, measured during kinesthetic demonstrations,
to compute force-informed actions for policy learning. We collect
demonstrations for six tasks and show that policies trained on our
force-informed actions achieve an average success rate of 76% across all tasks.
In contrast, policies trained directly on actions that do not account for
contact forces have near-zero success rates. We also conduct a study ablating
the inclusion of force data in policy observations. We find that while using
force data never hurts policy performance, it helps most for tasks that require
advanced levels of precision and coordination, like opening an AirPods case and
unscrewing a nut.
| [
{
"version": "v1",
"created": "Fri, 17 Jan 2025 18:57:39 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 03:53:51 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Claire",
""
],
[
"Yu",
"Zhongchun",
""
],
[
"Choi",
"Hojung",
""
],
[
"Cutkosky",
"Mark",
""
],
[
"Bohg",
"Jeannette",
""
]
] |
2501.10954 | Mohammed Lemaalem | Mohammed Lemaalem, Selva Chandrasekaran Selvaraj, Ilias Papailias,
Naveen K. Dandu, Arash Namaeighasemi, Larry A. Curtiss, Amin Salehi-Khojin,
and Anh T. Ngo | Interactive Multiscale Modeling to Bridge Atomic Properties and
Electrochemical Performance in Li-CO$_2$ Battery Design | 40 pages (main paper), 5 figures, graphical abstract, and supporting
information file | null | null | null | cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | Li-CO$_2$ batteries show promise as energy storage solutions, offering high
theoretical energy density and CO$_2$ fixation. Their operation is based on the
formation and decomposition of Li$_2$CO$_3$/C during discharge and charge
cycles, respectively. We used a multiscale modeling framework that integrates
Density Functional Theory (DFT), Ab-Initio Molecular Dynamics (AIMD), classical
Molecular Dynamics (MD), and Finite Element Analysis (FEA) to investigate
atomic and cell-level properties. The considered Li-CO$_2$ battery consists of
a lithium metal anode, an ionic liquid electrolyte, and a carbon cloth porous
cathode with Sb$_{0.67}$Bi$_{1.33}$Te$_3$ as a catalyst. DFT and AIMD
determined the electrical conductivities of Sb$_{0.67}$Bi$_{1.33}$Te$_3$ and
Li$_2$CO$_3$ using the Kubo-Greenwood formalism and studied the CO$_2$
reduction mechanism on the cathode catalyst. MD simulations calculated the
CO$_2$ diffusion coefficient, Li$^+$ transference number, ionic conductivity,
and Li$^+$ solvation structure. The FEA model, incorporating results from
atomistic simulations, reproduced experimental voltage-capacity profiles at 1
mA/cm$^2$ and revealed spatio-temporal variations in Li$_2$CO$_3$/C deposition,
porosity, and CO$_2$ concentration dependence on discharge rates in the
cathode. Accordingly, Li$_2$CO$_3$ can form large and thin film deposits,
leading to dispersed and local porosity changes at 0.1 mA/cm$^2$ and 1
mA/cm$^2$, respectively. The capacity decreases exponentially from 81,570 mAh/g
at 0.1 mA/cm$^2$ to 6,200 mAh/g at 1 mA/cm$^2$, due to pore clogging from
excessive discharge product deposition that limits CO$_2$ transport to the
cathode interior. Therefore, the performance of Li-CO$_2$ batteries can be
improved by enhancing CO$_2$ transport, regulating Li$_2$CO$_3$ deposition, and
optimizing cathode architecture.
| [
{
"version": "v1",
"created": "Sun, 19 Jan 2025 05:54:10 GMT"
},
{
"version": "v2",
"created": "Fri, 24 Jan 2025 05:39:58 GMT"
},
{
"version": "v3",
"created": "Tue, 4 Feb 2025 20:23:35 GMT"
},
{
"version": "v4",
"created": "Mon, 3 Mar 2025 06:29:58 GMT"
},
{
"version": "v5",
"created": "Thu, 27 Mar 2025 07:39:48 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Lemaalem",
"Mohammed",
""
],
[
"Selvaraj",
"Selva Chandrasekaran",
""
],
[
"Papailias",
"Ilias",
""
],
[
"Dandu",
"Naveen K.",
""
],
[
"Namaeighasemi",
"Arash",
""
],
[
"Curtiss",
"Larry A.",
""
],
[
"Salehi-Khojin",
"Amin",
""
],
[
"Ngo",
"Anh T.",
""
]
] |
2501.11441 | Maria Taboada | Maria Taboada, Diego Martinez, Mohammed Arideh, Rosa Mosquera | Ontology Matching with Large Language Models and Prioritized Depth-First
Search | null | null | null | null | cs.IR cs.CL | http://creativecommons.org/licenses/by/4.0/ | Ontology matching (OM) plays a key role in enabling data interoperability and
knowledge sharing, but it remains challenging due to the need for large
training datasets and limited vocabulary processing in machine learning
approaches. Recently, methods based on Large Language Model (LLMs) have shown
great promise in OM, particularly through the use of a retrieve-then-prompt
pipeline. In this approach, relevant target entities are first retrieved and
then used to prompt the LLM to predict the final matches. Despite their
potential, these systems still present limited performance and high
computational overhead. To address these issues, we introduce MILA, a novel
approach that embeds a retrieve-identify-prompt pipeline within a prioritized
depth-first search (PDFS) strategy. This approach efficiently identifies a
large number of semantic correspondences with high accuracy, limiting LLM
requests to only the most borderline cases. We evaluated MILA using the
biomedical challenge proposed in the 2023 and 2024 editions of the Ontology
Alignment Evaluation Initiative. Our method achieved the highest F-Measure in
four of the five unsupervised tasks, outperforming state-of-the-art OM systems
by up to 17%. It also performed better than or comparable to the leading
supervised OM systems. MILA further exhibited task-agnostic performance,
remaining stable across all tasks and settings, while significantly reducing
LLM requests. These findings highlight that high-performance LLM-based OM can
be achieved through a combination of programmed (PDFS), learned (embedding
vectors), and prompting-based heuristics, without the need of domain-specific
heuristics or fine-tuning.
| [
{
"version": "v1",
"created": "Mon, 20 Jan 2025 12:29:09 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:29:21 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Taboada",
"Maria",
""
],
[
"Martinez",
"Diego",
""
],
[
"Arideh",
"Mohammed",
""
],
[
"Mosquera",
"Rosa",
""
]
] |
2501.11841 | Jiuling Zhang | Jiuling Zhang | Survey on Monocular Metric Depth Estimation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Monocular Depth Estimation (MDE) is fundamental to computer vision, enabling
spatial understanding, 3D reconstruction, and autonomous driving. Deep
learning-based MDE predicts relative depth from a single image, but the lack of
metric scale introduces inconsistencies, limiting applicability in tasks such
as visual SLAM, 3D reconstruction, and novel view synthesis. Monocular Metric
Depth Estimation (MMDE) overcomes this limitation by enabling precise
scene-scale inference, improving depth consistency, enhancing stability in
sequential tasks, and streamlining integration into practical systems. This
paper systematically reviews the evolution of depth estimation, from
traditional geometric methods to deep learning breakthroughs, emphasizing
scale-agnostic approaches in zero-shot generalization which is crucial for
advancing MMDE. Recent progress in zero-shot MMDE is examined, focusing on
challenges such as model generalization and boundary detail loss. To address
these issues, researchers have explored unlabeled data augmentation, image
patching, architectural optimization, and generative techniques. This review
analyzes these developments, assessing their impact and limitations. Key
findings are synthesized, unresolved challenges outlined, and future research
direction proposal. By providing a clear technical roadmap and insight into
emerging trends, this work aims to drive innovation and expand the real-world
applications of MMDE.
| [
{
"version": "v1",
"created": "Tue, 21 Jan 2025 02:51:10 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 03:21:42 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Jiuling",
""
]
] |
2501.12526 | Martin \v{C}ech | Martin \v{C}ech, Kaisa Matom\"aki | On optimality of mollifiers | 56 pages; compared to version 1, we now prove optimality of
Michel-Vanderkam mollifier among a wider class (Theorem 1.3) and have added a
discussion and proof of the optimality of the Iwaniec-Sarnak mollifier in its
class (Theorem 1.1) as well as a discussion on the unbalanced
Michel-Vanderkam mollifier (Section 10) | null | null | null | math.NT | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Mollifiers are used in a variety of contexts, for instance to study the
non-vanishing of $L$-functions. In this paper, we study the general question of
finding optimal mollifiers and provide criteria to identify them provided the
corresponding mollified moments can be computed. As an application, we study
the non-vanishing of central values of Dirichlet $L$-functions. In particular
we show that the Michel-Vanderkam mollifier is optimal in a wide class of
balanced two-piece mollifiers as well as provide the first published proof that
the Iwaniec-Sarnak mollifier is optimal in a wide class of one-piece
mollifiers.
| [
{
"version": "v1",
"created": "Tue, 21 Jan 2025 22:30:13 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:55:48 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Čech",
"Martin",
""
],
[
"Matomäki",
"Kaisa",
""
]
] |
2501.12911 | Abdulkadir Korkmaz | Abdulkadir Korkmaz and Praveen Rao | A Selective Homomorphic Encryption Approach for Faster
Privacy-Preserving Federated Learning | 23 pages, 32 figures | null | null | null | cs.CR cs.DC cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning is a machine learning method that supports training models
on decentralized devices or servers, where each holds its local data, removing
the need for data exchange. This approach is especially useful in healthcare,
as it enables training on sensitive data without needing to share them. The
nature of federated learning necessitates robust security precautions due to
data leakage concerns during communication. To address this issue, we propose a
new approach that employs selective encryption, homomorphic encryption,
differential privacy, and bit-wise scrambling to minimize data leakage while
achieving good execution performance. Our technique , FAS (fast and secure
federated learning) is used to train deep learning models on medical imaging
data. We implemented our technique using the Flower framework and compared with
a state-of-the-art federated learning approach that also uses selective
homomorphic encryption. Our experiments were run in a cluster of eleven
physical machines to create a real-world federated learning scenario on
different datasets. We observed that our approach is up to 90\% faster than
applying fully homomorphic encryption on the model weights. In addition, we can
avoid the pretraining step that is required by our competitor and can save up
to 46% in terms of total execution time. While our approach was faster, it
obtained similar security results as the competitor.
| [
{
"version": "v1",
"created": "Wed, 22 Jan 2025 14:37:44 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 21:23:54 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 17:44:27 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Korkmaz",
"Abdulkadir",
""
],
[
"Rao",
"Praveen",
""
]
] |
2501.13470 | Yuliang Gu | Yuliang Gu, Weilun Tsao, Bo Du, Thierry G\'eraud, Yongchao Xu | Leveraging Textual Anatomical Knowledge for Class-Imbalanced
Semi-Supervised Multi-Organ Segmentation | 11 pages | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Annotating 3D medical images demands substantial time and expertise, driving
the adoption of semi-supervised learning (SSL) for segmentation tasks. However,
the complex anatomical structures of organs often lead to significant class
imbalances, posing major challenges for deploying SSL in real-world scenarios.
Despite the availability of valuable prior information, such as inter-organ
relative positions and organ shape priors, existing SSL methods have yet to
fully leverage these insights. To address this gap, we propose a novel approach
that integrates textual anatomical knowledge (TAK) into the segmentation model.
Specifically, we use GPT-4o to generate textual descriptions of anatomical
priors, which are then encoded using a CLIP-based model. These encoded priors
are injected into the segmentation model as parameters of the segmentation
head. Additionally, contrastive learning is employed to enhance the alignment
between textual priors and visual features. Extensive experiments demonstrate
the superior performance of our method, significantly surpassing
state-of-the-art approaches. The source code will be available at:
https://github.com/Lunn88/TAK-Semi.
| [
{
"version": "v1",
"created": "Thu, 23 Jan 2025 08:40:54 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 09:23:15 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Gu",
"Yuliang",
""
],
[
"Tsao",
"Weilun",
""
],
[
"Du",
"Bo",
""
],
[
"Géraud",
"Thierry",
""
],
[
"Xu",
"Yongchao",
""
]
] |
2501.13834 | Pat Lank | Uttaran Dutta and Pat Lank and Kabeer Manali Rahul | Integral transforms on singularity categories for Noetherian schemes | Current: Upgraded results, add fully faithfullness between derived
splinters, kernels for equivalences between singular varieties. Previous: v1,
comments welcome! | null | null | null | math.AG math.AC | http://creativecommons.org/licenses/by/4.0/ | This work studies conditions under which integral transforms induce exact
functors on singularity categories between schemes that are proper over a
Noetherian base scheme. A complete characterization for this behavior is
provided, which extends earlier work of Ballard and Rizzardo. We leverage a
description of the bounded derived category of coherent sheaves as finite
cohomological functors on the category of perfect complexes, which is an
application of Neeman's approximable triangulated categories, to reduce
arguments to an affine local setting. Moreover, we study adjoints of such
functors, extend a result of Olander to varieties with mild singularities, and
provide an obstruction for derived equivalences between singular varieties.
| [
{
"version": "v1",
"created": "Thu, 23 Jan 2025 16:58:54 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:18:39 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Dutta",
"Uttaran",
""
],
[
"Lank",
"Pat",
""
],
[
"Rahul",
"Kabeer Manali",
""
]
] |
2501.13852 | Ibrahim Buba Garba | Ibrahim Buba Garba, Lorenzo Trombini, Claudine Katan, Jacky Even,
Marios Zacharias, Mikael Kepenekian, and George Volonakis | Three-Dimensional to Layered Halide Perovskites: A Parameter-Free Hybrid
Functional Method for Predicting Electronic Band Gaps | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate prediction of electronic band gaps in halide perovskites using
density functional theory is crucial for optoelectronic applications. Standard
hybrid functionals like HSE and PBE0 are becoming computationally accessible,
yet can fail at predicting the band gaps for three-dimensional (3D) and/or
layered halide perovskite. This study evaluates the doubly screened
dielectric-dependent hybrid (DSH) functional for predicting band gaps of Pb-
and Sn-based inorganic and hybrid 3D halide perovskites, as well as layered
hybrid perovskites. The DSH functional employs material-dependent mixing
parameters derived from macroscopic dielectric constants, and accurately
predicts band gaps for 3D perovskites only if structural local disorder is
taken into account. For layered hybrid perovskites, the DSH functional based on
average dielectric constants tends to overestimate the band gaps. To improve
predictions, we propose using the calculated dielectric constant of the
respective 3D perovskites to define the DSH screening. This method is then
applied to Pb- and Sn-based layered halide perovskites with various organic
spacers and multilayered structures, such as $BA_2MA_{n-1}Pb_{n}I_{3n-1}$ with
n =1, 2, 3, resulting in improved precision. The HSE functional systematically
underestimates band gaps in layered perovskites due to the missing non-local
long-range dielectric screening. On the other hand, the PBE0 is in good
agreement with the experimental values, in particular for the layered iodide
perovskites. The computational framework introduced here provides an efficient
parameter-free \textit{ab initio} methodology suitable for predicting the
electronic properties of 3D, layered halide perovskites and their
heterostructures, towards modelling materials for advanced optoelectronic
devices
| [
{
"version": "v1",
"created": "Thu, 23 Jan 2025 17:20:34 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:47:47 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Garba",
"Ibrahim Buba",
""
],
[
"Trombini",
"Lorenzo",
""
],
[
"Katan",
"Claudine",
""
],
[
"Even",
"Jacky",
""
],
[
"Zacharias",
"Marios",
""
],
[
"Kepenekian",
"Mikael",
""
],
[
"Volonakis",
"George",
""
]
] |
2501.14691 | Andrew Hardt | Andrew Hardt and David Wallach | Stability of products of double Grothendieck polynomials | 14 pages | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We prove that products of double Grothendieck polynomials have the same back-
and forward-stability numbers as products of Schubert polynomials, characterize
which simple reflections appear in such products, and also give a new proof of
a finiteness conjecture of Lam-Lee-Shimozono on products of back-stable
Grothendieck polynomials which was first proved by Anderson. To do this, we use
the main theorems from our recent work, as well as expansion formulas of
Lenart, Fomin-Kirillov, and Lam-Lee-Shimozono.
| [
{
"version": "v1",
"created": "Fri, 24 Jan 2025 18:05:11 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:21:00 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Hardt",
"Andrew",
""
],
[
"Wallach",
"David",
""
]
] |
2501.15128 | Haixia Liu | Pingping Tao, Haixia Liu, Jing Su, Xiaochen Yang and Hongchen Tan | MAP-based Problem-Agnostic diffusion model for Inverse Problems | 17 pages, 10 figures | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion models have indeed shown great promise in solving inverse problems
in image processing. In this paper, we propose a novel, problem-agnostic
diffusion model called the maximum a posteriori (MAP)-based guided term
estimation method for inverse problems. To leverage unconditionally pretrained
diffusion models to address conditional generation tasks, we divide the
conditional score function into two terms according to Bayes' rule: an
unconditional score function (approximated by a pretrained score network) and a
guided term, which is estimated using a novel MAP-based method that
incorporates a Gaussian-type prior of natural images. This innovation allows us
to better capture the intrinsic properties of the data, leading to improved
performance. Numerical results demonstrate that our method preserves contents
more effectively compared to state-of-the-art methods--for example, maintaining
the structure of glasses in super-resolution tasks and producing more coherent
results in the neighborhood of masked regions during inpainting.
| [
{
"version": "v1",
"created": "Sat, 25 Jan 2025 08:30:15 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 09:10:42 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Tao",
"Pingping",
""
],
[
"Liu",
"Haixia",
""
],
[
"Su",
"Jing",
""
],
[
"Yang",
"Xiaochen",
""
],
[
"Tan",
"Hongchen",
""
]
] |
2501.15303 | Yoshiki Nakamura | Yoshiki Nakamura | Guarded Negation Transitive Closure Logic is 2-EXPTIME-complete | null | null | null | null | cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider guarded negation transitive closure logic (GNTC). In this paper,
we show that the satisfiability problem for GNTC is in 2-EXPTIME (hence,
2-EXPTIME-complete from existing lower bound results), which improves the
previously known non-elementary time upper bound. This extends previously known
2-EXPTIME upper bound results, e.g., for the guarded negation fragment of
first-order logic, the unary negation fragment of first-order logic with
regular path expressions, propositional dynamic logic (PDL) with intersection
and converse, and CPDL+ (an extension of PDL with conjunctive queries) of
bounded treewidth. To this end, we present a sound and complete local model
checker on tree decompositions. This system has a closure property of size
single exponential, and it induces a reduction from the satisfiability problem
for GNTC into the non-emptiness problem for 2-way (weak) alternating parity
tree automata in single exponential time. Additionally, we investigate the
complexity of satisfiability and model checking for fragments of GNTC, such as
guarded (quantification) fragments, unary negation fragments, and existential
positive fragments.
| [
{
"version": "v1",
"created": "Sat, 25 Jan 2025 19:00:51 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:44:35 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Nakamura",
"Yoshiki",
""
]
] |
2501.15801 | Fei Hou | Fei Hou | Hybrid bounds for ${\rm{GL}}(4)\times {\rm{GL}}(1)$ twisted
$L$-functions | We added an appendix by showing Lemma 2.1 | null | null | null | math.NT | http://creativecommons.org/licenses/by/4.0/ | Let $P,M$ be a two primes such that $(P,M)=1$. Let $\Pi$ be a normalized
Hecke-Maa\ss\ form on ${\rm{GL}}(4)$ of level $P$, and $\chi$ a primitive
Dirichlet character modulo $M$. In this paper, we study the hybrid subconvexity
problem for $L(s, \Pi\otimes \chi)$ simultaneously in the level and conductor
aspects. Among other things, we prove a hybrid subconvex bound, so long as
$M^{1/5}<P<M^{2/5}$.
| [
{
"version": "v1",
"created": "Mon, 27 Jan 2025 06:06:29 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 15:46:56 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Hou",
"Fei",
""
]
] |
2501.16391 | Xiaoqing Lian | Xiaoqing Lian, Jie Zhu, Tianxu Lv, Shiyun Nie, Hang Fan, Guosheng Wu,
Yunjun Ge, Lihua Li, Xiangxiang Zeng, Xiang Pan | Inductive-Associative Meta-learning Pipeline with Human Cognitive
Patterns for Unseen Drug-Target Interaction Prediction | null | null | null | null | cs.LG cs.AI q-bio.BM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Significant differences in protein structures hinder the generalization of
existing drug-target interaction (DTI) models, which often rely heavily on
pre-learned binding principles or detailed annotations. In contrast, BioBridge
designs an Inductive-Associative pipeline inspired by the workflow of
scientists who base their accumulated expertise on drawing insights into novel
drug-target pairs from weakly related references. BioBridge predicts novel
drug-target interactions using limited sequence data, incorporating multi-level
encoders with adversarial training to accumulate transferable binding
principles. On these principles basis, BioBridge employs a dynamic prototype
meta-learning framework to associate insights from weakly related annotations,
enabling robust predictions for previously unseen drug-target pairs. Extensive
experiments demonstrate that BioBridge surpasses existing models, especially
for unseen proteins. Notably, when only homologous protein binding data is
available, BioBridge proves effective for virtual screening of the epidermal
growth factor receptor and adenosine receptor, underscoring its potential in
drug discovery.
| [
{
"version": "v1",
"created": "Sun, 26 Jan 2025 08:22:22 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 07:41:02 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Lian",
"Xiaoqing",
""
],
[
"Zhu",
"Jie",
""
],
[
"Lv",
"Tianxu",
""
],
[
"Nie",
"Shiyun",
""
],
[
"Fan",
"Hang",
""
],
[
"Wu",
"Guosheng",
""
],
[
"Ge",
"Yunjun",
""
],
[
"Li",
"Lihua",
""
],
[
"Zeng",
"Xiangxiang",
""
],
[
"Pan",
"Xiang",
""
]
] |
2501.16395 | Ivan Fernandez-Val | Jaap H. Abbring, Victor Chernozhukov and Iv\'an Fern\'andez-Val | Philip G. Wright, directed acyclic graphs, and instrumental variables | 21 pages, 5 figures, this version fixes a typo in the previous
version | The Econometrics Journal 28(1), (2025) | 10.1093/ectj/utaf006 | null | econ.EM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Wright (1928) deals with demand and supply of oils and butter. In Appendix B
of this book, Philip Wright made several fundamental contributions to causal
inference. He introduced a structural equation model of supply and demand,
established the identification of supply and demand elasticities via the method
of moments and directed acyclical graphs, developed empirical methods for
estimating demand elasticities using weather conditions as instruments, and
proposed methods for counterfactual analysis of the welfare effect of imposing
tariffs and taxes. Moreover, he took all of these methods to data. These ideas
were far ahead, and much more profound than, any contemporary theoretical and
empirical developments on causal inference in statistics or econometrics. This
editorial aims to present P. Wright's work in a more modern framework, in a
lecture note format that can be useful for teaching and linking to contemporary
research.
| [
{
"version": "v1",
"created": "Sun, 26 Jan 2025 16:29:40 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:34:20 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Abbring",
"Jaap H.",
""
],
[
"Chernozhukov",
"Victor",
""
],
[
"Fernández-Val",
"Iván",
""
]
] |
2501.16417 | Subir Sachdev | Subir Sachdev | The foot, the fan, and the cuprate phase diagram: Fermi-volume-changing
quantum phase transitions | 42 pages, 14 figures; Contribution to the memorial volume for Jan
Zaanen; Expanded discussion on evidence for fractionalization in the
pseudogap, and its consequences | null | null | null | cond-mat.str-el hep-th quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A Fermi liquid with a 'large' Fermi surface (FL) can have a quantum phase
transition to a spin density wave state (SDW) with reconstructed 'small' Fermi
pockets. Both FL and SDW phases obey the Luttinger constraints on the volume
enclosed by the Fermi surfaces. Critical spin fluctuations lead to spin-singlet
$d$-wave pairing, as observed in the cuprates. Studies of the influence of
spatial disorder on the FL-SDW quantum phase transition predict an extended
quantum-critical Griffiths-type phase at low temperatures on the large Fermi
surface side. These computations agree with the 'foot' of strange metal
transport, and recent low temperature neutron scattering observations on
La$_{2-x}$Sr$_x$CuO$_4$.
However, this theory cannot explain the higher temperature pseudogap and the
'fan' of strange metal behavior of the hole-doped cuprates. Here we need to
consider underlying Fermi-volume-changing quantum phase transitions without
symmetry breaking. Then the small Fermi surface phase does not obey the
Luttinger constraint, and the pseudogap metal is described by thermal
fluctuations above a 'fractionalized Fermi liquid' (FL*) or a 'holon metal',
with the descriptions related by a duality on a background spin liquid. The
quantum critical fan is described using a field theory for an underlying FL-FL*
quantum phase transition in the presence of spatial disorder. This field theory
can be mapped to a form which can be analyzed using the methods of the
Sachdev-Ye-Kitaev model. Such an analysis successfully models
linear-in-temperature resistivity, optical conductivity and thermopower
observations in the quantum critical fan.
The confinement crossover connecting these lower and higher temperature
descriptions is also discussed.
| [
{
"version": "v1",
"created": "Mon, 27 Jan 2025 19:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 29 Jan 2025 17:40:56 GMT"
},
{
"version": "v3",
"created": "Sun, 2 Feb 2025 21:59:00 GMT"
},
{
"version": "v4",
"created": "Sun, 16 Feb 2025 21:08:52 GMT"
},
{
"version": "v5",
"created": "Tue, 4 Mar 2025 18:49:26 GMT"
},
{
"version": "v6",
"created": "Wed, 26 Mar 2025 19:55:38 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Sachdev",
"Subir",
""
]
] |
2501.17463 | Caroline Haslebacher | Lutz Duembgen, Caroline Haslebacher | Nonparametric Smoothing of Directional and Axial Data | null | null | null | null | stat.ME astro-ph.EP stat.CO | http://creativecommons.org/licenses/by/4.0/ | We discuss generalized linear models for directional data where the
conditional distribution of the response is a von Mises-Fisher distribution in
arbitrary dimension or a Bingham distribution on the unit circle. To do this
properly, we parametrize von Mises-Fisher distributions by Euclidean parameters
and investigate computational aspects of this parametrization. Then we modify
this approach for local polynomial regression as a means of nonparametric
smoothing of distributional data. The methods are illustrated with simulated
data and a data set from planetary sciences involving covariate vectors on a
sphere with axial response.
| [
{
"version": "v1",
"created": "Wed, 29 Jan 2025 07:57:08 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Mar 2025 19:21:49 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Duembgen",
"Lutz",
""
],
[
"Haslebacher",
"Caroline",
""
]
] |
2501.17922 | Isabel Medlock | Isabel Medlock, Daisuke Nagai, Daniel Angl\'es Alc\'azar, Matthew
Gebhardt | Constraining Baryonic Feedback Effects on the Matter Power Spectrum with
Fast Radio Bursts | 15 pages, 7 figures, accepted to ApJ | null | null | null | astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | In the age of large-scale galaxy and lensing surveys, such as DESI, Euclid,
Roman and Rubin, we stand poised to usher in a transformative new phase of
data-driven cosmology. To fully harness the capabilities of these surveys, it
is critical to constrain the poorly understood influence of baryon feedback
physics on the matter power spectrum. We investigate the use of a powerful and
novel cosmological probe - fast radio bursts (FRBs) - to capture baryonic
effects on the matter power spectrum, leveraging simulations from the CAMELS
projects, including IllustrisTNG, SIMBA, and Astrid. We find that FRB
statistics exhibit a strong correlation, independent of the subgrid model and
cosmology, with quantities known to encapsulate baryonic impacts on the matter
power spectrum, such as baryon spread and the halo baryon fraction. We propose
an innovative method utilizing FRB observations to quantify the effects of
feedback physics and enhance weak lensing measurements of $S_{8}$. We outline
the necessary steps to prepare for the imminent detection of large FRB
populations in the coming years, focusing on understanding the redshift
evolution of FRB observables and mitigating the effects of cosmic variance.
| [
{
"version": "v1",
"created": "Wed, 29 Jan 2025 19:00:02 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Mar 2025 11:02:21 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 09:17:53 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Medlock",
"Isabel",
""
],
[
"Nagai",
"Daisuke",
""
],
[
"Alcázar",
"Daniel Anglés",
""
],
[
"Gebhardt",
"Matthew",
""
]
] |
2501.18038 | James Brusseau | James Brusseau | A Case Study in Acceleration AI Ethics: The TELUS GenAI Conversational
Agent | null | null | null | null | cs.CY | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Acceleration ethics addresses the tension between innovation and safety in
artificial intelligence. The acceleration argument is that risks raised by
innovation should be answered with still more innovating. This paper summarizes
the theoretical position, and then shows how acceleration ethics works in a
real case. To begin, the paper summarizes acceleration ethics as composed of
five elements: innovation solves innovation problems, innovation is
intrinsically valuable, the unknown is encouraging, governance is
decentralized, ethics is embedded. Subsequently, the paper illustrates the
acceleration framework with a use-case, a generative artificial intelligence
language tool developed by the Canadian telecommunications company Telus. While
the purity of theoretical positions is blurred by real-world ambiguities, the
Telus experience indicates that acceleration AI ethics is a way of maximizing
social responsibility through innovation, as opposed to sacrificing social
responsibility for innovation, or sacrificing innovation for social
responsibility.
| [
{
"version": "v1",
"created": "Wed, 29 Jan 2025 22:57:56 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 17:50:32 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Brusseau",
"James",
""
]
] |
2501.18185 | Wenjian Liu | Ning Zhang, Qingpeng Wang, and Wenjian Liu | MetaWave: A Platform for Unified Implementation of Nonrelativistic and
Relativistic Wavefunctions | 59 papges, 11 figures | null | null | null | physics.chem-ph | http://creativecommons.org/licenses/by/4.0/ | \texttt{MetaWave} is a C++ template-based architecture designed for unified
implementation of nonrelativistic and relativistic wavefunction-based quantum
chemical methods. It is highly modular, extendable, and efficient. This is
achieved by decoupling the three distinct aspects of quantum chemical methods
(i.e., nature of Hamiltonian, structure of wavefunction, and strategy of
parallelization ), thereby allowing for separate treatment of them through
their internal type-trait and tagging systems furnished by C++ metaprogramming.
Once the second-quantized Hamiltonians, whether nonrelativistic (spin-free) or
relativistic (spin-dependent), are decomposed into topologically equivalent
diagrams for a unified evaluation of the basic coupling coefficients between
(randomly selected) spin-free or spin-dependent configuration state functions
or Slater determinants incorporating full molecular symmetry (including single
or double point group and spin or time reversal symmetry), the many-electron
wavefunctions, whether built up with scalar or spinor orbitals, can be
assembled with the same templates. As for parallelization, \texttt{MetaWave}
supports both OpenMP and MPI, with the majority of the latter being translated
automatically from its OpenMP counterparts.The whole structure of
\texttt{MetaWave} is reviewed here, with some showcases for illustrating its
performance.
| [
{
"version": "v1",
"created": "Thu, 30 Jan 2025 07:35:52 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 08:23:48 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Ning",
""
],
[
"Wang",
"Qingpeng",
""
],
[
"Liu",
"Wenjian",
""
]
] |
2501.18504 | Dr Peter J. Bentley | Peter J. Bentley, Soo Ling Lim, Fuyuki Ishikawa | CLEAR: Cue Learning using Evolution for Accurate Recognition Applied to
Sustainability Data Extraction | 9 pages plus 2 pages of supplemental material | Proceedings of the Genetic and Evolutionary Computation Conference
2025 (GECCO 25). ACM, Malaga, Spain | null | null | cs.CV cs.AI cs.NE | http://creativecommons.org/licenses/by/4.0/ | Large Language Model (LLM) image recognition is a powerful tool for
extracting data from images, but accuracy depends on providing sufficient cues
in the prompt - requiring a domain expert for specialized tasks. We introduce
Cue Learning using Evolution for Accurate Recognition (CLEAR), which uses a
combination of LLMs and evolutionary computation to generate and optimize cues
such that recognition of specialized features in images is improved. It
achieves this by auto-generating a novel domain-specific representation and
then using it to optimize suitable textual cues with a genetic algorithm. We
apply CLEAR to the real-world task of identifying sustainability data from
interior and exterior images of buildings. We investigate the effects of using
a variable-length representation compared to fixed-length and show how LLM
consistency can be improved by refactoring from categorical to real-valued
estimates. We show that CLEAR enables higher accuracy compared to expert human
recognition and human-authored prompts in every task with error rates improved
by up to two orders of magnitude and an ablation study evincing solution
concision.
| [
{
"version": "v1",
"created": "Thu, 30 Jan 2025 17:13:32 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 22:19:20 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Bentley",
"Peter J.",
""
],
[
"Lim",
"Soo Ling",
""
],
[
"Ishikawa",
"Fuyuki",
""
]
] |
2501.18523 | Sumner Harris | Sumner B. Harris, Patrick T. Gemperline, Christopher M. Rouleau, Rama
K. Vasudevan, Ryan B. Comes | Deep learning with reflection high-energy electron diffraction images to
predict cation ratio in Sr$_{2x}$Ti$_{2(1-x)}$O$_{3}$ thin films | null | null | null | null | cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | Machine learning (ML) with in situ diagnostics offers a transformative
approach to accelerate, understand, and control thin film synthesis by
uncovering relationships between synthesis conditions and material properties.
In this study, we demonstrate the application of deep learning to predict the
stoichiometry of Sr$_{2x}$Ti$_{2(1-x)}$O$_{3}$ thin films using reflection
high-energy electron diffraction images acquired during pulsed laser
deposition. A gated convolutional neural network trained for regression of the
Sr atomic fraction achieved accurate predictions with a small dataset of 31
samples. Explainable AI techniques revealed a previously unknown correlation
between diffraction streak features and cation stoichiometry in
Sr$_{2x}$Ti$_{2(1-x)}$O$_{3}$ thin films. Our results demonstrate how ML can be
used to transform a ubiquitous in situ diagnostic tool, that is usually limited
to qualitative assessments, into a quantitative surrogate measurement of
continuously valued thin film properties. Such methods are critically needed to
enable real-time control, autonomous workflows, and accelerate traditional
synthesis approaches.
| [
{
"version": "v1",
"created": "Thu, 30 Jan 2025 17:41:20 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:17:38 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Harris",
"Sumner B.",
""
],
[
"Gemperline",
"Patrick T.",
""
],
[
"Rouleau",
"Christopher M.",
""
],
[
"Vasudevan",
"Rama K.",
""
],
[
"Comes",
"Ryan B.",
""
]
] |
2501.19096 | Alexis Gu\'erin | Alexis Gu\'erin and Felix Roz | An action of the Witt algebra on Khovanov-Rozansky homology | 49 pages, many figures | null | null | null | math.GT math.QA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We construct an action of the positive part of the Witt algebra on the
Khovanov-Rozansky $\mathfrak{gl}_N$-link homology and prove that this
construction is functorial.
| [
{
"version": "v1",
"created": "Fri, 31 Jan 2025 12:41:52 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:34:55 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Guérin",
"Alexis",
""
],
[
"Roz",
"Felix",
""
]
] |
2502.00471 | Pavel Chebotarev | Pavel Chebotarev | Evolution of Society Caused by Collective and Individual Decisions | 15 pages, 9 figures, a converence submission | null | null | null | physics.soc-ph cs.GT math.OC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Decision-making societies may vary in their level of cooperation and degree
of conservatism, both of which influence their overall performance. Moreover,
these factors are not fixed -- they can change based on the decisions agents in
the society make in their interests. But can these changes lead to cyclical
patterns in societal evolution? To explore this question, we use the ViSE
(Voting in Stochastic Environment) model. In this framework, the level of
cooperation can be measured by group size, while the degree of conservatism is
determined by the voting threshold. Agents can adopt either individualistic or
group-oriented strategies when voting on stochastically generated external
proposals. For Gaussian proposal generators, the expected capital gain (ECG) --
a measure of agents' performance -- can be expressed in standard mathematical
functions. Our findings show that in neutral environments, societal evolution
with open or democratic groups can follow cyclic patterns. We also find that
highly conservative societies or conservative societies with low levels of
cooperation can evolve into liberal (less conservative than majoritarian)
societies and that mafia groups never let their members go when they want to.
| [
{
"version": "v1",
"created": "Sat, 1 Feb 2025 15:56:19 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:31:20 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chebotarev",
"Pavel",
""
]
] |
2502.01251 | David MacTaggart | David MacTaggart | On field line slippage rates in the solar corona | Accepted for Solar Physics | null | null | null | astro-ph.SR | http://creativecommons.org/licenses/by/4.0/ | Magnetic reconnection is one of the fundamental dynamical processes in the
solar corona. The method of studying reconnection in active region-scale
magnetic fields generally depends on non-local methods (i.e. requiring
information across the magnetic field under study) of magnetic topology, such
as separatrix skeletons and quasi-separatrix layers. The theory of General
Magnetic Reconnection is also non-local, in that its measure of the
reconnection rate depends on determining the maxima of integrals along field
lines. In this work, we complement the above approaches by introducing a local
description of magnetic reconnection, that is one in which information about
reconnection at a particular location depends only on quantities at that
location. This description connects the concept of the field line slippage
rate, relative to ideal motion, to the underlying local geometry of the
magnetic field characterized in terms of the Lorentz force and field-aligned
current density. It is argued that the dominant non-ideal term for the solar
corona, discussed in relation to this new description, is mathematically
equivalent to the anomalous resistivity employed by many magnetohrdrodynamic
simulations. However, the general application of this new approach is adaptable
to the inclusion of other non-ideal terms, which may arise from turbulence
modelling or the inclusion of a generalized Ohm's law. The approach is
illustrated with two examples of coronal magnetic fields related to flux ropes:
an analytical model and a nonlinear force-free extrapolation. In terms of the
latter, the slippage rate corresponds to the reconnection which would happen if
the given (static) force-free equilibrium were the instantaneous form of the
magnetic field governed by an Ohm's law with non-ideal terms.
| [
{
"version": "v1",
"created": "Mon, 3 Feb 2025 11:20:47 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 10:03:38 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"MacTaggart",
"David",
""
]
] |
2502.01749 | David Sutter | Giulia Mazzola, David Sutter, Renato Renner | Uhlmann's theorem for relative entropies | 25 pages, added Appendix A about the relation between sufficiency and
Uhlmann property | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Uhlmann's theorem states that, for any two quantum states $\rho_{AB}$ and
$\sigma_A$, there exists an extension $\sigma_{AB}$ of $\sigma_A$ such that the
fidelity between $\rho_{AB}$ and $\sigma_{AB}$ equals the fidelity between
their reduced states $\rho_A$ and $\sigma_A$. In this work, we generalize
Uhlmann's theorem to $\alpha$-R\'enyi relative entropies for $\alpha \in
[\frac{1}{2},\infty]$, a family of divergences that encompasses fidelity,
relative entropy, and max-relative entropy corresponding to
$\alpha=\frac{1}{2}$, $\alpha=1$, and $\alpha=\infty$, respectively.
| [
{
"version": "v1",
"created": "Mon, 3 Feb 2025 19:00:12 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 10:10:30 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Mazzola",
"Giulia",
""
],
[
"Sutter",
"David",
""
],
[
"Renner",
"Renato",
""
]
] |
2502.01894 | Goodarz Mehr | Goodarz Mehr and Azim Eskandarian | SimBEV: A Synthetic Multi-Task Multi-Sensor Driving Data Generation Tool
and Dataset | null | null | null | null | cs.CV cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Bird's-eye view (BEV) perception has garnered significant attention in
autonomous driving in recent years, in part because BEV representation
facilitates multi-modal sensor fusion. BEV representation enables a variety of
perception tasks including BEV segmentation, a concise view of the environment
useful for planning a vehicle's trajectory. However, this representation is not
fully supported by existing datasets, and creation of new datasets for this
purpose can be a time-consuming endeavor. To address this challenge, we
introduce SimBEV. SimBEV is a randomized synthetic data generation tool that is
extensively configurable and scalable, supports a wide array of sensors,
incorporates information from multiple sources to capture accurate BEV ground
truth, and enables a variety of perception tasks including BEV segmentation and
3D object detection. SimBEV is used to create the SimBEV dataset, a large
collection of annotated perception data from diverse driving scenarios. SimBEV
and the SimBEV dataset are open and available to the public.
| [
{
"version": "v1",
"created": "Tue, 4 Feb 2025 00:00:06 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 20:42:44 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Mehr",
"Goodarz",
""
],
[
"Eskandarian",
"Azim",
""
]
] |
2502.02389 | Pau Colomer | Pau Colomer, Christian Deppe, Holger Boche, Andreas Winter | Rate-reliability functions for deterministic identification | 12 pages, 2 figures. A preliminary version of this work has been
accepted for presentation at the 2025 IEEE International Conference on
Communications, Montreal (Canada) 8-12 June 2025 | null | null | null | cs.IT math.IT quant-ph | http://creativecommons.org/licenses/by/4.0/ | We investigate deterministic identification over arbitrary memoryless
channels under the constraint that the error probabilities of first and second
kind are exponentially small in the block length $n$, controlled by reliability
exponents $E_1,E_2 \geq 0$. In contrast to the regime of slowly vanishing
errors, where the identifiable message length scales as $\Theta(n\log n)$, here
we find that for positive exponents linear scaling is restored, now with a rate
that is a function of the reliability exponents. We give upper and lower bounds
on the ensuing rate-reliability function in terms of (the logarithm of) the
packing and covering numbers of the channel output set, which for small error
exponents $E_1,E_2>0$ can be expanded in leading order as the product of the
Minkowski dimension of a certain parametrisation the channel output set and
$\log\min\{E_1,E_2\}$. These allow us to recover the previously observed
slightly superlinear identification rates, and offer a different perspective
for understanding them in more traditional information theory terms. We further
illustrate our results with a discussion of the case of dimension zero, and
extend them to classical-quantum channels and quantum channels with tensor
product input restriction.
| [
{
"version": "v1",
"created": "Tue, 4 Feb 2025 15:09:14 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:42:43 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Colomer",
"Pau",
""
],
[
"Deppe",
"Christian",
""
],
[
"Boche",
"Holger",
""
],
[
"Winter",
"Andreas",
""
]
] |
2502.02474 | Yanbiao Gan | Zhongyi He, Yanbiao Gan, Bin Yang, Demei Li, Huilin Lai, Aiguo Xu | Multiscale thermodynamic nonequilibrium effects in Kelvin-Helmholtz
instability and their relative importance | null | Physics of Fluids 37, 034131 (2025) | null | null | physics.flu-dyn physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study investigates the complex kinetics of thermodynamic nonequilibrium
effects (TNEs) and their relative importance during the development of
Kelvin-Helmholtz instability (KHI) using high-order discrete Boltzmann models
(DBMs). First, the capabilities and differences among various discrete velocity
sets in capturing TNEs and distribution functions are assessed. Practical
guidelines for constructing discrete velocity stencils are proposed to enhance
phase-space discretization and improve the robustness of high-order DBM
simulation. At different stages of KHI and under varying initial conditions,
multiscale TNEs, such as viscous stresses of different orders, emerge with
distinct dominant roles. Specifically, three scenarios are identified: (i)
regimes dominated by first-order TNEs,(ii) alternation between first- and
second-order TNEs, and (iii) states where second-order TNEs govern the system's
behavior. To quantitatively capture these transitions, criteria for TNE
dominance at different orders in KHI evolution are established based on the
relative thermodynamic nonequilibrium intensity (\(R_{\text{TNE}}\)). In
scenarios dominated by second-order TNEs, differences between first-order and
second-order models are compared in terms of macroscopic quantities,
nonequilibrium effects, and kinetic moments, revealing the physical limitations
of low-order models in capturing TNEs. Furthermore, the effectiveness,
extensibility, and limitations of a representative high-order model are
examined under second-order TNE-dominated conditions. To encapsulate these
findings, a nonequilibrium phase diagram that visually maps the multiscale
characteristics of KHI is constructed. This diagram not only provides intuitive
insights into the dynamic interplay of different nonequilibrium effects but
also serves as a kinetic roadmap for selecting suitable models under diverse
nonequilibrium conditions.
| [
{
"version": "v1",
"created": "Tue, 4 Feb 2025 16:48:57 GMT"
},
{
"version": "v2",
"created": "Thu, 6 Mar 2025 10:57:42 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 14:14:36 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"He",
"Zhongyi",
""
],
[
"Gan",
"Yanbiao",
""
],
[
"Yang",
"Bin",
""
],
[
"Li",
"Demei",
""
],
[
"Lai",
"Huilin",
""
],
[
"Xu",
"Aiguo",
""
]
] |
2502.04831 | Manpreet Kaur | Manpreet Kaur, Sanjeev Kumar and T. Nandi | Precise charge state distribution of projectile ions through solid
targets | 11 pages, 16 figures, submitting in journal soon | null | null | null | physics.atom-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The charge state distribution (CSD) of the projectile ions through solid
targets in the intermediate energy range (1 MeV/u $<$ E $<$ 4 MeV/u) has a
major impact on the collision of the ion atom and accelerator physics. We
explore the mean charge states taken from the empirical formula [Schiwietz
$et~al.$, Nucl. Inst. Meths. {\bf 225}, 4(2004)] are only good for projectile
ions with $Z_1 \le 16$. To solve this issue, we develop a model in which
instead of a single formula, if we employ four formulae, the comparative
picture between experimental and empirical data becomes impressive.
Furthermore, the CSDs with the mean charge state so obtained and the Gaussian
distribution function having distribution width given by [Novikov and Teplove,
Phys. Lett. {\bf378}, 1286(2014)] compare well with the experimentally measured
CSDs for the entire range of projectile ions. We believe that precise CSDs will
be highly useful in both ion-atom collision and accelerator physics.
| [
{
"version": "v1",
"created": "Fri, 7 Feb 2025 11:05:54 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Feb 2025 15:48:33 GMT"
},
{
"version": "v3",
"created": "Wed, 26 Mar 2025 18:03:26 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kaur",
"Manpreet",
""
],
[
"Kumar",
"Sanjeev",
""
],
[
"Nandi",
"T.",
""
]
] |
2502.04921 | Zhandos Moldabekov Abdykanovich | Zhandos A. Moldabekov, Sebastian Schwalbe, Thomas Gawne, Thomas R.
Preston, Jan Vorberger, and Tobias Dornheim | Applying the Liouville-Lanczos Method of Time-Dependent
Density-Functional Theory to Warm Dense Matter | null | null | null | null | physics.plasm-ph physics.chem-ph physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Ab initio modeling of dynamic structure factors (DSF) and related density
response properties in the warm dense matter (WDM) regime is a challenging
computational task. The DSF, convolved with a probing X-ray beam and instrument
function, is measured in X-ray Thomson scattering (XRTS) experiments, which
allows for the study of electronic structure properties at the microscopic
level. Among the various ab initio methods, linear response time-dependent
density functional theory (LR-TDDFT) is a key framework for simulating the DSF.
The standard approach in LR-TDDFT for computing the DSF relies on the orbital
representation. A significant drawback of this method is the unfavorable
scaling of the number of required empty bands as the wavenumber increases,
making LR-TDDFT impractical for modeling XRTS measurements over large energy
scales, such as in backward scattering geometry. We consider and test an
alternative approach that employs the Liouville-Lanczos (LL) method for
simulating the DSF. This approach does not require empty states and allows the
DSF at large momentum transfer values and over a broad frequency range to be
accessed. We compare the results obtained from the LL method with those from
the standard LR-TDDFT within the projector augmented-wave formalism for
isochorically heated aluminum and warm dense hydrogen. Additionally, we utilize
exact path integral Monte Carlo (PIMC) results for the imaginary-time
density-density correlation function (ITCF) of warm dense hydrogen to
rigorously benchmark the LL approach. We discuss the application of the LL
method for calculating DSFs and ITCFs at different wavenumbers, the effects of
pseudopotentials, and the role of Lorentzian smearing. The successful
validation of the LL method under WDM conditions makes it a valuable addition
to the ab initio simulation landscape, supporting experimental efforts and
advancing WDM theory.
| [
{
"version": "v1",
"created": "Fri, 7 Feb 2025 13:38:00 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 09:53:50 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Moldabekov",
"Zhandos A.",
""
],
[
"Schwalbe",
"Sebastian",
""
],
[
"Gawne",
"Thomas",
""
],
[
"Preston",
"Thomas R.",
""
],
[
"Vorberger",
"Jan",
""
],
[
"Dornheim",
"Tobias",
""
]
] |
2502.05628 | Houcheng Jiang | Houcheng Jiang, Junfeng Fang, Ningyu Zhang, Guojun Ma, Mingyang Wan,
Xiang Wang, Xiangnan He, Tat-seng Chua | AnyEdit: Edit Any Knowledge Encoded in Language Models | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) often produce incorrect or outdated information,
necessitating efficient and precise knowledge updates. Current model editing
methods, however, struggle with long-form knowledge in diverse formats, such as
poetry, code snippets, and mathematical derivations. These limitations arise
from their reliance on editing a single token's hidden state, a limitation we
term "efficacy barrier". To solve this, we propose AnyEdit, a new
autoregressive editing paradigm. It decomposes long-form knowledge into
sequential chunks and iteratively edits the key token in each chunk, ensuring
consistent and accurate outputs. Theoretically, we ground AnyEdit in the Chain
Rule of Mutual Information, showing its ability to update any knowledge within
LLMs. Empirically, it outperforms strong baselines by 21.5% on benchmarks
including UnKEBench, AKEW, and our new EditEverything dataset for long-form
diverse-formatted knowledge. Additionally, AnyEdit serves as a plug-and-play
framework, enabling current editing methods to update knowledge with arbitrary
length and format, significantly advancing the scope and practicality of LLM
knowledge editing.
| [
{
"version": "v1",
"created": "Sat, 8 Feb 2025 16:18:37 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 03:21:36 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Jiang",
"Houcheng",
""
],
[
"Fang",
"Junfeng",
""
],
[
"Zhang",
"Ningyu",
""
],
[
"Ma",
"Guojun",
""
],
[
"Wan",
"Mingyang",
""
],
[
"Wang",
"Xiang",
""
],
[
"He",
"Xiangnan",
""
],
[
"Chua",
"Tat-seng",
""
]
] |
2502.05655 | Drew Duffield | Drew Damien Duffield | Folded Gentle Algebras | 58 pages | null | null | null | math.RT math.CO math.RA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We use folding techniques to define a new class of gentle-like algebras that
generalise the iterated tilted algebras of type $C$ and $\widetilde{C}$, which
we call folded gentle algebras. We then show that folded gentle algebras
satisfy many of the same remarkable properties of gentle algebras, and that the
proof of these properties follows directly from folding arguments. In
particular, we classify the indecomposable modules of folded gentle algebras in
terms symmetric and asymmetric string and band modules. We classify the
Auslander-Reiten sequences over these algebras, showing that irreducible
morphisms between string modules are given by adding/deleting hooks and cohooks
to/from strings. Finally, we show that the class of folded gentle algebras are
closed under derived equivalence.
| [
{
"version": "v1",
"created": "Sat, 8 Feb 2025 18:05:41 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 08:41:14 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Duffield",
"Drew Damien",
""
]
] |
2502.05759 | Zherui Li | Zherui Li, Houcheng Jiang, Hao Chen, Baolong Bi, Zhenhong Zhou, Fei
Sun, Junfeng Fang, Xiang Wang | Reinforced Lifelong Editing for Language Models | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) acquire information from pre-training corpora,
but their stored knowledge can become inaccurate or outdated over time. Model
editing addresses this challenge by modifying model parameters without
retraining, and prevalent approaches leverage hypernetworks to generate these
parameter updates. However, they face significant challenges in lifelong
editing due to their incompatibility with LLM parameters that dynamically
change during the editing process. To address this, we observed that
hypernetwork-based lifelong editing aligns with reinforcement learning modeling
and proposed RLEdit, an RL-based editing method. By treating editing losses as
rewards and optimizing hypernetwork parameters at the full knowledge sequence
level, we enable it to precisely capture LLM changes and generate appropriate
parameter updates. Our extensive empirical evaluation across several LLMs
demonstrates that RLEdit outperforms existing methods in lifelong editing with
superior effectiveness and efficiency, achieving a 59.24% improvement while
requiring only 2.11% of the time compared to most approaches. Our code is
available at: https://github.com/zhrli324/RLEdit.
| [
{
"version": "v1",
"created": "Sun, 9 Feb 2025 03:37:06 GMT"
},
{
"version": "v2",
"created": "Tue, 18 Feb 2025 15:07:53 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 05:46:13 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Li",
"Zherui",
""
],
[
"Jiang",
"Houcheng",
""
],
[
"Chen",
"Hao",
""
],
[
"Bi",
"Baolong",
""
],
[
"Zhou",
"Zhenhong",
""
],
[
"Sun",
"Fei",
""
],
[
"Fang",
"Junfeng",
""
],
[
"Wang",
"Xiang",
""
]
] |
2502.05785 | Deping Guo | Deping Guo, Canbo Zong, Weihan Zhang, Cong Wang, Junwei Liu and Wei Ji | Tunable altermagnetism via inter-chain engineering in parallelassembled
atomic chains | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Altermagnetism has recently drawn considerable attention in three- and
twodimensional materials. Here, we extend this concept to quasi-one-dimensional
(Q1D) monolayers assembled from single-atomic magnetic chains. Through
systematically examining nine types of structures, two stacking orders, and
intra-/inter-chain magnetic couplings, we identify four out of thirty promising
structural prototypes for hosting altermagnetism, which yields 192 potential
monolayer materials. We further confirm eight thermodynamically stable Q1D
monolayers via high-throughput calculations. Using symmetry analysis and
first-principles calculations, we find that the existence of altermagnetism is
determined by the type of inter-chain magnetic coupling and predict three
intrinsic altermagnets,\(CrBr_3\),\(VBr_3\),\(MnBr_3\),due to their
ferromagnetic inter-chain couplings and five extrinsic
ones,\(CrF_3\),\(CrCl_3\),\(CrI_3\),\(FeCl_3\)and \(CoTe_3\), ascribed to their
neglectable or antiferromagnetic inter-chain couplings. Moreover, the
inter-chain magnetic coupling here is highly tunable by manipulating the
inter-chain spacing, leading to experimentally feasible transitions between
altermagnetic and nodal-line semiconducting states. In addition, applying
external electric fields can further modulate the spin splitting. Our findings
establish a highly tunable family of Q1D altermagnets, offering fundamental
insights into the intricate relationship between geometry, electronic
structure, and magnetism.These discoveries hold significant promises for
experimental realization and future spintronic applications.
| [
{
"version": "v1",
"created": "Sun, 9 Feb 2025 06:01:18 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 15:34:39 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Guo",
"Deping",
""
],
[
"Zong",
"Canbo",
""
],
[
"Zhang",
"Weihan",
""
],
[
"Wang",
"Cong",
""
],
[
"Liu",
"Junwei",
""
],
[
"Ji",
"Wei",
""
]
] |
2502.06352 | Sihwan Park | Sihwan Park, Doohyuk Jang, Sungyub Kim, Souvik Kundu, Eunho Yang | LANTERN++: Enhancing Relaxed Speculative Decoding with Static Tree
Drafting for Visual Auto-regressive Models | ICLR 2025 Workshop at SCOPE (Oral), 16 pages, 5 figures, short paper
(6 pages exclude reference and appendix) | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Speculative decoding has been widely used to accelerate auto-regressive (AR)
text generation. However, its effectiveness for visual AR models remains
limited due to token selection ambiguity, where multiple tokens share similarly
low probabilities and thus reduce acceptance rates. Recently, relaxed
speculative decoding with dynamic tree drafting was proposed to mitigate this
ambiguity, demonstrating promising results in accelerating visual AR models.
However, we observe that token selection ambiguity still negatively affects
dynamic tree drafting, resulting in shallow draft trees and limited
acceleration. To overcome this issue, we introduce LANTERN++, a refined
framework that integrates static tree drafting with a tailored relaxed
acceptance condition, allowing drafts to be selected independently of
low-confidence predictions. This enables the acceptance of deeper sequences,
improving decoding efficiency while preserving image quality. Extensive
experiments on state-of-the-art visual AR models demonstrate that LANTERN++
significantly accelerates inference, achieving up to $\mathbf{\times 2.56}$
speedup over standard AR decoding while maintaining high image quality. The
code is publicly available at https://github.com/jadohu/LANTERN.
| [
{
"version": "v1",
"created": "Mon, 10 Feb 2025 11:05:18 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 11:53:23 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Park",
"Sihwan",
""
],
[
"Jang",
"Doohyuk",
""
],
[
"Kim",
"Sungyub",
""
],
[
"Kundu",
"Souvik",
""
],
[
"Yang",
"Eunho",
""
]
] |
2502.06386 | Alexander Ebenbichler | A. Ebenbichler, M. On\v{c}\'ak, N. Przybilla, H. R. Hrodmarsson, J. V.
Smoker, R. Lallement, A. Farhang, C. Bhatt, J. Cami, M. Cordiner, P.
Ehrenfreund, N. L. J. Cox, J. Th. van Loon, B. Foing | The EDIBLES survey. X. The 6196 {\AA} diffuse interstellar band:
Identification of side DIBs as an indication of a small carrier molecule | 13 pages, 12 figures, Accepted for publication in Astronomy &
Astrophysics | A&A 695, A212 (2025) | 10.1051/0004-6361/202453259 | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Context: Numerous studies of diffuse interstellar band (DIB) profiles have
detected substructures, which in turn suggests that large molecules are acting
as their carriers. However, some of the narrowest DIBs generally do not show
such substructures, suggesting the possibility of very small carriers. Aims:
Based on the previously found tight correlation of the three narrow DIBs at
6196, 6440 and 6623 A and the present detection of weaker side DIBs to each of
them in the extensive data set from the ESO Diffuse Interstellar Bands Large
Exploration Survey, we investigated whether they may stem from small linear
carrier molecules. This approach can lead to concrete DIB carrier suggestions,
which can be tested in laboratory measurements in future studies. Methods: We
suggest that the DIBs we studied here represent individual rotational
transitions of a small molecule. We determined the molecular constants from
observations and compared them with data from a large set of quantum-chemical
calculations to constrain possible carrier candidates. Furthermore, we
determined the rotational temperatures by fitting line ratios using the fitted
molecular models. Results: We determined molecular constants for three DIB
systems and the corresponding transition types. The fitted rotational
temperatures lie within the range of known interstellar diatomic molecules. We
identified several DIB carrier candidates, almost all of them molecular ions.
Some of them are metastable species, indicating the possibility of collision
complexes as DIB carriers. Conclusions: If our hypothesis holds, this would be
a major step towards the identification of a carrier molecule of the 6196 A
DIB, the strongest among the narrow DIBs.
| [
{
"version": "v1",
"created": "Mon, 10 Feb 2025 12:14:38 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:31:40 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ebenbichler",
"A.",
""
],
[
"Ončák",
"M.",
""
],
[
"Przybilla",
"N.",
""
],
[
"Hrodmarsson",
"H. R.",
""
],
[
"Smoker",
"J. V.",
""
],
[
"Lallement",
"R.",
""
],
[
"Farhang",
"A.",
""
],
[
"Bhatt",
"C.",
""
],
[
"Cami",
"J.",
""
],
[
"Cordiner",
"M.",
""
],
[
"Ehrenfreund",
"P.",
""
],
[
"Cox",
"N. L. J.",
""
],
[
"van Loon",
"J. Th.",
""
],
[
"Foing",
"B.",
""
]
] |
2502.06608 | Yangguang Li | Yangguang Li, Zi-Xin Zou, Zexiang Liu, Dehu Wang, Yuan Liang, Zhipeng
Yu, Xingchao Liu, Yuan-Chen Guo, Ding Liang, Wanli Ouyang, Yan-Pei Cao | TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified
Flow Models | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advancements in diffusion techniques have propelled image and video
generation to unprecedented levels of quality, significantly accelerating the
deployment and application of generative AI. However, 3D shape generation
technology has so far lagged behind, constrained by limitations in 3D data
scale, complexity of 3D data processing, and insufficient exploration of
advanced techniques in the 3D domain. Current approaches to 3D shape generation
face substantial challenges in terms of output quality, generalization
capability, and alignment with input conditions. We present TripoSG, a new
streamlined shape diffusion paradigm capable of generating high-fidelity 3D
meshes with precise correspondence to input images. Specifically, we propose:
1) A large-scale rectified flow transformer for 3D shape generation, achieving
state-of-the-art fidelity through training on extensive, high-quality data. 2)
A hybrid supervised training strategy combining SDF, normal, and eikonal losses
for 3D VAE, achieving high-quality 3D reconstruction performance. 3) A data
processing pipeline to generate 2 million high-quality 3D samples, highlighting
the crucial rules for data quality and quantity in training 3D generative
models. Through comprehensive experiments, we have validated the effectiveness
of each component in our new framework. The seamless integration of these parts
has enabled TripoSG to achieve state-of-the-art performance in 3D shape
generation. The resulting 3D shapes exhibit enhanced detail due to
high-resolution capabilities and demonstrate exceptional fidelity to input
images. Moreover, TripoSG demonstrates improved versatility in generating 3D
models from diverse image styles and contents, showcasing strong generalization
capabilities. To foster progress and innovation in the field of 3D generation,
we will make our model publicly available.
| [
{
"version": "v1",
"created": "Mon, 10 Feb 2025 16:07:54 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Feb 2025 06:03:40 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 17:25:50 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Li",
"Yangguang",
""
],
[
"Zou",
"Zi-Xin",
""
],
[
"Liu",
"Zexiang",
""
],
[
"Wang",
"Dehu",
""
],
[
"Liang",
"Yuan",
""
],
[
"Yu",
"Zhipeng",
""
],
[
"Liu",
"Xingchao",
""
],
[
"Guo",
"Yuan-Chen",
""
],
[
"Liang",
"Ding",
""
],
[
"Ouyang",
"Wanli",
""
],
[
"Cao",
"Yan-Pei",
""
]
] |
2502.06818 | Jingyun Wang | Jingyun Wang, Cilin Yan, Guoliang Kang | Rethinking the Global Knowledge of CLIP in Training-Free Open-Vocabulary
Semantic Segmentation | Under review | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent works modify CLIP to perform open-vocabulary semantic segmentation in
a training-free manner (TF-OVSS). In vanilla CLIP, patch-wise image
representations mainly encode homogeneous image-level properties, which hinders
the application of CLIP to the dense prediction task. Previous TF-OVSS works
sacrifice globality to enhance the locality of CLIP features, by making each
patch mainly attend to itself or its neighboring patches within a narrow local
window. With their modifications,the ability of CLIP to aggregate global
context information is largely weakened. Differently, in this paper, we rethink
the global knowledge encoded by CLIP and propose GCLIP to answer how to extract
and utilize beneficial global knowledge of CLIP for TF-OVSS. As the
representation of each patch is finally determined by the attention weights and
the Value embeddings, we propose to reshape the last-block attention and Value
embeddings to aggregate useful global context into final features. Firstly, we
aim to equip the last-block attention with image-level properties while not
introducing homogeneous attention patterns across patches. To realize the goal,
we fuse the attention from the global-token emerging blocks with the
Query-Query attention. Secondly, we aim to make Value embeddings of the
last-block attention module more semantically correlated. To realize this, we
design a novel channel suppression strategy.Extensive experiments on five
standard benchmarks demonstrate that our method consistently outperforms
previous state-of-the-arts.
| [
{
"version": "v1",
"created": "Wed, 5 Feb 2025 03:37:50 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 01:18:44 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wang",
"Jingyun",
""
],
[
"Yan",
"Cilin",
""
],
[
"Kang",
"Guoliang",
""
]
] |
2502.06874 | Yanming Guo | Yanming Guo, Xiao Qian, Kevin Credit, Jin Ma | Group Reasoning Emission Estimation Networks | null | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Accurate greenhouse gas (GHG) emission reporting is critical for governments,
businesses, and investors. However, adoption remains limited particularly among
small and medium enterprises due to high implementation costs, fragmented
emission factor databases, and a lack of robust sector classification methods.
To address these challenges, we introduce Group Reasoning Emission Estimation
Networks (GREEN), an AI-driven carbon accounting framework that standardizes
enterprise-level emission estimation, constructs a large-scale benchmark
dataset, and leverages a novel reasoning approach with large language models
(LLMs). Specifically, we compile textual descriptions for 20,850 companies with
validated North American Industry Classification System (NAICS) labels and
align these with an economic model of carbon intensity factors. By reframing
sector classification as an information retrieval task, we fine-tune
Sentence-BERT models using a contrastive learning loss. To overcome the
limitations of single-stage models in handling thousands of hierarchical
categories, we propose a Group Reasoning method that ensembles LLM classifiers
based on the natural NAICS ontology, decomposing the task into multiple
sub-classification steps. We theoretically prove that this approach reduces
classification uncertainty and computational complexity. Experiments on 1,114
NAICS categories yield state-of-the-art performance (83.68% Top-1, 91.47%
Top-10 accuracy), and case studies on 20 companies report a mean absolute
percentage error (MAPE) of 45.88%. The project is available at:
https://huggingface.co/datasets/Yvnminc/ExioNAICS.
| [
{
"version": "v1",
"created": "Sat, 8 Feb 2025 09:02:43 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:37:40 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Guo",
"Yanming",
""
],
[
"Qian",
"Xiao",
""
],
[
"Credit",
"Kevin",
""
],
[
"Ma",
"Jin",
""
]
] |
2502.06942 | Hao Fu | Hao Fu, Lumen Boco, Francesco Shankar, Andrea Lapi, Mohammadreza
Ayromlou, Daniel Roberts, Yingjie Peng, Aldo Rodr\'iguez-Puebla, Feng Yuan,
Cressida Cleland, Simona Mei, Nicola Menci | Shedding light on the star formation rate-halo accretion rate connection
and halo quenching mechanism via DECODE, the Discrete statistical
sEmi-empiriCal mODEl | 14 pages, 13 figures, Published in Astronomy & Astrophysics | A&A 695, A252 (2025) | 10.1051/0004-6361/202453218 | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Aims: The relative roles of the physical mechanisms involved in quenching
galaxy star formation are still unclear. We tackle this fundamental problem
with our cosmological semi-empirical model DECODE (Discrete statistical
sEmi-empiriCal mODEl), designed to predict galaxy stellar mass assembly
histories, from minimal input assumptions. Methods: Specifically, in this work
the star formation history of each galaxy is calculated along its progenitor
dark matter halo by assigning at each redshift a star formation rate extracted
from a monotonic star formation rate-halo accretion rate (SFR-HAR) relation
derived from abundance matching between the (observed) SFR function and the
(numerically predicted) HAR function, a relation that is also predicted by the
TNG100 simulation. SFRs are integrated across cosmic time to build up the mass
of galaxies, which may halt their star formation following input physical
quenching recipes. Results: In this work we test the popular halo quenching
scenario and we find that: 1) the assumption of a monotonic relation between
SFR and HAR allows to reproduce the number densities of the bulk of
star-forming galaxies in the local Universe; 2) the halo quenching is
sufficient to reproduce the statistics of the quenched galaxies and flat
(steep) high-mass end of the SMHM relation (SMF); and 3) to align with the
observed steep (flat) low-mass end of the SMHM (SMF) additional quenching
processes in the least massive haloes are needed. Conclusions: DECODE is an
invaluable tool and will pave the way to investigate the origin of newly
observed high-redshift objects from the latest ongoing facilities such as JWST
and Euclid.
| [
{
"version": "v1",
"created": "Mon, 10 Feb 2025 19:00:01 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 08:12:28 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Fu",
"Hao",
""
],
[
"Boco",
"Lumen",
""
],
[
"Shankar",
"Francesco",
""
],
[
"Lapi",
"Andrea",
""
],
[
"Ayromlou",
"Mohammadreza",
""
],
[
"Roberts",
"Daniel",
""
],
[
"Peng",
"Yingjie",
""
],
[
"Rodríguez-Puebla",
"Aldo",
""
],
[
"Yuan",
"Feng",
""
],
[
"Cleland",
"Cressida",
""
],
[
"Mei",
"Simona",
""
],
[
"Menci",
"Nicola",
""
]
] |
2502.06955 | Howard Baer | Howard Baer, Vernon Barger, Dibyashree Sengupta and Kairui Zhang | All axion dark matter from supersymmetric models | 11 pages plus 1 .png figure; version 2 contains some added text and
improved figure | null | null | OU-HEP-250204 | hep-ph hep-th | http://creativecommons.org/licenses/by/4.0/ | Supersymmetric models accompanied by certain anomaly-free discrete
R-symmetries Z_n^R are attractive in that 1. the R-symmetry (which can arise
from compactified string theory as a remnant of the broken 10-d Lorentz
symmetry) forbids unwanted superpotential terms while allowing for the
generation of an accidental, approximate global U(1)_{PQ} symmetry needed to
solve the strong CP problem and 2. they provide a raison d'etre for an
otherwise ad-hoc R-parity conservation. We augment the minimal supersymmetric
Standard Model (MSSM) by two additional Z_n^R- and PQ-charged fields X and Y
wherein SUSY breaking at an intermediate scale m_{hidden} leads to PQ breaking
at a scale f_a\sim 10^{11} GeV leading to a SUSY DFSZ axion. The same SUSY
breaking can trigger R-parity breaking via higher-dimensional operators leading
to tiny R-violating couplings of order (f_a/m_P)^N and a WIMP quality problem.
For Z_4^R and Z_8^R, we find only an N=1 suppression. Then the lightest SUSY
particle (LSP) of the MSSM becomes unstable with a lifetime of order ~
10^{-3}-10 seconds so the LSPs all decay away before the present epoch. That
leaves a universe with all axion cold dark matter and no WIMPs in accord with
recent LZ-2024 WIMP search results.
| [
{
"version": "v1",
"created": "Mon, 10 Feb 2025 19:00:23 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 16:40:43 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Baer",
"Howard",
""
],
[
"Barger",
"Vernon",
""
],
[
"Sengupta",
"Dibyashree",
""
],
[
"Zhang",
"Kairui",
""
]
] |
2502.08180 | Zhen Xiong | Zhen Xiong, Yujun Cai, Bryan Hooi, Nanyun Peng, Zhecheng Li, Yiwei
Wang | Enhancing LLM Character-Level Manipulation via Divide and Conquer | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have demonstrated strong generalization
capabilities across a wide range of natural language processing (NLP) tasks.
However, they exhibit notable weaknesses in character-level string
manipulation, struggling with fundamental operations such as character
deletion, insertion, and substitution. These challenges stem primarily from
tokenization constraints, despite the critical role of such operations in data
preprocessing and code generation. Through systematic analysis, we derive two
key insights: (1) LLMs face significant difficulties in leveraging intrinsic
token knowledge for character-level reasoning, and (2) atomized word structures
can substantially enhance LLMs' ability to process token-level structural
information. Building on these insights, we propose Character-Level
Manipulation via Divide and Conquer, a novel approach designed to bridge the
gap between token-level processing and character-level manipulation. Our method
decomposes complex operations into explicit character-level subtasks coupled
with controlled token reconstruction phases, leading to significant
improvements in accuracy. Without additional training, our method significantly
improves accuracies on the $\texttt{Deletion}$, $\texttt{Insertion}$, and
$\texttt{Substitution}$ tasks. To support further research, we open-source our
implementation and benchmarks.
| [
{
"version": "v1",
"created": "Wed, 12 Feb 2025 07:37:39 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 16:07:18 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Xiong",
"Zhen",
""
],
[
"Cai",
"Yujun",
""
],
[
"Hooi",
"Bryan",
""
],
[
"Peng",
"Nanyun",
""
],
[
"Li",
"Zhecheng",
""
],
[
"Wang",
"Yiwei",
""
]
] |
2502.08356 | Kushagra Bhushan Mr. | Kushagra Bhushan, Yatin Nandwani, Dinesh Khandelwal, Sonam Gupta,
Gaurav Pandey, Dinesh Raghu, Sachindra Joshi | Systematic Knowledge Injection into Large Language Models via Diverse
Augmentation for Domain-Specific RAG | 22 pages, 14 tables, to be published in NAACL 2025 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Retrieval-Augmented Generation (RAG) has emerged as a prominent method for
incorporating domain knowledge into Large Language Models (LLMs). While RAG
enhances response relevance by incorporating retrieved domain knowledge in the
context, retrieval errors can still lead to hallucinations and incorrect
answers. To recover from retriever failures, domain knowledge is injected by
fine-tuning the model to generate the correct response, even in the case of
retrieval errors. However, we observe that without systematic knowledge
augmentation, fine-tuned LLMs may memorize new information but still fail to
extract relevant domain knowledge, leading to poor performance. In this work,
we present a novel framework that significantly enhances the fine-tuning
process by augmenting the training data in two ways -- context augmentation and
knowledge paraphrasing. In context augmentation, we create multiple training
samples for a given QA pair by varying the relevance of the retrieved
information, teaching the model when to ignore and when to rely on retrieved
content. In knowledge paraphrasing, we fine-tune with multiple answers to the
same question, enabling LLMs to better internalize specialized knowledge. To
mitigate catastrophic forgetting due to fine-tuning, we add a domain-specific
identifier to a question and also utilize a replay buffer containing general QA
pairs. Experimental results demonstrate the efficacy of our method over
existing techniques, achieving up to 10\% relative gain in token-level recall
while preserving the LLM's generalization capabilities.
| [
{
"version": "v1",
"created": "Wed, 12 Feb 2025 12:39:51 GMT"
},
{
"version": "v2",
"created": "Mon, 17 Feb 2025 14:29:48 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 11:31:39 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Bhushan",
"Kushagra",
""
],
[
"Nandwani",
"Yatin",
""
],
[
"Khandelwal",
"Dinesh",
""
],
[
"Gupta",
"Sonam",
""
],
[
"Pandey",
"Gaurav",
""
],
[
"Raghu",
"Dinesh",
""
],
[
"Joshi",
"Sachindra",
""
]
] |
2502.08745 | Zhihan Zhang | Zhihan Zhang, Shiyang Li, Zixuan Zhang, Xin Liu, Haoming Jiang,
Xianfeng Tang, Yifan Gao, Zheng Li, Haodong Wang, Zhaoxuan Tan, Yichuan Li,
Qingyu Yin, Bing Yin, Meng Jiang | IHEval: Evaluating Language Models on Following the Instruction
Hierarchy | Accepted to NAACL 2025 for oral presentation. Our project page is
located at https://ytyz1307zzh.github.io/iheval.github.io | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The instruction hierarchy, which establishes a priority order from system
messages to user messages, conversation history, and tool outputs, is essential
for ensuring consistent and safe behavior in language models (LMs). Despite its
importance, this topic receives limited attention, and there is a lack of
comprehensive benchmarks for evaluating models' ability to follow the
instruction hierarchy. We bridge this gap by introducing IHEval, a novel
benchmark comprising 3,538 examples across nine tasks, covering cases where
instructions in different priorities either align or conflict. Our evaluation
of popular LMs highlights their struggle to recognize instruction priorities.
All evaluated models experience a sharp performance decline when facing
conflicting instructions, compared to their original instruction-following
performance. Moreover, the most competitive open-source model only achieves 48%
accuracy in resolving such conflicts. Our results underscore the need for
targeted optimization in the future development of LMs.
| [
{
"version": "v1",
"created": "Wed, 12 Feb 2025 19:35:28 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 20:35:30 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhang",
"Zhihan",
""
],
[
"Li",
"Shiyang",
""
],
[
"Zhang",
"Zixuan",
""
],
[
"Liu",
"Xin",
""
],
[
"Jiang",
"Haoming",
""
],
[
"Tang",
"Xianfeng",
""
],
[
"Gao",
"Yifan",
""
],
[
"Li",
"Zheng",
""
],
[
"Wang",
"Haodong",
""
],
[
"Tan",
"Zhaoxuan",
""
],
[
"Li",
"Yichuan",
""
],
[
"Yin",
"Qingyu",
""
],
[
"Yin",
"Bing",
""
],
[
"Jiang",
"Meng",
""
]
] |
2502.08879 | Kai-Chi Chang | Kai-Chi Chang, Xiang Cheng, Murat Can Sarihan, and Chee Wei Wong | Recent advances in high-dimensional mode-locked quantum frequency combs | 45 pages, 8 figures | Newton 1, 100024 (2025) | 10.1016/j.newton.2025.100024 | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | High-dimensional entanglement in qudit states offers a promising pathway
towards the realization of practical, large-scale quantum systems that are
highly controllable. These systems can be leveraged for various applications,
including advanced quantum information processing, secure communications,
computation, and metrology. In this context, quantum frequency combs have a
crucial role as they inherently support multiple modes in both temporal and
frequency domains, while preserving a single spatial mode. The multiple
temporal and frequency modes of quantum frequency combs facilitate the
generation, characterization, and control of high-dimensional time-frequency
entanglement in extensive quantum systems. In this review article, we provide
an overview of recent technological advancements in high-dimensional
energy-time entangled quantum frequency combs. We explore how these
time-frequency qudits, achieved using scalable telecommunications-wavelength
components, can empower the creation of large-scale quantum states. Advances in
quantum frequency combs can unlock new capabilities and versatility for
promising developments in quantum science and technology.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2025 01:33:58 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 05:40:36 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chang",
"Kai-Chi",
""
],
[
"Cheng",
"Xiang",
""
],
[
"Sarihan",
"Murat Can",
""
],
[
"Wong",
"Chee Wei",
""
]
] |
2502.08972 | Hyundong Cho | Hyundong Cho, Karishma Sharma, Nicolaas Jedema, Leonardo F. R.
Ribeiro, Alessandro Moschitti, Ravi Krishnan, Jonathan May | Tuning-Free Personalized Alignment via Trial-Error-Explain In-Context
Learning | NAACL 2025 Findings | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Language models are aligned to the collective voice of many, resulting in
generic outputs that do not align with specific users' styles. In this work, we
present Trial-Error-Explain In-Context Learning} (ITCL), a tuning-free method
that personalizes language models for text generation tasks with fewer than 10
examples per user. TICL iteratively expands an in-context learning prompt via a
trial-error-explain process, adding model-generated negative samples and
explanations that provide fine-grained guidance towards a specific user's
style. TICL achieves favorable win rates on pairwise comparisons with
LLM-as-a-judge up to 91.5% against the previous state-of-the-art and
outperforms competitive tuning-free baselines for personalized alignment tasks
of writing emails, essays and news articles. Both lexical and qualitative
analyses show that the negative samples and explanations enable language models
to learn stylistic context more effectively and overcome the bias towards
structural and formal phrases observed in their zero-shot outputs. By
front-loading inference compute to create a user-specific in-context learning
prompt that does not require extra generation steps at test time, TICL presents
a novel yet simple approach for personalized alignment.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2025 05:20:21 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 07:21:27 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Cho",
"Hyundong",
""
],
[
"Sharma",
"Karishma",
""
],
[
"Jedema",
"Nicolaas",
""
],
[
"Ribeiro",
"Leonardo F. R.",
""
],
[
"Moschitti",
"Alessandro",
""
],
[
"Krishnan",
"Ravi",
""
],
[
"May",
"Jonathan",
""
]
] |
2502.09042 | Pittawat Taveekitworachai | Pittawat Taveekitworachai, Potsawee Manakul, Kasima Tharnpipitchai,
Kunat Pipatanakul | Typhoon T1: An Open Thai Reasoning Model | 25 pages, 6 figures | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper introduces Typhoon T1, an open effort to develop an open Thai
reasoning model. A reasoning model is a relatively new type of generative model
built on top of large language models (LLMs). A reasoning model generates a
long chain of thought before arriving at a final answer, an approach found to
improve performance on complex tasks. However, details on developing such a
model are limited, especially for reasoning models that can generate traces in
a low-resource language. Typhoon T1 presents an open effort that dives into the
details of developing a reasoning model in a more cost-effective way by
leveraging supervised fine-tuning using open datasets, instead of reinforcement
learning. This paper shares the details about synthetic data generation and
training, as well as our dataset and model weights. Additionally, we provide
insights gained from developing a reasoning model that generalizes across
domains and is capable of generating reasoning traces in a low-resource
language, using Thai as an example. We hope this open effort provides a
foundation for further research in this field.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2025 07:55:54 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:45:15 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Taveekitworachai",
"Pittawat",
""
],
[
"Manakul",
"Potsawee",
""
],
[
"Tharnpipitchai",
"Kasima",
""
],
[
"Pipatanakul",
"Kunat",
""
]
] |
2502.09056 | Kunat Pipatanakul | Kunat Pipatanakul, Pittawat Taveekitworachai, Potsawee Manakul, Kasima
Tharnpipitchai | Adapting Language-Specific LLMs to a Reasoning Model in One Day via
Model Merging -- An Open Recipe | 9 pages | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper investigates data selection and model merging methodologies aimed
at incorporating advanced reasoning capabilities such as those of DeepSeek R1
into language-specific large language models (LLMs), with a particular focus on
the Thai LLM. Our goal is to enhance the reasoning capabilities of
language-specific LLMs while maintaining their target language abilities.
DeepSeek R1 excels in reasoning but primarily benefits high-resource languages
such as English and Chinese. However, low-resource languages remain underserved
due to the dominance of English-centric training data and model optimizations,
which limit performance in these languages. This limitation results in
unreliable code-switching and diminished effectiveness on tasks in low-resource
languages. Meanwhile, local and regional LLM initiatives have attempted to
bridge this gap by developing language-specific LLMs that focus on improving
local linguistic fidelity. We demonstrate that, with only publicly available
datasets and a computational budget of $120, it is possible to enhance the
reasoning capabilities of language-specific LLMs to match the level of DeepSeek
R1, without compromising their performance on target language tasks.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2025 08:10:45 GMT"
},
{
"version": "v2",
"created": "Mon, 17 Feb 2025 13:16:00 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 06:45:16 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Pipatanakul",
"Kunat",
""
],
[
"Taveekitworachai",
"Pittawat",
""
],
[
"Manakul",
"Potsawee",
""
],
[
"Tharnpipitchai",
"Kasima",
""
]
] |
2502.09514 | Ansgar Burchards | Ansgar G. Burchards | Continuous-Variable Quantum MacWilliams Identities | 26 pages (20 main + 6 appendices), 4 figures | null | null | null | quant-ph math.MG | http://creativecommons.org/licenses/by/4.0/ | We derive bounds on general quantum error correcting codes against the
displacement noise channel. The bounds limit the distances attainable by codes
and also apply in an approximate setting. Our main result is a quantum analogue
of the classical Cohn-Elkies bound on sphere packing densities attainable in
Euclidean space. We further derive a quantum version of Levenshtein's sphere
packing bound and argue that Gottesman--Kitaev--Preskill (GKP) codes based on
the $E_8$ and Leech lattices achieve optimal distances. The main technical tool
is a continuous-variable version of the quantum MacWilliams identities, which
we introduce. The identities relate a pair of weight distributions which can be
obtained for any two trace-class operators. General properties of these weight
distributions are discussed, along with several examples.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2025 17:30:22 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 12:58:13 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Burchards",
"Ansgar G.",
""
]
] |
2502.09535 | Carlton Shepherd | Carlton Shepherd and Elliot Hurley | Entropy Collapse in Mobile Sensors: The Hidden Risks of Sensor-Based
Security | null | null | null | null | cs.CR | http://creativecommons.org/licenses/by/4.0/ | Mobile sensor data has been proposed for security-critical applications such
as device pairing, proximity detection, and continuous authentication. However,
the foundational assumption that these signals provide sufficient entropy
remains under-explored. In this work, we systematically analyse the entropy of
mobile sensor data across four diverse datasets spanning multiple application
contexts. Our findings reveal pervasive biases, with single-sensor mean
min-entropy values ranging from 3.408-4.483 bits (S.D.=1.018-1.574) despite
Shannon entropy being several multiples higher. We further demonstrate that
correlations between sensor modalities reduce the worst-case entropy of using
multiple sensors by up to approx. 75% compared to average-case Shannon entropy.
This brings joint min-entropy well below 10 bits in many cases and, in the best
case, yielding only approx. 24 bits of min-entropy when combining 20 sensor
modalities. These results call into question the widely held assumption that
adding more sensors inherently yields higher security. We ultimately caution
against relying on raw sensor data as a primary source of randomness.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2025 17:50:58 GMT"
},
{
"version": "v2",
"created": "Fri, 14 Feb 2025 10:43:47 GMT"
},
{
"version": "v3",
"created": "Mon, 17 Feb 2025 22:41:20 GMT"
},
{
"version": "v4",
"created": "Tue, 25 Mar 2025 11:42:52 GMT"
},
{
"version": "v5",
"created": "Wed, 26 Mar 2025 22:14:50 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Shepherd",
"Carlton",
""
],
[
"Hurley",
"Elliot",
""
]
] |
2502.09698 | Elias Zapusek | Elias Zapusek, Kristina Kirova, Walter Hahn, Michael Marthaler,
Florentin Reiter | Variational quantum thermalizers based on weakly-symmetric nonunitary
multi-qubit operations | 19 pages, 10 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose incorporating multi-qubit nonunitary operations in Variational
Quantum Thermalizers (VQTs). VQTs are hybrid quantum-classical algorithms that
generate the thermal (Gibbs) state of a given Hamiltonian, with applications in
quantum algorithms and simulations. However, current algorithms struggle at
intermediate temperatures, where the target state is nonpure but exhibits
entanglement. We devise multi-qubit nonunitary operations that harness weak
symmetries and thereby improve the performance of the algorithm. Utilizing
dissipation engineering, we create these nonunitary multi-qubit operations
without the need for measurements or additional qubits. To train the ansatz, we
develop and benchmark novel methods for entropy estimation of quantum states,
expanding the toolbox for quantum state characterization. We demonstrate that
our approach can prepare thermal states of paradigmatic spin models at all
temperatures. Our work thus creates new opportunities for simulating open
quantum many-body systems.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2025 19:00:00 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:23:15 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zapusek",
"Elias",
""
],
[
"Kirova",
"Kristina",
""
],
[
"Hahn",
"Walter",
""
],
[
"Marthaler",
"Michael",
""
],
[
"Reiter",
"Florentin",
""
]
] |
2502.09972 | Shiba Behera | S. P. Behera | Sensitivity study of a sapphire detector using Coherent Elastic
Neutrino-Nucleus Scattering process | 11 pages, 7 figures, Published version | Physical Review D 111, 053007 (2025) | 10.1103/PhysRevD.111.053007 | null | hep-ph hep-ex nucl-ex nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Indian Coherent Neutrino-nucleus Scattering Experiment(ICNSE) has been
proposed at Bhabha Atomic Research Centre in India to measure the coherent
elastic neutrino-nucleus scattering process using electron antineutrinos
produced from reactors. Phenomenological studies are performed to find out the
sensitivity of a sapphire detector for various fundamental physics parameters
at an exposure of one year. Reactors of different core compositions, sizes, and
thermal powers have been considered as sources of electron antineutrinos. The
potential of the ICNSE to measure the weak mixing angle at a low energy regime
has been extracted. Furthermore, the detector's capability has been
investigated for examining the electromagnetic properties of neutrinos,
including their magnetic moment. Additionally, an exploration has been
conducted on the detector's sensitivity in restricting new interactions between
neutrinos and electrons or nuclei, thereby constraining the parameter space
related to light mediators. It is found that the ICNSE detector can put a
stronger constraints on the scalar and vector mediators masses.
| [
{
"version": "v1",
"created": "Fri, 14 Feb 2025 07:58:53 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Feb 2025 04:15:51 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 07:19:46 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Behera",
"S. P.",
""
]
] |
2502.10581 | Zeyu Jia | Zeyu Jia, Alexander Rakhlin, Tengyang Xie | Do We Need to Verify Step by Step? Rethinking Process Supervision from a
Theoretical Perspective | null | null | null | null | cs.LG cs.AI stat.ML | http://creativecommons.org/licenses/by/4.0/ | As large language models have evolved, it has become crucial to distinguish
between process supervision and outcome supervision -- two key reinforcement
learning approaches to complex reasoning tasks. While process supervision
offers intuitive advantages for long-term credit assignment, the precise
relationship between these paradigms has remained an open question.
Conventional wisdom suggests that outcome supervision is fundamentally more
challenging due to the trajectory-level coverage problem, leading to
significant investment in collecting fine-grained process supervision data.
In this paper, we take steps towards resolving this debate. Our main theorem
shows that, under standard data coverage assumptions, reinforcement learning
through outcome supervision is no more statistically difficult than through
process supervision, up to polynomial factors in horizon. At the core of this
result lies the novel Change of Trajectory Measure Lemma -- a technical tool
that bridges return-based trajectory measure and step-level distribution shift.
Furthermore, for settings with access to a verifier or a rollout capability, we
prove that any policy's advantage function can serve as an optimal process
reward model, providing a direct connection between outcome and process
supervision. These findings suggest that the empirically observed performance
gap -- if any -- between outcome and process supervision likely stems from
algorithmic limitations rather than inherent statistical difficulties,
potentially transforming how we approach data collection and algorithm design
for reinforcement learning.
| [
{
"version": "v1",
"created": "Fri, 14 Feb 2025 22:21:56 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 22:45:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Jia",
"Zeyu",
""
],
[
"Rakhlin",
"Alexander",
""
],
[
"Xie",
"Tengyang",
""
]
] |
2502.11011 | Howard E. Haber | Howard E. Haber and P. M. Ferreira | RG-stable parameter relations of a scalar field theory in absence of a
symmetry | 29 pages. Version 2 adds a new appendix and includes two new
footnotes and two additional references | null | null | null | hep-ph hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The stability of tree-level relations among the parameters of a quantum field
theory with respect to renormalization group (RG) running is typically
explained by the existence of a symmetry. We examine a toy model of a quantum
field theory of two real scalars in which a tree-level relation among the
squared-mass parameters of the scalar potential appears to be RG-stable without
the presence of an appropriate underlying symmetry. The stability of this
relation with respect to renormalization group running can be explained by
complexifying the original scalar field theory. It is then possible to exhibit
a symmetry that guarantees the relations of relevant beta functions of
squared-mass parameters of the complexified theory. Among these relations, we
can identify equations that are algebraically identical to the corresponding
equations that guarantee the stability of the relations among the squared-mass
parameters of the original real scalar field theory where the symmetry of the
complexified theory is no longer present.
| [
{
"version": "v1",
"created": "Sun, 16 Feb 2025 06:33:21 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:02:06 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Haber",
"Howard E.",
""
],
[
"Ferreira",
"P. M.",
""
]
] |
2502.11316 | Woojoo Lee | Seonghyun Choi, Kyeongwon Lee, Jae-Jin Lee, Woojoo Lee | Standalone FPGA-Based QAOA Emulator for Weighted-MaxCut on Embedded
Devices | 9 pages, 6 figures, 3 tables | null | null | null | cs.ET quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum computing QC emulation is crucial for advancing QC applications,
especially given the scalability constraints of current devices. FPGA-based
designs offer an efficient and scalable alternative to traditional large-scale
platforms, but most are tightly integrated with high-performance systems,
limiting their use in mobile and edge environments. This study introduces a
compact, standalone FPGA-based QC emulator designed for embedded systems,
leveraging the Quantum Approximate Optimization Algorithm (QAOA) to solve the
Weighted-MaxCut problem. By restructuring QAOA operations for hardware
compatibility, the proposed design reduces time complexity from O(N^2) to O(N),
where N equals 2^n for n qubits. This reduction, coupled with a pipeline
architecture, significantly minimizes resource consumption, enabling support
for up to nine qubits on mid-tier FPGAs, roughly three times more than
comparable designs. Additionally, the emulator achieved energy savings ranging
from 1.53 times for two-qubit configurations to up to 852 times for nine-qubit
configurations, compared to software-based QAOA on embedded processors. These
results highlight the practical scalability and resource efficiency of the
proposed design, providing a robust foundation for QC emulation in
resource-constrained edge devices.
| [
{
"version": "v1",
"created": "Sun, 16 Feb 2025 23:30:16 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 06:17:28 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Choi",
"Seonghyun",
""
],
[
"Lee",
"Kyeongwon",
""
],
[
"Lee",
"Jae-Jin",
""
],
[
"Lee",
"Woojoo",
""
]
] |
2502.11503 | Mahmoud Benkhalifa | Mahmoud Benkhalifa | Self-Homotopy Equivalence Group of an Elliptic Space and Its Embedding
in general Linear Groups | null | null | null | null | math.AT | http://creativecommons.org/licenses/by/4.0/ | For a rational elliptic space, this paper examines the relationship between
its homotopy groups and its self-homotopy equivalence group. Moreover, we
investigate how this group is embedded in general linear groups.
| [
{
"version": "v1",
"created": "Mon, 17 Feb 2025 07:07:56 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:44:16 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Benkhalifa",
"Mahmoud",
""
]
] |
2502.11748 | Giorgos Kordopatis-Zilos | Giorgos Kordopatis-Zilos, Vladan Stojni\'c, Anna Manko, Pavel
\v{S}uma, Nikolaos-Antonios Ypsilantis, Nikos Efthymiadis, Zakaria Laskar,
Ji\v{r}\'i Matas, Ond\v{r}ej Chum, Giorgos Tolias | ILIAS: Instance-Level Image retrieval At Scale | CVPR 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work introduces ILIAS, a new test dataset for Instance-Level Image
retrieval At Scale. It is designed to evaluate the ability of current and
future foundation models and retrieval techniques to recognize particular
objects. The key benefits over existing datasets include large scale, domain
diversity, accurate ground truth, and a performance that is far from saturated.
ILIAS includes query and positive images for 1,000 object instances, manually
collected to capture challenging conditions and diverse domains. Large-scale
retrieval is conducted against 100 million distractor images from YFCC100M. To
avoid false negatives without extra annotation effort, we include only query
objects confirmed to have emerged after 2014, i.e. the compilation date of
YFCC100M. An extensive benchmarking is performed with the following
observations: i) models fine-tuned on specific domains, such as landmarks or
products, excel in that domain but fail on ILIAS ii) learning a linear
adaptation layer using multi-domain class supervision results in performance
improvements, especially for vision-language models iii) local descriptors in
retrieval re-ranking are still a key ingredient, especially in the presence of
severe background clutter iv) the text-to-image performance of the
vision-language foundation models is surprisingly close to the corresponding
image-to-image case. website: https://vrg.fel.cvut.cz/ilias/
| [
{
"version": "v1",
"created": "Mon, 17 Feb 2025 12:42:38 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 17:27:09 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kordopatis-Zilos",
"Giorgos",
""
],
[
"Stojnić",
"Vladan",
""
],
[
"Manko",
"Anna",
""
],
[
"Šuma",
"Pavel",
""
],
[
"Ypsilantis",
"Nikolaos-Antonios",
""
],
[
"Efthymiadis",
"Nikos",
""
],
[
"Laskar",
"Zakaria",
""
],
[
"Matas",
"Jiří",
""
],
[
"Chum",
"Ondřej",
""
],
[
"Tolias",
"Giorgos",
""
]
] |
2502.11931 | Mrinal Sarkar | Mrinal Sarkar and Shamik Gupta | Asymmetric simple exclusion process on a random comb: Transport
properties in the stationary state | 20 pages, 4 figures (Text revised, no changes to results) | null | null | null | cond-mat.stat-mech | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We address the dynamics of interacting particles on a disordered lattice
formed by a random comb. The dynamics comprises that of the asymmetric simple
exclusion process, whereby motion to nearest-neighour sites that are empty is
more likely in the direction of a bias than in the opposite direction. The
random comb comprises a backbone lattice from each site of which emanates a
branch with a random number of sites. The backbone and the branches run in the
direction of the bias. The number of branch sites or alternatively the branch
lengths are sampled independently from a common distribution, specifically, an
exponential distribution. The system relaxes at long times into a
nonequilibrium stationary state. We analyse the stationary-state density of
sites across the random comb, and also explore the transport properties, in
particular, the stationary-state drift velocity of particles along the
backbone. We show that in the stationary state, the density is uniform along
the backbone and nonuniform along the branches, decreasing monotonically from
the free-end of a branch to its intersection with the backbone. On the other
hand, the drift velocity as a function of the bias strength has a non-monotonic
dependence, first increasing and then decreasing with increase of bias.
However, remarkably, as the particle density increases, the dependence becomes
no more non-monotonic. We understand this effect as a consequence of an
interplay between biased hopping and hard-core exclusion, whereby sites towards
the free end of the branches remain occupied for long times and become
effectively non-participatory in the dynamics of the system. This results in an
effective reduction of the branch lengths and a motion of the particles that
takes place primarily along the backbone.
| [
{
"version": "v1",
"created": "Mon, 17 Feb 2025 15:41:48 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 16:25:33 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Sarkar",
"Mrinal",
""
],
[
"Gupta",
"Shamik",
""
]
] |
2502.12282 | Fatemeh Kayanikhoo | Fatemeh Kayanikhoo and W{\l}odek Klu\'zniak and Miljenko
\v{C}emelji\'c | ULX collimation by outflows in moderately magnetized neutron stars | 10 pages, 6 figures, Accepted in APJ | ApJ 982 95 (2025) | 10.3847/1538-4357/adb714 | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | We perform radiative magnetohydrodynamics simulations in general relativity
(GRRMHD) of super-Eddington disk accretion onto neutron stars endowed with a
magnetic dipole corresponding to surface strengths not exceeding 100 GigaGauss.
Accretion is found to power strong outflows which collimate the emergent
radiation of the accretion columns, leading to apparent radiative luminosities
of $\sim 100$ Eddington, when the true luminosity is a few Eddington units.
Surprisingly, the collimation cone/angle widens with increasing magnetic field.
Thus, in our simulations the apparent luminosity of the neutron star is
substantially larger for the weaker magnetic fields ($10^{10}\,$G) than for the
stronger ones ($10^{11}\,$G). We conclude that a super-Eddington accreting
neutron star with the dipole magnetic field $10^{10}\,$G is the most likely
source of ultraluminous X-rays.
| [
{
"version": "v1",
"created": "Mon, 17 Feb 2025 19:38:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kayanikhoo",
"Fatemeh",
""
],
[
"Kluźniak",
"Włodek",
""
],
[
"Čemeljić",
"Miljenko",
""
]
] |
2502.12679 | Oswaldo Oliveira de | Oswaldo Rio Branco de Oliveira | The Change of Variable Formula Integrals, do they have equal value? | 17 pages, 3 figures | null | null | null | math.CA | http://creativecommons.org/licenses/by/4.0/ | Assuming that the two integrals in the Change of Variable Formula for the
Riemann integral on the real line are finite, one can ask rightfully ask if
they have equal value. We give a positive answer to this question. The proof is
very easy to follow and to keep in mind. Two examples are given.
| [
{
"version": "v1",
"created": "Tue, 18 Feb 2025 09:35:13 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 22:36:32 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 12:20:00 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"de Oliveira",
"Oswaldo Rio Branco",
""
]
] |
2502.12767 | Junseong Choi | Sumin Jo, Junseong Choi, Jiho Kim, Edward Choi | R2-KG: General-Purpose Dual-Agent Framework for Reliable Reasoning on
Knowledge Graphs | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent studies have combined Large Language Models (LLMs) with Knowledge
Graphs (KGs) to enhance reasoning, improving inference accuracy without
additional training while mitigating hallucination. However, existing
frameworks are often rigid, struggling to adapt to KG or task changes. They
also rely heavily on powerful LLMs for reliable (i.e., trustworthy) reasoning.
To address this, We introduce R2-KG, a plug-and-play, dual-agent framework that
separates reasoning into two roles: an Operator (a low-capacity LLM) that
gathers evidence and a Supervisor (a high-capacity LLM) that makes final
judgments. This design is cost-efficient for LLM inference while still
maintaining strong reasoning accuracy. Additionally, R2-KG employs an
Abstention mechanism, generating answers only when sufficient evidence is
collected from KG, which significantly enhances reliability. Experiments across
multiple KG-based reasoning tasks show that R2-KG consistently outperforms
baselines in both accuracy and reliability, regardless of the inherent
capability of LLMs used as the Operator. Further experiments reveal that the
single-agent version of R2-KG, equipped with a strict self-consistency
strategy, achieves significantly higher-than-baseline reliability while
reducing inference cost. However, it also leads to a higher abstention rate in
complex KGs. Our findings establish R2-KG as a flexible and cost-effective
solution for KG-based reasoning. It reduces reliance on high-capacity LLMs
while ensuring trustworthy inference. The code is available at
https://github.com/ekrxjwh2009/R2-KG/.
| [
{
"version": "v1",
"created": "Tue, 18 Feb 2025 11:31:52 GMT"
},
{
"version": "v2",
"created": "Thu, 6 Mar 2025 06:41:40 GMT"
},
{
"version": "v3",
"created": "Mon, 24 Mar 2025 02:24:19 GMT"
},
{
"version": "v4",
"created": "Thu, 27 Mar 2025 07:49:51 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Jo",
"Sumin",
""
],
[
"Choi",
"Junseong",
""
],
[
"Kim",
"Jiho",
""
],
[
"Choi",
"Edward",
""
]
] |
2502.12902 | Christopher B\"ulte | Christopher B\"ulte, Philipp Scholl, Gitta Kutyniok | Probabilistic neural operators for functional uncertainty quantification | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Neural operators aim to approximate the solution operator of a system of
differential equations purely from data. They have shown immense success in
modeling complex dynamical systems across various domains. However, the
occurrence of uncertainties inherent in both model and data has so far rarely
been taken into account\textemdash{}a critical limitation in complex, chaotic
systems such as weather forecasting. In this paper, we introduce the
probabilistic neural operator (PNO), a framework for learning probability
distributions over the output function space of neural operators. PNO extends
neural operators with generative modeling based on strictly proper scoring
rules, integrating uncertainty information directly into the training process.
We provide a theoretical justification for the approach and demonstrate
improved performance in quantifying uncertainty across different domains and
with respect to different baselines. Furthermore, PNO requires minimal
adjustment to existing architectures, shows improved performance for most
probabilistic prediction tasks, and leads to well-calibrated predictive
distributions and adequate uncertainty representations even for long dynamical
trajectories. Implementing our approach into large-scale models for physical
applications can lead to improvements in corresponding uncertainty
quantification and extreme event identification, ultimately leading to a deeper
understanding of the prediction of such surrogate models.
| [
{
"version": "v1",
"created": "Tue, 18 Feb 2025 14:42:11 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 13:19:41 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Bülte",
"Christopher",
""
],
[
"Scholl",
"Philipp",
""
],
[
"Kutyniok",
"Gitta",
""
]
] |
2502.12920 | Thomas Lee | Thomas L. Lee, William Toner, Rajkarn Singh, Artjom Joosen and Martin
Asenov | Lightweight Online Adaption for Time Series Foundation Model Forecasts | 8 pages, Preprint | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Foundation models (FMs) have emerged as a promising approach for time series
forecasting. While effective, FMs typically remain fixed during deployment due
to the high computational costs of learning them online. Consequently, deployed
FMs fail to adapt their forecasts to current data characteristics, despite the
availability of online feedback from newly arriving data. This raises the
question of whether FM performance can be enhanced by the efficient usage of
this feedback. We propose AdapTS to answer this question.
AdapTS is a lightweight mechanism for the online adaption of FM forecasts in
response to online feedback. AdapTS consists of two parts: a) the
AdapTS-Forecaster which is used to learn the current data distribution; and b)
the AdapTS-Weighter which is used to combine the forecasts of the FM and the
AdapTS-Forecaster. We evaluate the performance of AdapTS in conjunction with
several recent FMs across a suite of standard time series datasets. In all of
our experiments we find that using AdapTS improves performance. This work
demonstrates how efficient usage of online feedback can be used to improve FM
forecasts.
| [
{
"version": "v1",
"created": "Tue, 18 Feb 2025 15:01:02 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 21:36:47 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Lee",
"Thomas L.",
""
],
[
"Toner",
"William",
""
],
[
"Singh",
"Rajkarn",
""
],
[
"Joosen",
"Artjom",
""
],
[
"Asenov",
"Martin",
""
]
] |
2502.13012 | Chaoran Chen | Chaoran Chen, Bingsheng Yao, Ruishi Zou, Wenyue Hua, Weimin Lyu,
Yanfang Ye, Toby Jia-Jun Li, Dakuo Wang | Towards a Design Guideline for RPA Evaluation: A Survey of Large
Language Model-Based Role-Playing Agents | null | null | null | null | cs.HC cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Role-Playing Agent (RPA) is an increasingly popular type of LLM Agent that
simulates human-like behaviors in a variety of tasks. However, evaluating RPAs
is challenging due to diverse task requirements and agent designs. This paper
proposes an evidence-based, actionable, and generalizable evaluation design
guideline for LLM-based RPA by systematically reviewing 1,676 papers published
between Jan. 2021 and Dec. 2024. Our analysis identifies six agent attributes,
seven task attributes, and seven evaluation metrics from existing literature.
Based on these findings, we present an RPA evaluation design guideline to help
researchers develop more systematic and consistent evaluation methods.
| [
{
"version": "v1",
"created": "Tue, 18 Feb 2025 16:33:33 GMT"
},
{
"version": "v2",
"created": "Sat, 15 Mar 2025 17:09:44 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 04:07:19 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Chen",
"Chaoran",
""
],
[
"Yao",
"Bingsheng",
""
],
[
"Zou",
"Ruishi",
""
],
[
"Hua",
"Wenyue",
""
],
[
"Lyu",
"Weimin",
""
],
[
"Ye",
"Yanfang",
""
],
[
"Li",
"Toby Jia-Jun",
""
],
[
"Wang",
"Dakuo",
""
]
] |
2502.13731 | Jessica Lally | Jessica Lally, Milad Kazemi, Nicola Paoletti | Robust Counterfactual Inference in Markov Decision Processes | Fixed typo in Equation (5) | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper addresses a key limitation in existing counterfactual inference
methods for Markov Decision Processes (MDPs). Current approaches assume a
specific causal model to make counterfactuals identifiable. However, there are
usually many causal models that align with the observational and interventional
distributions of an MDP, each yielding different counterfactual distributions,
so fixing a particular causal model limits the validity (and usefulness) of
counterfactual inference. We propose a novel non-parametric approach that
computes tight bounds on counterfactual transition probabilities across all
compatible causal models. Unlike previous methods that require solving
prohibitively large optimisation problems (with variables that grow
exponentially in the size of the MDP), our approach provides closed-form
expressions for these bounds, making computation highly efficient and scalable
for non-trivial MDPs. Once such an interval counterfactual MDP is constructed,
our method identifies robust counterfactual policies that optimise the
worst-case reward w.r.t. the uncertain interval MDP probabilities. We evaluate
our method on various case studies, demonstrating improved robustness over
existing methods.
| [
{
"version": "v1",
"created": "Wed, 19 Feb 2025 13:56:20 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 14:20:21 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Lally",
"Jessica",
""
],
[
"Kazemi",
"Milad",
""
],
[
"Paoletti",
"Nicola",
""
]
] |
2502.13744 | Ken Kangda Wren Dr | Ken Kangda Wren | The Risk-Neutral Equivalent Pricing of Model-Uncertainty | 25 pages of main text, 13 pages of Appendix and Bibliography | null | null | null | q-fin.MF econ.EM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing approaches to asset-pricing under model-uncertainty adapt classical
utility-maximisation frameworks and seek theoretical comprehensiveness. We move
toward practice by considering binary model-uncertainties and by switching
attention from 'preference' to 'constraints'. Economic asset-pricing in this
setting is found to decompose into the viable pricing of model-risk and of
non-model risk separately such that the former has a unique and intuitive
risk-neutral equivalent formulation with convenient properties. Its parameter,
a dynamically conserved constant of model-risk inference, allows an integrated
representation of ex-ante risk-pricing and bias, such that their ex-post
price-effects can be disentangled, through well-known price anomalies such as
Momentum and Low-Risk.
| [
{
"version": "v1",
"created": "Wed, 19 Feb 2025 14:07:37 GMT"
},
{
"version": "v2",
"created": "Fri, 7 Mar 2025 11:26:20 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 10:27:56 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wren",
"Ken Kangda",
""
]
] |
2502.14232 | Elizabeth Silber | Trevor C. Wilson, Elizabeth A. Silber, Thomas A. Colston, Brian R.
Elbing, Thom R. Edwards | Bolide infrasound signal morphology and yield estimates: A case study of
two events detected by a dense acoustic sensor network | 37 pages, 10 figures, 4 tables | The Astronomical Journal, Vol. 169, No. 4, 15pp (2025) | 10.3847/1538-3881/adbb70 | SAND2025-00966O | astro-ph.EP astro-ph.IM physics.ao-ph physics.geo-ph physics.ins-det physics.space-ph | http://creativecommons.org/licenses/by/4.0/ | Two bolides (2 June 2016 and 4 April 2019) were detected at multiple regional
infrasound stations with many of the locations receiving multiple detections.
Analysis of the received signals was used to estimate the yield, location and
trajectory, and the type of shock that produced the received signal. The
results from the infrasound analysis were compared with ground truth
information that was collected through other sensing modalities. This
multi-modal framework offers an expanded perspective on the processes governing
bolide shock generation and propagation. The majority of signal features showed
reasonable agreement between the infrasound-based interpretation and the other
observational modalities, though the yield estimate from the 2019 bolide was
significantly lower using the infrasound detections. There was also evidence
suggesting that one of the detections was from a cylindrical shock that was
initially propagating upward, which is unusual though not impossible.
| [
{
"version": "v1",
"created": "Thu, 20 Feb 2025 03:50:31 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wilson",
"Trevor C.",
""
],
[
"Silber",
"Elizabeth A.",
""
],
[
"Colston",
"Thomas A.",
""
],
[
"Elbing",
"Brian R.",
""
],
[
"Edwards",
"Thom R.",
""
]
] |
2502.14249 | Gauri Batra | Gauri Batra and Henry W. Lin | Giant gravitons in D$p$-brane holography | 22 pages. v2: refs added, improved Appendix B | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider half BPS operators in maximally supersymmetric Yang Mills (SYM)
in $p+1$ dimensions. These operators satisfy trace relations that are identical
to those discussed in the $p=3$ case ($\mathcal{N} = 4$ SYM). Nevertheless, the
bulk explanation of these trace relations must differ from the $p = 3$ case as
their holographic duals are not AdS spacetimes. We identify giant graviton
solutions in the dual holographic backgrounds for $-1 \le p \le 4$. In the 't
Hooft limit, these giants are D$(6-p)$-branes that wrap a $S_{6-p} \subset
S_{8-p}$. We also follow the giants into the strong coupling region where they
become other branes. Despite propagating in a non-AdS geometry, we find that
the branes "feel" like they are in AdS. This is closely related to the emergent
scaling symmetry present in these boundary theories.
| [
{
"version": "v1",
"created": "Thu, 20 Feb 2025 04:32:49 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 04:50:14 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Batra",
"Gauri",
""
],
[
"Lin",
"Henry W.",
""
]
] |
2502.14535 | Ron Folman | Or Dobkowski, Barak Trok, Peter Skakunenko, Yonathan Japha, David
Groswasser, Maxim Efremov, Chiara Marletto, Ivette Fuentes, Roger Penrose,
Vlatko Vedral, Wolfgang P. Schleich, Ron Folman | Observation of the quantum equivalence principle for matter-waves | null | null | null | null | quant-ph cond-mat.quant-gas gr-qc hep-th physics.atom-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Einstein's general theory of relativity is based on the principle of
equivalence - in essence, dating back to Galileo - which asserts that, locally,
the effect of a gravitational field is equivalent to that of an accelerating
reference frame, so that the local gravitational field is eliminated in a
freely-falling frame. Einstein's theory enables this principle to extend to a
global description of relativistic space-time, at the expense of allowing
space-time to become curved, realising a consistent frame-independent
description of nature at the classical level. Einstein's theory has been
confirmed to great accuracy for astrophysical bodies. However, in the quantum
domain the equivalence principle has been predicted to take a unique form
involving a gauge phase that is observable if the wavefunction - fundamental to
quantum descriptions - allows an object to interfere with itself after being
simultaneously at rest in two differently accelerating frames, one being the
laboratory (Newtonian) frame and the other in the freely-falling (Einsteinian)
frame. To measure this gauge phase we realise a novel cold-atom interferometer
in which one wave packet stays static in the laboratory frame while the other
is in free fall. We follow the relative-phase evolution of the wave packets in
the two frames, confirming the equivalence principle in the quantum domain. Our
observation is yet another fundamental test of the interface between quantum
theory and gravity. The new interferometer also opens the door for further
probing of the latter interface, as well as to searches for new physics.
| [
{
"version": "v1",
"created": "Thu, 20 Feb 2025 13:11:26 GMT"
},
{
"version": "v2",
"created": "Mon, 3 Mar 2025 18:28:25 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Mar 2025 15:29:46 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Dobkowski",
"Or",
""
],
[
"Trok",
"Barak",
""
],
[
"Skakunenko",
"Peter",
""
],
[
"Japha",
"Yonathan",
""
],
[
"Groswasser",
"David",
""
],
[
"Efremov",
"Maxim",
""
],
[
"Marletto",
"Chiara",
""
],
[
"Fuentes",
"Ivette",
""
],
[
"Penrose",
"Roger",
""
],
[
"Vedral",
"Vlatko",
""
],
[
"Schleich",
"Wolfgang P.",
""
],
[
"Folman",
"Ron",
""
]
] |
2502.15394 | Matthias Schymura | Bj\"orn Kriepke and Matthias Schymura | On generic $\Delta$-modular integer matrices with two rows | 27 pages, 1 figure, added application to matroids | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The column number question asks for the maximal number of columns of an
integer matrix with the property that all its rank size minors are bounded by a
fixed parameter $\Delta$ in absolute value. Polynomial upper bounds have been
proved in various settings in recent years, with consequences for algorithmic
questions in integer linear programming and matroid theory. In this paper, we
focus on the exact determination of the maximal column number of such matrices
with two rows and no vanishing $2$-minors. We prove that for large enough
$\Delta$, this number is a quasi-linear function, non-decreasing and always
even. Such basic structural properties of column number functions are barely
known, but expected to hold in other settings as well. Moreover, our results
identify the unique excluded (co)rank two minors for the class of matroids that
are representable as a $\Delta$-submodular matrix.
| [
{
"version": "v1",
"created": "Fri, 21 Feb 2025 11:42:18 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:34:58 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kriepke",
"Björn",
""
],
[
"Schymura",
"Matthias",
""
]
] |
2502.15682 | Guanqi Zhan | Guanqi Zhan, Yuanpei Liu, Kai Han, Weidi Xie, Andrew Zisserman | ELIP: Enhanced Visual-Language Foundation Models for Image Retrieval | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The objective in this paper is to improve the performance of text-to-image
retrieval. To this end, we introduce a new framework that can boost the
performance of large-scale pre-trained vision-language models, so that they can
be used for text-to-image re-ranking. The approach, Enhanced Language-Image
Pre-training (ELIP), uses the text query, via a simple MLP mapping network, to
predict a set of visual prompts to condition the ViT image encoding. ELIP can
easily be applied to the commonly used CLIP, SigLIP and BLIP-2 networks. To
train the architecture with limited computing resources, we develop a 'student
friendly' best practice, involving global hard sample mining, and curation of a
large-scale dataset. On the evaluation side, we set up two new
out-of-distribution (OOD) benchmarks, Occluded COCO and ImageNet-R, to assess
the zero-shot generalisation of the models to different domains. The results
demonstrate that ELIP significantly boosts CLIP/SigLIP/SigLIP-2 text-to-image
retrieval performance and outperforms BLIP-2 on several benchmarks, as well as
providing an easy means to adapt to OOD datasets.
| [
{
"version": "v1",
"created": "Fri, 21 Feb 2025 18:59:57 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:57:43 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Zhan",
"Guanqi",
""
],
[
"Liu",
"Yuanpei",
""
],
[
"Han",
"Kai",
""
],
[
"Xie",
"Weidi",
""
],
[
"Zisserman",
"Andrew",
""
]
] |
2502.16398 | Lasse Wulf | Lasse Wulf | Computing the Polytope Diameter is Even Harder than NP-hard (Already for
Perfect Matchings) | null | null | null | null | cs.CC cs.CG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The diameter of a polytope is a fundamental geometric parameter that plays a
crucial role in understanding the efficiency of the simplex method. Despite its
central nature, the computational complexity of computing the diameter of a
given polytope is poorly understood. Already in 1994, Frieze and Teng [Comp.
Compl.] recognized the possibility that this task could potentially be harder
than NP-hard, and asked whether the corresponding decision problem is complete
for the second stage of the polynomial hierarchy, i.e. $\Pi^p_2$-complete. In
the following years, partial results could be obtained. In a cornerstone
result, Frieze and Teng themselves proved weak NP-hardness for a family of
custom defined polytopes. Sanit\`a [FOCS18] in a break-through result proved
that already for the much simpler fractional matching polytope the problem is
strongly NP-hard. Very recently, Steiner and N\"obel [SODA25] generalized this
result to the even simpler bipartite perfect matching polytope and the circuit
diameter. In this paper, we finally show that computing the diameter of the
bipartite perfect matching polytope is $\Pi^p_2$-hard. Since the corresponding
decision problem is also trivially contained in $\Pi^p_2$, this decidedly
answers Frieze and Teng's 30 year old question. Our results also hold when the
diameter is replaced by the circuit diameter. As our second main result, we
prove that for some $\varepsilon > 0$ the (circuit) diameter of the bipartite
perfect matching polytope cannot be approximated by a factor better than $(1 +
\varepsilon)$. This answers a recent question by N\"obel and Steiner. It is the
first known inapproximability result for the circuit diameter, and extends
Sanit\`a's inapproximability result of the diameter to the totally unimodular
case.
| [
{
"version": "v1",
"created": "Sun, 23 Feb 2025 01:17:39 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 16:09:11 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Wulf",
"Lasse",
""
]
] |
2502.17190 | Ralf Meyer | Bartosz K. Kwa\'sniewski, Ralf Meyer, Akshara Prasad | Type semigroups for twisted groupoids and a dichotomy for groupoid
C*-algebras | 54 pages; small changes based on comments by readers | null | null | null | math.OA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We develop a theory of type semigroups for arbitrary twisted, not necessarily
Hausdorff \'etale groupoids. The type semigroup is a dynamical version of the
Cuntz semigroup. We relate it to traces, ideals, pure infiniteness, and stable
finiteness of the reduced and essential C*-algebras. If the reduced C*-algebra
of a twisted groupoid is simple and the type semigroup satisfies a weak version
of almost unperforation, then the C*-algebra is either stably finite or purely
infinite. We apply our theory to Cartan inclusions. We calculate the type
semigroup for the possibly non-Hausdorff groupoids associated to self-similar
group actions on graphs and deduce a dichotomy for the resulting Exel-Pardo
algebras.
| [
{
"version": "v1",
"created": "Mon, 24 Feb 2025 14:24:40 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 01:12:56 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Kwaśniewski",
"Bartosz K.",
""
],
[
"Meyer",
"Ralf",
""
],
[
"Prasad",
"Akshara",
""
]
] |
2502.17456 | Zhen Huang | Hu Ding, Pengxiang Hua, Zhen Huang | Survey on Recent Progress of AI for Chemistry: Methods, Applications,
and Opportunities | 22 pages, 8 figures, 4 tables | null | null | null | physics.chem-ph cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The development of artificial intelligence (AI) techniques has brought
revolutionary changes across various realms. In particular, the use of
AI-assisted methods to accelerate chemical research has become a popular and
rapidly growing trend, leading to numerous groundbreaking works. In this paper,
we provide a comprehensive review of current AI techniques in chemistry from a
computational perspective, considering various aspects in the design of
methods. We begin by discussing the characteristics of data from diverse
sources, followed by an overview of various representation methods. Next, we
review existing models for several topical tasks in the field, and conclude by
highlighting some key challenges that warrant further attention.
| [
{
"version": "v1",
"created": "Sun, 9 Feb 2025 13:39:49 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Ding",
"Hu",
""
],
[
"Hua",
"Pengxiang",
""
],
[
"Huang",
"Zhen",
""
]
] |
2502.17827 | Entejar Alam | Entejar Alam, Paul J. Rathouz, Peter M\"uller | DPGLM: A Semiparametric Bayesian GLM with Inhomogeneous Normalized
Random Measures | null | null | null | null | stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a novel varying-weight dependent Dirichlet process (DDP) model
that extends a recently developed semi-parametric generalized linear model
(SPGLM) by adding a nonparametric Bayesian prior on the baseline distribution
of the GLM. We show that the resulting model takes the form of an inhomogeneous
completely random measure that arises from exponential tilting of a normalized
completely random measure. Building on familiar posterior sampling methods for
mixtures with respect to normalized random measures, we introduce posterior
simulation in the resulting model. We validate the proposed methodology through
extensive simulation studies and illustrate its application using data from a
speech intelligibility study.
| [
{
"version": "v1",
"created": "Tue, 25 Feb 2025 04:07:59 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Mar 2025 17:10:43 GMT"
}
] | 2025-03-28T00:00:00 | [
[
"Alam",
"Entejar",
""
],
[
"Rathouz",
"Paul J.",
""
],
[
"Müller",
"Peter",
""
]
] |
Subsets and Splits