id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2309.17270 | Robert Webber | Jonathan Weare, Robert J. Webber | Randomly sparsified Richardson iteration: A dimension-independent sparse
linear solver | 31 pages, 2 figures | null | null | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, a class of algorithms combining classical fixed point iterations
with repeated random sparsification of approximate solution vectors has been
successfully applied to eigenproblems with matrices as large as $10^{108}
\times 10^{108}$. So far, a complete mathematical explanation for their success
has proven elusive.
The family of methods has not yet been extended to the important case of
linear system solves. In this paper we propose a new scheme based on repeated
random sparsification that is capable of solving sparse linear systems in
arbitrarily high dimensions. We provide a complete mathematical analysis of
this new algorithm. Our analysis establishes a faster-than-Monte Carlo
convergence rate and justifies use of the scheme even when the solution vector
itself is too large to store.
| [
{
"version": "v1",
"created": "Fri, 29 Sep 2023 14:20:23 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Nov 2023 19:28:36 GMT"
},
{
"version": "v3",
"created": "Sat, 12 Apr 2025 04:18:50 GMT"
},
{
"version": "v4",
"created": "Thu, 24 Apr 2025 19:25:25 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Weare",
"Jonathan",
""
],
[
"Webber",
"Robert J.",
""
]
] |
2309.17401 | Milin Zhang | Milin Zhang, Mohammad Abdi, Jonathan Ashdown and Francesco Restuccia | Adversarial Attacks to Latent Representations of Distributed Neural
Networks in Split Computing | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Distributed deep neural networks (DNNs) have been shown to reduce the
computational burden of mobile devices and decrease the end-to-end inference
latency in edge computing scenarios. While distributed DNNs have been studied,
to the best of our knowledge, the resilience of distributed DNNs to adversarial
action remains an open problem. In this paper, we fill the existing research
gap by rigorously analyzing the robustness of distributed DNNs against
adversarial action. We cast this problem in the context of information theory
and rigorously proved that (i) the compressed latent dimension improves the
robustness but also affect task-oriented performance; and (ii) the deeper
splitting point enhances the robustness but also increases the computational
burden. These two trade-offs provide a novel perspective to design robust
distributed DNN. To test our theoretical findings, we perform extensive
experimental analysis by considering 6 different DNN architectures, 6 different
approaches for distributed DNN and 10 different adversarial attacks using the
ImageNet-1K dataset.
| [
{
"version": "v1",
"created": "Fri, 29 Sep 2023 17:01:29 GMT"
},
{
"version": "v2",
"created": "Fri, 24 Nov 2023 20:50:35 GMT"
},
{
"version": "v3",
"created": "Tue, 30 Jan 2024 16:24:01 GMT"
},
{
"version": "v4",
"created": "Fri, 25 Apr 2025 00:32:41 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Zhang",
"Milin",
""
],
[
"Abdi",
"Mohammad",
""
],
[
"Ashdown",
"Jonathan",
""
],
[
"Restuccia",
"Francesco",
""
]
] |
2310.00205 | Mingjie Shen | Mingjie Shen, Akul Abhilash Pillai, Brian A. Yuan, James C. Davis,
Aravind Machiry | Finding 709 Defects in 258 Projects: An Experience Report on Applying
CodeQL to Open-Source Embedded Software (Experience Paper) -- Extended Report | This is the extended version of: Mingjie Shen, Akul Abhilash Pillai,
Brian A. Yuan, James C. Davis, and Aravind Machiry. 2025. Finding 709 Defects
in 258 Projects: An Experience Report on Applying CodeQL to Open-Source
Embedded Software (Experience Paper). Proc. ACM Softw. Eng. 2, ISSTA, Article
ISSTA048 (July 2025), 24 pages. https://doi.org/10.1145/3728923 | null | null | null | cs.SE cs.CR | http://creativecommons.org/licenses/by/4.0/ | In this experience paper, we report on a large-scale empirical study of
Static Application Security Testing (SAST) in Open-Source Embedded Software
(EMBOSS) repositories. We collected a corpus of 258 of the most popular EMBOSS
projects, and then measured their use of SAST tools via program analysis and a
survey (N=25) of their developers. Advanced SAST tools are rarely used -- only
3% of projects go beyond trivial compiler analyses. Developers cited the
perception of ineffectiveness and false positives as reasons for limited
adoption. Motivated by this deficit, we applied the state-of-the-art (SOTA)
CodeQL SAST tool and measured its ease of use and actual effectiveness. Across
the 258 projects, CodeQL reported 709 true defects with a false positive rate
of 34%. There were 535 (75%) likely security vulnerabilities, including in
major projects maintained by Microsoft, Amazon, and the Apache Foundation.
EMBOSS engineers have confirmed 376 (53%) of these defects, mainly by accepting
our pull requests. Two CVEs were issued. Based on these results, we proposed
pull requests to include our workflows as part of EMBOSS Continuous Integration
(CI) pipelines, 37 (71% of active repositories) of these are already merged. In
summary, we urge EMBOSS engineers to adopt the current generation of SAST
tools, which offer low false positive rates and are effective at finding
security-relevant defects.
| [
{
"version": "v1",
"created": "Sat, 30 Sep 2023 00:49:55 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 05:18:45 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Shen",
"Mingjie",
""
],
[
"Pillai",
"Akul Abhilash",
""
],
[
"Yuan",
"Brian A.",
""
],
[
"Davis",
"James C.",
""
],
[
"Machiry",
"Aravind",
""
]
] |
2310.10484 | Emir Syahreza Fadhilla | Emir Syahreza Fadhilla, Ardian Nata Atmaja, Bobby Eka Gunara, Mir
Faizal | Topologically Stable BPS and Non-BPS States in Supersymmetric
$\mathcal{N}=2$ Baby-Skyrme Model | 14 pages, 1 figures, abstract and title changed | Phys.Lett.B 864 (2025) 139447 | 10.1016/j.physletb.2025.139447 | null | hep-th | http://creativecommons.org/licenses/by/4.0/ | The supersymmetric baby-Skyrme model is an interesting field theoretical
model, and its BPS states have been studied using the usual methods. Here, we
propose a novel method to rigorously obtain both topologically stable BPS and
non-BPS states in the $\mathcal{N}=2$ baby Skyrme Model. It is observed that
the BPS states found using this novel method coincide with the BPS states found
using the usual methods. However, we are also able to obtain the non-BPS
states, which break all of the supersymmetry of the theory. Furthermore, there
exists a one-parameter family of non-BPS solutions that are connected to the
half-BPS solutions, where half of the supersymmetry is restored when the
parameter is set to zero. The proposed method is general, and we expect that it
might be useful for investigating the topologically stable non-BPS states of
other theories. Thus, this method could possibly have wide applications for the
study of non-BPS states in supersymmetric theories.
| [
{
"version": "v1",
"created": "Mon, 16 Oct 2023 15:04:11 GMT"
},
{
"version": "v2",
"created": "Mon, 15 Apr 2024 09:40:30 GMT"
},
{
"version": "v3",
"created": "Thu, 2 Jan 2025 03:51:01 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Fadhilla",
"Emir Syahreza",
""
],
[
"Atmaja",
"Ardian Nata",
""
],
[
"Gunara",
"Bobby Eka",
""
],
[
"Faizal",
"Mir",
""
]
] |
2310.11135 | Shakti Singh | Shakti Singh, Manan Dholakia and Sharat Chandra | Towards atomistic understanding of Iron phosphate glass: a
first-principles based DFT modeling and study of its physical properties | 40 pages, 21 figures, manuscript for journal | The Journal of Physical Chemistry B 128 (13), 3258-3272 (2024) | 10.1021/acs.jpcb.3c07670 | null | cond-mat.mtrl-sci | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Iron phosphate glasses (IPG) have been proposed as futuristic glass material
for nuclear waste immobilization, anode material for lithium batteries and also
as bioactive glass. In the last decade, there have been attempts to propose
atomistic models of IPG to explain their properties from atomistic viewpoint
and to predict their behavior in radioactive environment. In this paper, we
seek to produce small scale models of IPG that can be handled within the scheme
of Density Functional Theory (DFT) to study the electronic structure of this
material. The starting models generated using Monte Carlo (MC) method [S. Singh
and S. Chandra, Comp. Mat. Sci., 202, 110943, (2022)] were subsequently
annealed (at 1000 K) using ab-initio molecular dynamics (AIMD). This removes
coordination defects present in the MC models. The equilibrated structure at
this temperature was then force-relaxed using conjugate-gradient (CG)
optimization. This hybrid approach (MC + AIMD + 0K DFT-CG optimization)
produced good atomistic models of IPG which can reproduce experimentally
observed electronic band-gap, vibrational density of states (VDOS), magnetic
moment of Fe, the elastic constants as well as optical and dielectric
properties. Computationally expensive melt-quench simulation can be avoided
using present approach allowing the use of DFT for accurate calculations of
properties of complex glass like IPG.
| [
{
"version": "v1",
"created": "Tue, 17 Oct 2023 10:33:15 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Singh",
"Shakti",
""
],
[
"Dholakia",
"Manan",
""
],
[
"Chandra",
"Sharat",
""
]
] |
2310.13561 | Guillem Ram\'irez | Guillem Ram\'irez and Matthias Lindemann and Alexandra Birch and Ivan
Titov | Cache & Distil: Optimising API Calls to Large Language Models | null | Findings of the Association for Computational Linguistics: ACL
2024 | 10.18653/v1/2024.findings-acl.704 | null | cs.CL cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Large-scale deployment of generative AI tools often depends on costly API
calls to a Large Language Model (LLM) to fulfil user queries. To curtail the
frequency of these calls, one can employ a smaller language model -- a student
-- which is continuously trained on the responses of the LLM. This student
gradually gains proficiency in independently handling an increasing number of
user requests, a process we term neural caching. The crucial element in neural
caching is a policy that decides which requests should be processed by the
student alone and which should be redirected to the LLM, subsequently aiding
the student's learning. In this study, we focus on classification tasks, and we
consider a range of classic active learning-based selection criteria as the
policy. Our experiments suggest that Margin Sampling and Query by Committee
bring consistent benefits across tasks and budgets.
| [
{
"version": "v1",
"created": "Fri, 20 Oct 2023 15:01:55 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Ramírez",
"Guillem",
""
],
[
"Lindemann",
"Matthias",
""
],
[
"Birch",
"Alexandra",
""
],
[
"Titov",
"Ivan",
""
]
] |
2310.17295 | Hans Lei{\ss} | Mark Hopkins and Hans Lei{\ss} | Normal Forms for Elements of ${}^*$-Continuous Kleene Algebras
Representing the Context-Free Languages | 2nd Revised version. 42 pages, 4 figures. Additional results in Sec.2
and Sec.5, Cor.2.7 corrected | null | null | null | cs.FL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Within the tensor product $K \mathop{\otimes_{\cal R}} C_2'$ of any
${}^*$-continuous Kleene algebra $K$ with the polycyclic ${}^*$-continuous
Kleene algebra $C_2'$ over two bracket pairs there is a copy of the fixed-point
closure of $K$: the centralizer of $C_2'$ in $K \mathop{\otimes_{\cal R}}
C_2'$. Using an automata-theoretic representation of elements of
$K\mathop{\otimes_{\cal R}} C_2'$ \`a la Kleene, with the aid of normal form
theorems that restrict the occurrences of brackets on paths through the
automata, we develop a foundation for a calculus of context-free expressions
without variable binders. We also give some results on the bra-ket
${}^*$-continuous Kleene algebra $C_2$, motivate the ``completeness equation''
that distinguishes $C_2$ from $C_2'$, and show that $C_2'$ already validates a
relativized form of this equation.
| [
{
"version": "v1",
"created": "Thu, 26 Oct 2023 10:23:02 GMT"
},
{
"version": "v2",
"created": "Sat, 14 Sep 2024 18:42:20 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 13:12:05 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Hopkins",
"Mark",
""
],
[
"Leiß",
"Hans",
""
]
] |
2310.19380 | Meng Lou | Meng Lou, Shu Zhang, Hong-Yu Zhou, Sibei Yang, Chuan Wu, Yizhou Yu | TransXNet: Learning Both Global and Local Dynamics with a Dual Dynamic
Token Mixer for Visual Recognition | Accepted by IEEE TNNLS | null | 10.1109/TNNLS.2025.3550979 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recent studies have integrated convolutions into transformers to introduce
inductive bias and improve generalization performance. However, the static
nature of conventional convolution prevents it from dynamically adapting to
input variations, resulting in a representation discrepancy between convolution
and self-attention as the latter computes attention maps dynamically.
Furthermore, when stacking token mixers that consist of convolution and
self-attention to form a deep network, the static nature of convolution hinders
the fusion of features previously generated by self-attention into convolution
kernels. These two limitations result in a sub-optimal representation capacity
of the entire network. To find a solution, we propose a lightweight Dual
Dynamic Token Mixer (D-Mixer) to simultaneously learn global and local dynamics
via computing input-dependent global and local aggregation weights. D-Mixer
works by applying an efficient global attention module and an input-dependent
depthwise convolution separately on evenly split feature segments, endowing the
network with strong inductive bias and an enlarged receptive field. We use
D-Mixer as the basic building block to design TransXNet, a novel hybrid
CNN-Transformer vision backbone network that delivers compelling performance.
In the ImageNet-1K classification, TransXNet-T surpasses Swin-T by 0.3% in
top-1 accuracy while requiring less than half of the computational cost.
Furthermore, TransXNet-S and TransXNet-B exhibit excellent model scalability,
achieving top-1 accuracy of 83.8% and 84.6% respectively, with reasonable
computational costs. Additionally, our proposed network architecture
demonstrates strong generalization capabilities in various dense prediction
tasks, outperforming other state-of-the-art networks while having lower
computational costs. Code is publicly available at
https://github.com/LMMMEng/TransXNet.
| [
{
"version": "v1",
"created": "Mon, 30 Oct 2023 09:35:56 GMT"
},
{
"version": "v2",
"created": "Thu, 30 Nov 2023 01:48:03 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Mar 2025 09:09:26 GMT"
},
{
"version": "v4",
"created": "Mon, 31 Mar 2025 08:29:53 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Lou",
"Meng",
""
],
[
"Zhang",
"Shu",
""
],
[
"Zhou",
"Hong-Yu",
""
],
[
"Yang",
"Sibei",
""
],
[
"Wu",
"Chuan",
""
],
[
"Yu",
"Yizhou",
""
]
] |
2311.02050 | Sariel Har-Peled | Pankaj K. Agarwal, Sariel Har-Peled, Rahul Raychaudhury and Stavros
Sintos | Fast Approximation Algorithms for Piercing Boxes by Points | Appeared in SODA 2024 | null | null | null | cs.CG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | $\newcommand{\popt}{{\mathcal{p}}}
\newcommand{\Re}{\mathbb{R}}\newcommand{\N}{{\mathcal{N}}}
\newcommand{\BX}{\mathcal{B}} \newcommand{\bb}{\mathsf{b}}
\newcommand{\eps}{\varepsilon} \newcommand{\polylog}{\mathrm{polylog}} $ Let
$\mathcal{B}=\{\mathsf{b}_1, \ldots ,\mathsf{b}_n\}$ be a set of $n$
axis-aligned boxes in $\Re^d$ where $d\geq2$ is a constant. The \emph{piercing
problem} is to compute a smallest set of points $\N \subset \Re^d$ that hits
every box in $\mathcal{B}$, i.e., $\N\cap \mathsf{b}_i\neq \emptyset$, for
$i=1,\ldots, n$. Let $\popt=\popt(\mathcal{B})$, the \emph{piercing number} be
the minimum size of a piercing set of $\mathcal{B}$. We present a randomized
$O(d^2\log\log \popt)$-approximation algorithm with expected running time
$O(n^{d/2}\polylog n)$. Next, we present a faster $O(n^{\log d+1})$-time
algorithm but with a slightly inferior approximation factor of
$O(2^{4d}\log\log\popt)$. The running time of both algorithms can be improved
to near-linear using a sampling-based technique, if $\popt = O(n^{1/d})$.
For the dynamic version of the problem in the plane, we obtain a randomized
$O(\log\log\popt)$-approximation algorithm with $O(n^{1/2}\polylog n )$
amortized expected update time for insertion or deletion of boxes. For squares
in $\Re^2$, the update time can be improved to $O(n^{1/3}\polylog n )$.
| [
{
"version": "v1",
"created": "Fri, 3 Nov 2023 17:27:07 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 03:24:44 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Agarwal",
"Pankaj K.",
""
],
[
"Har-Peled",
"Sariel",
""
],
[
"Raychaudhury",
"Rahul",
""
],
[
"Sintos",
"Stavros",
""
]
] |
2311.09924 | Wolfgang Pitsch | Wolfgang Pitsch and Ricard Riba | Finite type invariants in low degrees and the Johnson filtration | 32 pages. Corrected a proof that needed a boundary argument. Added a
section on the surgery formula for the new invariant. Corrected minor typos | null | null | null | math.GT | http://creativecommons.org/licenses/by/4.0/ | We study the behaviour of the Casson invariant $\lambda$, its square, and
Othsuki's second invariant $\lambda_2$ as functions on the Johnson subgroup of
the mapping class group. We show that since $\lambda$ and $d_2 = \lambda_2 - 18
\lambda^2$ are invariants that are morphisms on respectively the second and the
third level of the Johnson filtration they never vanish on any level of this
filtration. In contrast we prove that the invariant $\lambda_2-18\lambda^2
+3\lambda$ vanishes on the fifth level of the Johnson filtration,
$\mathcal{M}_{g,1}(5)$, and as a consequence we prove that, for instance, the
Poincar\'e homology sphere does not admit any Heegaard splitting with gluing
map an element in $\mathcal{M}_{g,1}(5)$. Finally we determine a surgery
formula for Othsuki's second invariant $\lambda_2$.
| [
{
"version": "v1",
"created": "Thu, 16 Nov 2023 14:23:33 GMT"
},
{
"version": "v2",
"created": "Thu, 18 Jan 2024 13:16:09 GMT"
},
{
"version": "v3",
"created": "Mon, 25 Nov 2024 11:04:41 GMT"
},
{
"version": "v4",
"created": "Fri, 25 Apr 2025 13:55:17 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Pitsch",
"Wolfgang",
""
],
[
"Riba",
"Ricard",
""
]
] |
2311.15585 | Tian-Yu Zhao | Tianyu Zhao, Ruijun Shi, Yue Zhou, Zhoujian Cao, Zhixiang Ren | Dawning of a New Era in Gravitational Wave Data Analysis: Unveiling
Cosmic Mysteries via Artificial Intelligence -- A Systematic Review | 24 pages, 12 figures | Front. Phys. 20, 045301 (2025) | 10.15302/frontphys.2025.045301 | null | gr-qc astro-ph.HE astro-ph.IM | http://creativecommons.org/licenses/by/4.0/ | Gravitational wave data analysis (GWDA) faces significant challenges due to
high-dimensional parameter spaces and non-Gaussian, non-stationary artifacts in
the interferometer background, which traditional methods have made significant
progress in addressing but continue to face limitations. Artificial
intelligence (AI), particularly deep learning (DL) algorithms, offers potential
advantages, including computational efficiency, scalability, and adaptability,
which may complement traditional approaches in tackling these challenges more
effectively. In this review, we explore AI-driven approaches to GWDA, covering
every stage of the pipeline and presenting first explorations in waveform
modeling and parameter estimation. This work represents the most comprehensive
review to date, integrating the latest AI advancements with practical GWDA
applications. Our meta-analysis reveals insights and trends, highlighting the
transformative potential of AI in revolutionizing gravitational wave research
and paving the way for future discoveries.
| [
{
"version": "v1",
"created": "Mon, 27 Nov 2023 07:21:24 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 06:57:43 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Zhao",
"Tianyu",
""
],
[
"Shi",
"Ruijun",
""
],
[
"Zhou",
"Yue",
""
],
[
"Cao",
"Zhoujian",
""
],
[
"Ren",
"Zhixiang",
""
]
] |
2312.01259 | Fumika Suzuki | Fumika Suzuki and Wojciech H. Zurek | Topological defect formation in a phase transition with tunable order | Supplementary movie showing time evolution of the order parameter is
available at https://fumikaphys.com/tunable.mp4 (Simplified explanation; see
paper for details.) 8 pages, 5 figures | Phys. Rev. Lett. 132, 241601 (2024) | 10.1103/PhysRevLett.132.241601 | LA-UR-23-32921 | cond-mat.stat-mech gr-qc hep-ph hep-th quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Kibble-Zurek mechanism (KZM) describes the non-equilibrium dynamics and
topological defect formation in systems undergoing second-order phase
transitions. KZM has found applications in fields such as cosmology and
condensed matter physics. However, it is generally not suitable for describing
first-order phase transitions. It has been demonstrated that transitions in
systems like superconductors or charged superfluids, typically classified as
second-order, can exhibit weakly first-order characteristics when the influence
of fluctuations is taken into account. Moreover, the order of the phase
transition (i.e., the extent to which it becomes first rather than second
order) can be tuned. We explore quench-induced formation of topological defects
in such tunable phase transitions and propose that their density can be
predicted by combining KZM with nucleation theory.
| [
{
"version": "v1",
"created": "Sun, 3 Dec 2023 02:11:19 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Mar 2024 06:41:17 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 11:33:35 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Suzuki",
"Fumika",
""
],
[
"Zurek",
"Wojciech H.",
""
]
] |
2312.02252 | Xiaoqian Shen | Xiaoqian Shen and Mohamed Elhoseiny | StoryGPT-V: Large Language Models as Consistent Story Visualizers | Accepted to CVPR 2025; Project page:
https://xiaoqian-shen.github.io/StoryGPT-V | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recent generative models have demonstrated impressive capabilities in
generating realistic and visually pleasing images grounded on textual prompts.
Nevertheless, a significant challenge remains in applying these models for the
more intricate task of story visualization. Since it requires resolving
pronouns (he, she, they) in the frame descriptions, i.e., anaphora resolution,
and ensuring consistent characters and background synthesis across frames. Yet,
the emerging Large Language Model (LLM) showcases robust reasoning abilities to
navigate through ambiguous references and process extensive sequences.
Therefore, we introduce \emph{StoryGPT-V}, which leverages the merits of the
latent diffusion (LDM) and LLM to produce images with consistent and
high-quality characters grounded on given story descriptions. First, we train a
character-aware LDM, which takes character-augmented semantic embedding as
input and includes the supervision of the cross-attention map using character
segmentation masks, aiming to enhance character generation accuracy and
faithfulness. In the second stage, we enable an alignment between the output of
LLM and the character-augmented embedding residing in the input space of the
first-stage model. This harnesses the reasoning ability of LLM to address
ambiguous references and the comprehension capability to memorize the context.
We conduct comprehensive experiments on two visual story visualization
benchmarks. Our model reports superior quantitative results and consistently
generates accurate characters of remarkable quality with low memory
consumption. Our code is publicly available at:
\href{https://xiaoqian-shen.github.io/StoryGPT-V}{https://xiaoqian-shen.github.io/StoryGPT-V}.
| [
{
"version": "v1",
"created": "Mon, 4 Dec 2023 18:14:29 GMT"
},
{
"version": "v2",
"created": "Wed, 13 Dec 2023 11:21:32 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 12:47:21 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Shen",
"Xiaoqian",
""
],
[
"Elhoseiny",
"Mohamed",
""
]
] |
2312.03116 | Charles Puelz | Dan Lior, Craig G. Rusin, Justin Weigand, Kristina V. Montez, Yimo
Wang, Silvana Molossi, Daniel J. Penny, Charles Puelz | Deformable Registration of MRA and 4D Flow Images to Facilitate Accurate
Estimation of Flow Properties within Blood Vessels | null | null | null | null | q-bio.QM physics.med-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A method is presented for the registration of MRA and 4D Flow images, with
the goal of calculating blood flow properties using both modalities
simultaneously. In particular, the method produces an alignment of
segmentations of vessel networks, from MRA images, with the blood velocity
field within those networks, from the corresponding 4D Flow images. The
alignment procedure is driven by the registration of centerlines of vessels
extracted from the two modalities. Our approach is robust to noise, small
deformations, and partial omissions of vessel surfaces and/or blood velocities.
The alignment procedure is tested on 7 patient data sets acquired at Texas
Children's Hospital. The quality of the resulting alignment is assessed by (i)
an illustration of the aligned and unaligned surface segmentations for a sample
patient, (ii) histograms of distances between centerline networks, and (iii)
graphs of estimated blood flow. For each of the 7 analyzed data sets, medians
of the distance histograms decreased an average of 83.5%, and the estimated
blood flow increased significantly as a result of the alignment procedure.
| [
{
"version": "v1",
"created": "Tue, 5 Dec 2023 20:24:39 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 12:42:06 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Lior",
"Dan",
""
],
[
"Rusin",
"Craig G.",
""
],
[
"Weigand",
"Justin",
""
],
[
"Montez",
"Kristina V.",
""
],
[
"Wang",
"Yimo",
""
],
[
"Molossi",
"Silvana",
""
],
[
"Penny",
"Daniel J.",
""
],
[
"Puelz",
"Charles",
""
]
] |
2312.08868 | A Daassou | Hayat Laassiri, Ahmed Daassou, Rachid Benbrik | Responses of AdS black holes to the collective influence of quintessence
and string cloud | 17 pages, 4 figures. Accepted in International Journal of Theoretical
Physics, 2025 | International Journal of Theoretical Physics, 2025 | 10.1007/s10773-025-05918-6 | null | hep-th gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This research paper explores the interplay between quintessence and a cloud
of strings, focusing on their influence on critical points, behaviors, and
fractional-order phase transitions in AdS black holes. We analyze the
thermodynamic properties of AdS black holes surrounded by quintessence and a
cloud of strings, with particular attention to the effects of thermal
fluctuations on their thermodynamics. Using critical conditions, we derive
approximate analytical expressions for the critical points applicable to all
types of these black holes. We further study the impact of the quintessence
parameter and the intensity of the cloud of strings on critical behaviors,
employing three-dimensional visualizations to depict these effects for
different values of \(\omega_q\). Our results show how these factors affect the
locations of phase transitions and the regions where both phases coexist. In
addition, we examine fractional-order phase transitions, which reveal
significant transformations in the black hole system.
| [
{
"version": "v1",
"created": "Thu, 14 Dec 2023 12:44:28 GMT"
},
{
"version": "v2",
"created": "Mon, 25 Dec 2023 09:20:18 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 10:04:47 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Laassiri",
"Hayat",
""
],
[
"Daassou",
"Ahmed",
""
],
[
"Benbrik",
"Rachid",
""
]
] |
2312.09314 | Ivan Matchenya | Ivan Matchenya, Anton Khanas, Roman Podgornyi, Daniil Shirkin, Alexey
Ekgardt, Nikita Sizykh, Sergey Anoshkin, Dmitry V. Krasnikov, Alexei Yulin,
Alexey Zhukov, Albert G. Nasibulin, Ivan Scheblykin, Anatoly Pushkarev,
Andrei Zenkevich, Juan Bisquert, Alexandr Marunchenko | Biorealistic Response in Optoelectrically-Driven Flexible
Halide-Perovskite Single-Crystal Memristors | null | null | null | null | cond-mat.mtrl-sci physics.app-ph physics.optics | http://creativecommons.org/licenses/by/4.0/ | The transition to smart wearable and flexible optoelectronic devices
communicating with each other and performing neuromorphic computing at the edge
is a big goal in next-generation optoelectronics. These devices should perform
their regular tasks supported by energy-efficient in-memory calculations. Here,
we study the response of the CsPbBr$_3$ halide-perovskite single crystal
fabricated on the flexible polymer substrate and integrated with the
single-walled carbon nanotube thin film electrodes in a lateral geometry. We
show both photodetection functions combined with the synaptic functionality in
our device under the application of hybrid optoelectrical stimuli. Furthermore,
we demonstrate that our device exhibits frequency-dependent bidirectional
modification of synaptic weight with a sliding threshold similar to
biologically plausible Bienenstock-Cooper-Munro learning. The demonstrated
optoelectronic synaptic behavior in halide-perovskite single-crystals opens the
opportunity for the development of hybrid organic-inorganic artificial visual
systems.
| [
{
"version": "v1",
"created": "Thu, 14 Dec 2023 19:47:29 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 14:57:55 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Matchenya",
"Ivan",
""
],
[
"Khanas",
"Anton",
""
],
[
"Podgornyi",
"Roman",
""
],
[
"Shirkin",
"Daniil",
""
],
[
"Ekgardt",
"Alexey",
""
],
[
"Sizykh",
"Nikita",
""
],
[
"Anoshkin",
"Sergey",
""
],
[
"Krasnikov",
"Dmitry V.",
""
],
[
"Yulin",
"Alexei",
""
],
[
"Zhukov",
"Alexey",
""
],
[
"Nasibulin",
"Albert G.",
""
],
[
"Scheblykin",
"Ivan",
""
],
[
"Pushkarev",
"Anatoly",
""
],
[
"Zenkevich",
"Andrei",
""
],
[
"Bisquert",
"Juan",
""
],
[
"Marunchenko",
"Alexandr",
""
]
] |
2312.12176 | Manuel Silva | Jose L. G\'omez, Manuel Silva, Antonio Seoane, Agn\`es Borr\'as, Mario
Noriega, Germ\'an Ros, Jose A. Iglesias-Guitian, Antonio M. L\'opez | All for One, and One for All: UrbanSyn Dataset, the third Musketeer of
Synthetic Driving Scenes | The UrbanSyn Dataset is available in http://urbansyn.org/ | Neurocomputing volume 637 page 130038, 2025 | 10.1016/j.neucom.2025.130038 | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce UrbanSyn, a photorealistic dataset acquired through
semi-procedurally generated synthetic urban driving scenarios. Developed using
high-quality geometry and materials, UrbanSyn provides pixel-level ground
truth, including depth, semantic segmentation, and instance segmentation with
object bounding boxes and occlusion degree. It complements GTAV and Synscapes
datasets to form what we coin as the 'Three Musketeers'. We demonstrate the
value of the Three Musketeers in unsupervised domain adaptation for image
semantic segmentation. Results on real-world datasets, Cityscapes, Mapillary
Vistas, and BDD100K, establish new benchmarks, largely attributed to UrbanSyn.
We make UrbanSyn openly and freely accessible (www.urbansyn.org).
| [
{
"version": "v1",
"created": "Tue, 19 Dec 2023 14:09:12 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 11:35:12 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Gómez",
"Jose L.",
""
],
[
"Silva",
"Manuel",
""
],
[
"Seoane",
"Antonio",
""
],
[
"Borrás",
"Agnès",
""
],
[
"Noriega",
"Mario",
""
],
[
"Ros",
"Germán",
""
],
[
"Iglesias-Guitian",
"Jose A.",
""
],
[
"López",
"Antonio M.",
""
]
] |
2312.17545 | Amar Kumar | Amar Kumar, Sujeet Chaudhary and Sharat Chandra | Effect of Point Defects and Lattice Distortions on the Structural,
Electronic, and Magnetic properties of Co$_2$MnAl Heusler alloy | null | Physical Review Materials 8 (3), 034405 (2024) | 10.1103/PhysRevMaterials.8.034405 | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The effects of various point defects and lattice distortions on the
structural, electronic, and magnetic properties of Co$_2$MnAl alloy are
investigated using density functional theory calculations. For the point
defects, six types of binary antisites, three types of ternary antisites, and
three kinds of vacancies have been simulated with different disorder degrees,
up to a maximum of 12.50%. For the lattice distortions, cubic strain within
-10% $\leq$$\Delta{V/V_0}$$\leq$ 10% (corresponding to 5.50\r{A} $\leq$ a
$\leq$5.88\r{A}) and tetragonal distortions with
0.5$\leq$$\textit{c/a}$$\leq$1.5 at three different unit-cell volumes -
$\textit{V}_0$ and ($\textit{V}_0$$\pm5$%$\textit{V}_0$) have been considered.
The Co$_{Al}$ and Mn$_{Al}$ binary antisite disordered structures (namely,
Co$_{2.0625}$MnAl$_{0.9375}$, Co$_{2.125}$MnAl$_{0.875}$,
Co$_2$Mn$_{1.0625}$Al$_{0.9375}$ and Co$_2$Mn$_{1.125}$Al$_{0.875}$) and
(Co$_{Al}$+Mn$_{Al}$) ternary antisite disordered structure
(Co$_{2.0625}$Mn$_{1.0625}$Al$_{0.875}$) exhibit perfect half-metallicity. The
rest of the antisite disorders have a marginal effect on the half-metallic
properties of Co$_2$MnAl, along with high spin polarization ($\textit{P}$
$\geq$ 70%) and nearly same magnetization ($\textit{M$_s$}$) as that for ideal
structure. Conversely, the vacancy defects significantly affect the electronic
and magnetic properties. The cubic strained structures exhibit high
$\textit{P}$ and constant $\textit{M$_s$}$. Under negative strain within -10%
$\leq$$\Delta{V/V_0}$$\leq$ -7% (for 5.50\r{A} $\leq$ a $\leq$ 5.58\r{A}), the
strained structures have perfect half-metallicity. On the other hand,
tetragonal distortions lead to significant degradation in half-metallic
behavior, except for small distortion values $\Delta{c/a}$, irrespective of
their volume.
| [
{
"version": "v1",
"created": "Fri, 29 Dec 2023 10:30:50 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Kumar",
"Amar",
""
],
[
"Chaudhary",
"Sujeet",
""
],
[
"Chandra",
"Sharat",
""
]
] |
2401.00867 | Borja Aizpurua Altuna | Borja Aizpurua, Samuel Palmer, Roman Orus | Tensor Networks for Explainable Machine Learning in Cybersecurity | 12 pages, 9 figures, 3 tables | Neurocomputing, Volume 639, 28 July 2025, Article 130211 | 10.1016/j.neucom.2025.130211 | null | cs.LG cs.AI quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we show how tensor networks help in developing explainability
of machine learning algorithms. Specifically, we develop an unsupervised
clustering algorithm based on Matrix Product States (MPS) and apply it in the
context of a real use-case of adversary-generated threat intelligence. Our
investigation proves that MPS rival traditional deep learning models such as
autoencoders and GANs in terms of performance, while providing much richer
model interpretability. Our approach naturally facilitates the extraction of
feature-wise probabilities, Von Neumann Entropy, and mutual information,
offering a compelling narrative for classification of anomalies and fostering
an unprecedented level of transparency and interpretability, something
fundamental to understand the rationale behind artificial intelligence
decisions.
| [
{
"version": "v1",
"created": "Fri, 29 Dec 2023 22:35:45 GMT"
},
{
"version": "v2",
"created": "Fri, 5 Jan 2024 11:40:56 GMT"
},
{
"version": "v3",
"created": "Fri, 5 Apr 2024 09:15:05 GMT"
},
{
"version": "v4",
"created": "Fri, 25 Apr 2025 14:56:42 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Aizpurua",
"Borja",
""
],
[
"Palmer",
"Samuel",
""
],
[
"Orus",
"Roman",
""
]
] |
2401.05100 | Ryuta Moriyasu | Ryuta Moriyasu, Sho Kawaguchi, Kenji Kashima | Sampled-Data Primal-Dual Gradient Dynamics in Model Predictive Control | null | null | null | null | eess.SY cs.SY math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Model Predictive Control (MPC) is a versatile approach capable of
accommodating diverse control requirements that holds significant promise for a
broad spectrum of industrial applications. Noteworthy challenges associated
with MPC include the substantial computational burden, which is sometimes
considered excessive even for linear systems. Recently, a rapid computation
method that guides the input toward convergence with the optimal control
problem solution by employing primal-dual gradient (PDG) dynamics as a
controller has been proposed for linear MPCs. However, stability has been
ensured under the assumption that the controller is a continuous-time system,
leading to potential instability when the controller undergoes discretization
and is implemented as a sampled-data system. In this paper, we propose a
discrete-time dynamical controller, incorporating specific modifications to the
PDG approach, and present stability conditions relevant to the resulting
sampled-data system. Additionally, we introduce an extension designed to
enhance control performance, that was traded off in the original. Numerical
examples substantiate that our proposed method, which can be executed in only 1
$\mu$s in a standard laptop, not only ensures stability with considering
sampled-data implementation but also effectively enhances control performance.
| [
{
"version": "v1",
"created": "Wed, 10 Jan 2024 12:05:18 GMT"
},
{
"version": "v2",
"created": "Mon, 7 Apr 2025 02:48:05 GMT"
},
{
"version": "v3",
"created": "Thu, 24 Apr 2025 21:39:55 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Moriyasu",
"Ryuta",
""
],
[
"Kawaguchi",
"Sho",
""
],
[
"Kashima",
"Kenji",
""
]
] |
2401.05242 | Yuan Zhang | Yifei Pan, Yuan Zhang | On solutions to $-\Delta u = V u$ near infinity | 19 pages, accepted at Communications in Partial Differential
Equations | null | 10.1080/03605302.2025.2485479 | null | math.AP math.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the unique continuation property and the sign changing
behavior of weak solutions to $-\Delta u =Vu$ near infinity under certain
conditions on the blow-up rate of the potential $V$ near infinity.
| [
{
"version": "v1",
"created": "Wed, 10 Jan 2024 16:15:03 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2025 23:01:02 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Pan",
"Yifei",
""
],
[
"Zhang",
"Yuan",
""
]
] |
2401.07807 | Fabian Heimann | Fabian Heimann | A Higher Order Unfitted Space-Time Finite Element Method for Coupled
Surface-Bulk problems | null | null | 10.1007/978-3-031-86173-4_43 | null | math.NA cs.NA | http://creativecommons.org/licenses/by/4.0/ | We present a higher order space-time unfitted finite element method for
convection-diffusion problems on coupled (surface and bulk) domains. In that
way, we combine a method suggested by Heimann, Lehrenfeld, Preu{\ss} (SIAM J.
Sci. Comput. 45(2), 2023, B139 - B165) for the bulk case with a method
suggested by Sass, Reusken (Comput. Math. Appl. 146(15), 2023, 253-270) for the
surface case. The geometry is allowed to change with time, and the higher order
discrete approximation of this geometry is ensured by a time-dependent
isoparametric mapping. The space-time discretisation approach allows for
straightforward handling of arbitrary high orders. In that way, we also
generalise results of Hansbo, Larson, Zahedi (Comput. Methods Appl. Mech.
Engrg. 307, 2016, 96-116) to higher orders. The convergence of the proposed
higher order discretisations is confirmed numerically.
| [
{
"version": "v1",
"created": "Mon, 15 Jan 2024 16:25:53 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Heimann",
"Fabian",
""
]
] |
2401.08676 | Qiyun Cheng | Qiyun Cheng, Huihua Yang, Shanbin Shi, and Wei Ji | A Data-Driven Based Concurrent Coupling Approach for Cryogenic
Propellant Tank Long-term Pressure Control Predictions | null | null | null | null | physics.flu-dyn physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The design and optimization of cryogenic propellant storage tanks for NASA's
future space missions require fast and accurate predictions of long-term fluid
behaviors. Computational fluid dynamics (CFD) techniques are high-fidelity but
computationally prohibitive. Coarse mesh nodal techniques are fast but heavily
rely on empirical correlations to capture prominent three-dimensional
phenomena. A data-driven based concurrent coupling (DCC) approach has been
developed to integrate CFD with nodal techniques for efficient and accurate
analysis. This concurrent coupling scheme generates case-specific correlations
on the fly through a short period of co-solving CFD and nodal simulations,
followed by a long-period nodal simulation with CFD-corrected solutions. This
paper presents the approach development, stability analysis, and efficiency
demonstration, specifically for modeling two-phase cryogenic propellant tank
self-pressurization and periodic mixing phenomena. Linear regression is
employed to derive the data-driven correlations. The self-pressurization
experiments of Multipurpose Hydrogen Test Bed experiments and K-Site tank are
used to validate the approach. The DCC approach accurately predicts temperature
stratifications while reducing computational time by as much as 70% compared to
pure CFD simulations. Additionally, the DCC approach mitigates the risks of
numerical instability and correlation loss inherent in current domain
decomposition or overlapping-based coupling methods, making it a flexible and
user-friendly approach for integrated CFD and nodal analysis of cryogenic tank
behaviors.
| [
{
"version": "v1",
"created": "Wed, 10 Jan 2024 15:29:35 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 01:35:29 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Cheng",
"Qiyun",
""
],
[
"Yang",
"Huihua",
""
],
[
"Shi",
"Shanbin",
""
],
[
"Ji",
"Wei",
""
]
] |
2401.10239 | Brenner Rego | Brenner S. Rego, Davide M. Raimondo, Guilherme V. Raffo | Line zonotopes: A tool for state estimation and fault diagnosis of
unbounded and descriptor systems | 15 pages, 6 figures. Revised manuscript v2 includes a new name for
the set representation, revised article structure, a new numerical example,
and several minor modifications. Theoretical results unchanged. Revised
manuscript v3 includes several clarifications on the contributions of the
paper. Revised manuscript v4 has a shorter title. arXiv admin note: text
overlap with arXiv:2306.07369 | null | null | null | eess.SY cs.SY math.DS | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This paper proposes new methods for set-based state estimation and active
fault diagnosis (AFD) of linear descriptor systems (LDS). Unlike intervals,
ellipsoids, and zonotopes, constrained zonotopes (CZs) can directly incorporate
linear static constraints on state variables - typical of descriptor systems -
into their mathematical representation, leading to less conservative
enclosures. However, for LDS that are unstable or not fully observable, a
bounded representation cannot ensure a valid enclosure of the states over time.
To address this limitation, we introduce line zonotopes, a new representation
for unbounded sets that retains key properties of CZs, including polynomial
time complexity reduction methods, while enabling the description of strips,
hyperplanes, and the entire n-dimensional Euclidean space. This extension not
only generalizes the use of CZs to unbounded settings but can also enhance
set-based estimation and AFD in both stable and unstable scenarios.
Additionally, we extend the AFD method for LDS from Rego et al. (2020) to
operate over reachable tubes rather than solely on the reachable set at the
final time of the considered horizon. This reduces conservatism in input
separation and enables more accurate fault diagnosis based on the entire output
sequence. The advantages of the proposed methods over existing CZ-based
approaches are demonstrated through numerical examples.
| [
{
"version": "v1",
"created": "Mon, 27 Nov 2023 17:55:31 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Aug 2024 19:29:24 GMT"
},
{
"version": "v3",
"created": "Tue, 18 Mar 2025 22:26:05 GMT"
},
{
"version": "v4",
"created": "Thu, 24 Apr 2025 20:05:08 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Rego",
"Brenner S.",
""
],
[
"Raimondo",
"Davide M.",
""
],
[
"Raffo",
"Guilherme V.",
""
]
] |
2401.11515 | Tsung-Hung Yao | Tsung-Hung Yao and Zhenke Wu and Karthik Bharath and Veerabhadran
Baladandayuthapani | Geometry-driven Bayesian Inference for Ultrametric Covariance Matrices | null | null | null | null | stat.ME | http://creativecommons.org/licenses/by/4.0/ | Ultrametric matrices are a class of covariance matrices that arise in latent
tree models. As a parameter space in a statistical model, the set of
ultrametric matrices is neither convex nor a smooth manifold. Focus in the
literature has hitherto been restricted to estimation through projections and
relaxation-based techniques, and inferential methods are lacking. Motivated by
this, we establish a bijection between the set of positive definite ultrametric
matrices and the set of rooted, leaf-labeled trees equipped with the stratified
geometry of the well-known phylogenetic treespace. Using the pullback geometry
under the bijection and by adapting sampling algorithms in Bayesian
phylogenetics, we develop algorithms to sample from the posterior distribution
on the set of ultrametric matrices in a Bayesian latent tree model where the
tree may be binary or multifurcating. We demonstrate the utility of the
algorithms in simulation studies, and illustrate them on a pre-clinical cancer
application to quantify uncertainty about treatment trees that identify
treatments with high mechanism similarity that target correlated pathways.
| [
{
"version": "v1",
"created": "Sun, 21 Jan 2024 15:08:27 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2025 22:04:21 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Yao",
"Tsung-Hung",
""
],
[
"Wu",
"Zhenke",
""
],
[
"Bharath",
"Karthik",
""
],
[
"Baladandayuthapani",
"Veerabhadran",
""
]
] |
2401.12012 | Mengdi Wang | Mengdi Wang, Anna Bodonhelyi, Efe Bozkir, Enkelejda Kasneci | TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for
Lazy Clients | Proceedings of the AAAI Conference on Artificial Intelligence 2024
(AAAI'24) | Proceedings of the AAAI Conference on Artificial Intelligence 2024
(AAAI'24) | 10.1609/aaai.v38i14.29481 | null | cs.LG cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning is a distributed collaborative machine learning paradigm
that has gained strong momentum in recent years. In federated learning, a
central server periodically coordinates models with clients and aggregates the
models trained locally by clients without necessitating access to local data.
Despite its potential, the implementation of federated learning continues to
encounter several challenges, predominantly the slow convergence that is
largely due to data heterogeneity. The slow convergence becomes particularly
problematic in cross-device federated learning scenarios where clients may be
strongly limited by computing power and storage space, and hence counteracting
methods that induce additional computation or memory cost on the client side
such as auxiliary objective terms and larger training iterations can be
impractical. In this paper, we propose a novel federated aggregation strategy,
TurboSVM-FL, that poses no additional computation burden on the client side and
can significantly accelerate convergence for federated classification task,
especially when clients are "lazy" and train their models solely for few epochs
for next global aggregation. TurboSVM-FL extensively utilizes support vector
machine to conduct selective aggregation and max-margin spread-out
regularization on class embeddings. We evaluate TurboSVM-FL on multiple
datasets including FEMNIST, CelebA, and Shakespeare using user-independent
validation with non-iid data distribution. Our results show that TurboSVM-FL
can significantly outperform existing popular algorithms on convergence rate
and reduce communication rounds while delivering better test metrics including
accuracy, F1 score, and MCC.
| [
{
"version": "v1",
"created": "Mon, 22 Jan 2024 14:59:11 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Jan 2024 17:27:10 GMT"
},
{
"version": "v3",
"created": "Mon, 29 Jan 2024 17:55:56 GMT"
},
{
"version": "v4",
"created": "Sun, 11 Feb 2024 12:09:51 GMT"
},
{
"version": "v5",
"created": "Thu, 19 Dec 2024 16:15:27 GMT"
},
{
"version": "v6",
"created": "Fri, 25 Apr 2025 11:16:13 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Wang",
"Mengdi",
""
],
[
"Bodonhelyi",
"Anna",
""
],
[
"Bozkir",
"Efe",
""
],
[
"Kasneci",
"Enkelejda",
""
]
] |
2401.12374 | Jordi Canela | Jordi Canela, Antonio Garijo, Pascale Roesch | Connected McMullen-like Julia sets in a Chebyshev-Halley Family | Minor modifications were made to the text and Figure 9 was added
following the referees' suggestions | SIAM J. Appl. Dyn. Syst., Vol. 24, No. 2, pp. 1289--1315 (2025) | 10.1137/24M1648090 | null | math.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we study a one parameter family of rational maps obtained by
applying the Chebyshev-Halley root finding algorithms. We show that the
dynamics near parameters where the family presents some degeneracy might be
understood from the point of view of singular perturbations. More precisely, we
relate the dynamics of those maps with the one of the McMullen family
$M_{\lambda}(z)=z^4 + \lambda /z^2$, using quasi-conformal surgery.
| [
{
"version": "v1",
"created": "Mon, 22 Jan 2024 21:55:41 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 17:30:31 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Canela",
"Jordi",
""
],
[
"Garijo",
"Antonio",
""
],
[
"Roesch",
"Pascale",
""
]
] |
2401.16751 | Matthias Frey | Matthias Frey and Igor Bjelakovi\'c and Michael C. Gastpar and Jingge
Zhu | Simultaneous Computation and Communication over MAC | null | IEEE Transactions on Information Theory (Early Access), vol.
71(5), 2025 | 10.1109/TIT.2025.3542673 | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study communication over a Gaussian multiple-access channel (MAC) with two
types of transmitters: Digital transmitters hold a message from a discrete set
that needs to be communicated to the receiver with vanishing error probability.
Analog transmitters hold sequences of analog values. Some functions of these
distributed values (but not the values themselves) need to be conveyed to the
receiver, subject to a fidelity criterion such as mean squared error (MSE) or a
certain maximum error with given confidence. For the case in which the computed
function for the analog transmitters is a sum of values in [-1,1], we derive
inner and outer bounds for the tradeoff of digital and analog rates of
communication under peak and average power constraints for digital transmitters
and a peak power constraint for analog transmitters. We then extend the
achievability result to a class of functions that includes all linear and some
non-linear functions. This extended scheme works over fading channels as long
as full channel state information is available at the transmitter. The
practicality of our proposed communication scheme is shown in channel
simulations that use a version of the scheme based on low density parity check
(LDPC) coding. We evaluate the system performance for different block lengths
and Gaussian as well as non-Gaussian noise distributions.
| [
{
"version": "v1",
"created": "Tue, 30 Jan 2024 05:10:40 GMT"
},
{
"version": "v2",
"created": "Tue, 23 Apr 2024 10:28:36 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Nov 2024 00:20:11 GMT"
},
{
"version": "v4",
"created": "Thu, 13 Feb 2025 21:33:44 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Frey",
"Matthias",
""
],
[
"Bjelaković",
"Igor",
""
],
[
"Gastpar",
"Michael C.",
""
],
[
"Zhu",
"Jingge",
""
]
] |
2401.16927 | Gerhard Roehrle | Michael Bate, S\"oren B\"ohm, Alastair Litterick, Benjamin Martin,
Gerhard Roehrle | $G$-complete reducibility and saturation | 15 pages; v2 minor changes; v3 18 pages, various changes; new is
Proposition 4.8 which shows that saturation is compatible with standard
Frobenius endomorphisms; v4, 19 pages, introduction rewritten, substantial
reorganization of material; v5, minor changes following referee's suggestion;
final version to appear in PJM; v6 minor changes to conform with published
version, bibliographic updates | null | null | null | math.RT math.GR | http://creativecommons.org/licenses/by/4.0/ | Let $H \subseteq G$ be connected reductive linear algebraic groups defined
over an algebraically closed field of characteristic $p> 0$. In our first main
theorem we show that if a closed subgroup $K$ of $H$ is $H$-completely
reducible, then it is also $G$-completely reducible in the sense of Serre,
under some restrictions on $p$, generalising the known case for $G = GL(V)$.
Our proof uses R.W. Richardson's notion of reductive pairs to reduce to the
$GL(V)$ case. We study Serre's notion of saturation and prove that saturation
behaves well with respect to products and regular subgroups. Our second main
theorem shows that if $K$ is $H$-completely reducible, then the saturation of
$K$ in $G$ is completely reducible in the saturation of $H$ in $G$ (which is
again a connected reductive subgroup of $G$), under suitable restrictions on
$p$, again generalising the known instance for $G = GL(V)$. We also study
saturation of finite subgroups of Lie type in $G$. We show that saturation is
compatible with standard Frobenius endomorphisms, and we use this to generalise
a result due to Nori from 1987 in case $G = GL(V)$.
| [
{
"version": "v1",
"created": "Tue, 30 Jan 2024 11:52:27 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Feb 2024 17:51:07 GMT"
},
{
"version": "v3",
"created": "Wed, 27 Mar 2024 08:03:55 GMT"
},
{
"version": "v4",
"created": "Tue, 30 Jul 2024 16:51:17 GMT"
},
{
"version": "v5",
"created": "Thu, 3 Apr 2025 09:10:04 GMT"
},
{
"version": "v6",
"created": "Fri, 25 Apr 2025 08:05:10 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Bate",
"Michael",
""
],
[
"Böhm",
"Sören",
""
],
[
"Litterick",
"Alastair",
""
],
[
"Martin",
"Benjamin",
""
],
[
"Roehrle",
"Gerhard",
""
]
] |
2401.17488 | Huaisong Zhao PhD | Huaisong Zhao, Rui Han, Ling Qin, Feng Yuan, and Peng Zou | A universal pairing gap measurement proposal by dynamical excitations in
2D doped attractive Fermi-Hubbard model with spin-orbit coupling | 12 pages, 8 figures | Phys. Rev. A 110, 063326 (2024) | 10.1103/PhysRevA.110.063326 | null | cond-mat.quant-gas cond-mat.supr-con | http://creativecommons.org/licenses/by/4.0/ | By calculating dynamical structure factor of two-dimensional doped attractive
Fermi-Hubbard model with Rashba spin-orbit coupling, we not only investigate
collective modes and single-particle excitations of the system during the phase
transition between Bardeen-Cooper-Schrieffer superfluid and topological
superfluid, but also propose a universal method to measure pairing gap
measurement in an optical lattice system. Our numerical results show that the
area of the molecular excitation peak at the transferred momentum ${\bf
q}=\left[\pi,\pi\right]$ is proportional to the square of the pairing gap in
the system with Rashba SOC. In particular, this method is very sensitive to the
pairing gap. This goes on verifying that this method is universal to measure
the pairing gap in a doped optical lattice with Rashba SOC. These theoretical
results are important for experimentally measuring the pairing gap and studying
the topological superfluid in an optical lattice.
| [
{
"version": "v1",
"created": "Tue, 30 Jan 2024 22:47:09 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Zhao",
"Huaisong",
""
],
[
"Han",
"Rui",
""
],
[
"Qin",
"Ling",
""
],
[
"Yuan",
"Feng",
""
],
[
"Zou",
"Peng",
""
]
] |
2401.17929 | Alexander Erlei | Alexander Erlei, Lukas Meub | Reputation-Driven Adoption and Avoidance of Algorithmic Decision Aids in
Credence Goods Markets | null | null | null | null | econ.GN cs.HC q-fin.EC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In credence goods markets such as health care or repair services, consumers
rely on experts with superior information to adequately diagnose and treat
them. Experts, however, are constrained in their diagnostic abilities, which
hurts market efficiency and consumer welfare. Technological breakthroughs that
substitute or complement expert judgments have the potential to alleviate
consumer mistreatment. This article studies how competitive experts adopt novel
diagnostic technologies when skills are heterogeneously distributed and
obfuscated to consumers. We differentiate between novel technologies that
increase expert abilities, and algorithmic decision aids that complement expert
judgments, but do not affect an expert's personal diagnostic precision. When
consumers build up beliefs about an expert's type through repeated
interactions, we show that high-ability experts may strategically forego the
decision aid in order to escape a pooling equilibrium by differentiating
themselves from low-ability experts. Without future visits, signaling concerns
cause all experts to randomize their investment choice, leading to
under-utilization from low-ability experts and over-utilization from
high-ability experts. Results from two online experiments support our
hypotheses. High-ability experts are significantly less likely than low-ability
experts to invests into an algorithmic decision aid if reputation building is
possible. Otherwise, there is no difference, and experts who believe that
consumers play a signaling game randomize their investment choice.
| [
{
"version": "v1",
"created": "Wed, 31 Jan 2024 15:41:37 GMT"
},
{
"version": "v2",
"created": "Mon, 15 Apr 2024 10:45:24 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 10:16:11 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Erlei",
"Alexander",
""
],
[
"Meub",
"Lukas",
""
]
] |
2402.02825 | Tian-Yu Zhao | Tianyu Zhao, Yue Zhou, Ruijun Shi, Zhoujian Cao, Zhixiang Ren | GWAI: Artificial Intelligence Platform for Enhanced Gravitational Wave
Data Analysis | 19 pages, 5 figures | SoftwareX 28, 101930 (2024) | 10.1016/j.softx.2024.101930 | null | gr-qc astro-ph.IM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Gravitational wave (GW) astronomy has opened new frontiers in understanding
the cosmos, while the integration of artificial intelligence (AI) in science
promises to revolutionize data analysis methodologies. However, a significant
gap exists, as there is currently no dedicated platform that enables scientists
to develop, test, and evaluate AI algorithms efficiently for GW data analysis.
To address this gap, we introduce GWAI, a pioneering AI-centered software
platform designed for GW data analysis. GWAI contains a three-layered
architecture that emphasizes simplicity, modularity, and flexibility, covering
the entire analysis pipeline. GWAI aims to accelerate scientific discoveries,
bridging the gap between advanced AI techniques and astrophysical research.
| [
{
"version": "v1",
"created": "Mon, 5 Feb 2024 09:15:53 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 07:28:25 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Zhao",
"Tianyu",
""
],
[
"Zhou",
"Yue",
""
],
[
"Shi",
"Ruijun",
""
],
[
"Cao",
"Zhoujian",
""
],
[
"Ren",
"Zhixiang",
""
]
] |
2402.03145 | Robin Str\"asser | Robin Str\"asser, Manuel Schaller, Karl Worthmann, Julian Berberich,
Frank Allg\"ower | SafEDMD: A Koopman-based data-driven controller design framework for
nonlinear dynamical systems | null | null | null | null | eess.SY cs.LG cs.SY math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Koopman operator serves as the theoretical backbone for machine learning
of dynamical control systems, where the operator is heuristically approximated
by extended dynamic mode decomposition (EDMD). In this paper, we propose
SafEDMD, a novel stability- and certificate-oriented EDMD-based controller
design framework. Our approach leverages a reliable surrogate model generated
in a data-driven fashion in order to provide closed-loop guarantees. In
particular, we establish a controller design based on semi-definite programming
with guaranteed stabilization of the underlying nonlinear system. As central
ingredient, we derive proportional error bounds that vanish at the origin and
are tailored to control tasks. We illustrate the developed method by means of
several benchmark examples and highlight the advantages over state-of-the-art
methods.
| [
{
"version": "v1",
"created": "Mon, 5 Feb 2024 16:12:36 GMT"
},
{
"version": "v2",
"created": "Fri, 17 May 2024 12:26:30 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 14:51:40 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Strässer",
"Robin",
""
],
[
"Schaller",
"Manuel",
""
],
[
"Worthmann",
"Karl",
""
],
[
"Berberich",
"Julian",
""
],
[
"Allgöwer",
"Frank",
""
]
] |
2402.03985 | Ossi R\"ais\"a | Ossi R\"ais\"a, Antti Honkela | A Bias-Variance Decomposition for Ensembles over Multiple Synthetic
Datasets | AISTATS 2025 | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Recent studies have highlighted the benefits of generating multiple synthetic
datasets for supervised learning, from increased accuracy to more effective
model selection and uncertainty estimation. These benefits have clear empirical
support, but the theoretical understanding of them is currently very light. We
seek to increase the theoretical understanding by deriving bias-variance
decompositions for several settings of using multiple synthetic datasets,
including differentially private synthetic data. Our theory yields a simple
rule of thumb to select the appropriate number of synthetic datasets in the
case of mean-squared error and Brier score. We investigate how our theory works
in practice with several real datasets, downstream predictors and error
metrics. As our theory predicts, multiple synthetic datasets often improve
accuracy, while a single large synthetic dataset gives at best minimal
improvement, showing that our insights are practically relevant.
| [
{
"version": "v1",
"created": "Tue, 6 Feb 2024 13:20:46 GMT"
},
{
"version": "v2",
"created": "Thu, 23 May 2024 13:35:55 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 10:38:41 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Räisä",
"Ossi",
""
],
[
"Honkela",
"Antti",
""
]
] |
2402.05411 | Zihan Yan | Aron C. Wall, Zihan Yan | Linearised Second Law for Higher Curvature Gravity and Non-Minimally
Coupled Vector Fields | 38 pages, 2 figures, 2 tables; references and comments added in v2 | Phys.Rev.D 110 (2024) 8, 084005 | 10.1103/PhysRevD.110.084005 | null | gr-qc hep-th | http://creativecommons.org/licenses/by/4.0/ | Expanding the work of arXiv:1504.08040, we show that black holes obey a
second law for linear perturbations to bifurcate Killing horizons, in any
covariant higher curvature gravity coupled to scalar and vector fields. The
vector fields do not need to be gauged, and (like the scalars) can have
arbitrary non-minimal couplings to the metric. The increasing entropy has a
natural expression in covariant phase space language, which makes it manifestly
invariant under JKM ambiguities. An explicit entropy formula is given for
f(Riemann) gravity coupled to vectors, where at most one derivative acts on
each vector. Besides the previously known curvature terms, there are three
extra terms involving differentiating the Lagrangian by the symmetric vector
derivative (which therefore vanish for gauge fields).
| [
{
"version": "v1",
"created": "Thu, 8 Feb 2024 05:06:23 GMT"
},
{
"version": "v2",
"created": "Sun, 28 Apr 2024 10:40:53 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Wall",
"Aron C.",
""
],
[
"Yan",
"Zihan",
""
]
] |
2402.05878 | Nicolas Nguyen | Nicolas Nguyen, Imad Aouali, Andr\'as Gy\"orgy, Claire Vernade | Prior-Dependent Allocations for Bayesian Fixed-Budget Best-Arm
Identification in Structured Bandits | null | International Conference on Artificial Intelligence and Statistics
(AISTATS 2025) | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | We study the problem of Bayesian fixed-budget best-arm identification (BAI)
in structured bandits. We propose an algorithm that uses fixed allocations
based on the prior information and the structure of the environment. We provide
theoretical bounds on its performance across diverse models, including the
first prior-dependent upper bounds for linear and hierarchical BAI. Our key
contribution is introducing new proof methods that result in tighter bounds for
multi-armed BAI compared to existing methods. We extensively compare our
approach to other fixed-budget BAI methods, demonstrating its consistent and
robust performance in various settings. Our work improves our understanding of
Bayesian fixed-budget BAI in structured bandits and highlights the
effectiveness of our approach in practical scenarios.
| [
{
"version": "v1",
"created": "Thu, 8 Feb 2024 18:13:26 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2025 17:26:00 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Nguyen",
"Nicolas",
""
],
[
"Aouali",
"Imad",
""
],
[
"György",
"András",
""
],
[
"Vernade",
"Claire",
""
]
] |
2402.09112 | Heather Battey Dr | Jakub Rybak, Heather Battey, Karthik Bharath | Regression graphs and sparsity-inducing reparametrizations | null | null | null | null | math.ST stat.TH | http://creativecommons.org/licenses/by/4.0/ | That parametrization and sparsity are inherently linked raises the
possibility that relevant models, not obviously sparse in their natural
formulation, exhibit a population-level sparsity after reparametrization. In
covariance models, positive-definiteness enforces additional constraints on how
sparsity can legitimately manifest. It is therefore natural to consider
reparametrization maps in which sparsity respects positive definiteness. The
main purpose of this paper is to provide insight into structures on the
physically-natural scale that induce and are induced by sparsity after
reparametrization. The richest of the four structures initially uncovered can
be generated, under a causal ordering, by the joint-response graphs studied by
Wermuth & Cox (2004), while the most restrictive is that induced by sparsity on
the scale of the matrix logarithm, studied by Battey (2017). The Iwasawa
decomposition of the general linear group, combined with the graphical-models
interpretation, points to a class of reparametrizations for the chain-graph
models (Andersson et al. 2001), with undirected and directed acyclic graphs as
special cases. An important insight is the interpretation of approximate zeros,
explaining the modelling implications of enforcing sparsity after
reparameterization: in effect, the relation between two variables would be
declared null if relatively direct regression effects were negligible and
others manifested through long paths. The insights have a bearing on
methodology, some aspects of which are presented. A detailed simulation uses
the theoretical insights to further explore regimes under which
reparametrization is beneficial.
| [
{
"version": "v1",
"created": "Wed, 14 Feb 2024 11:55:16 GMT"
},
{
"version": "v2",
"created": "Mon, 29 Jul 2024 14:07:18 GMT"
},
{
"version": "v3",
"created": "Fri, 28 Mar 2025 17:14:55 GMT"
},
{
"version": "v4",
"created": "Fri, 25 Apr 2025 16:54:59 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Rybak",
"Jakub",
""
],
[
"Battey",
"Heather",
""
],
[
"Bharath",
"Karthik",
""
]
] |
2402.11908 | Sayeh Gholipour Picha | Sayeh Gholipour Picha, Dawood Al Chanti, Alice Caplier | Semantic Textual Similarity Assessment in Chest X-ray Reports Using a
Domain-Specific Cosine-Based Metric | null | null | 10.5220/0012429600003657 | null | cs.CL cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Medical language processing and deep learning techniques have emerged as
critical tools for improving healthcare, particularly in the analysis of
medical imaging and medical text data. These multimodal data fusion techniques
help to improve the interpretation of medical imaging and lead to increased
diagnostic accuracy, informed clinical decisions, and improved patient
outcomes. The success of these models relies on the ability to extract and
consolidate semantic information from clinical text. This paper addresses the
need for more robust methods to evaluate the semantic content of medical
reports. Conventional natural language processing approaches and metrics are
initially designed for considering the semantic context in the natural language
domain and machine translation, often failing to capture the complex semantic
meanings inherent in medical content. In this study, we introduce a novel
approach designed specifically for assessing the semantic similarity between
generated medical reports and the ground truth. Our approach is validated,
demonstrating its efficiency in assessing domain-specific semantic similarity
within medical contexts. By applying our metric to state-of-the-art Chest X-ray
report generation models, we obtain results that not only align with
conventional metrics but also provide more contextually meaningful scores in
the considered medical domain.
| [
{
"version": "v1",
"created": "Mon, 19 Feb 2024 07:48:25 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Picha",
"Sayeh Gholipour",
""
],
[
"Chanti",
"Dawood Al",
""
],
[
"Caplier",
"Alice",
""
]
] |
2402.15459 | Alexander Zyuzin | Alexander A. Zyuzin | Magnetoelectric effect in superconductors with d-wave magnetization | 5 pages, 3 figures | null | null | null | cond-mat.supr-con cond-mat.mes-hall | http://creativecommons.org/licenses/by/4.0/ | We report on a study of the interplay between the supercurrent and
spin-polarization in a two-dimensional superconducting system in the presence
of a d-wave symmetric antiferromagnetic exchange interaction (altermagnetism).
It is demonstrated that the supercurrent exhibits a transverse contribution in
the presence of both constant and momentum-dependent exchange interactions. We
also discuss the analog of the Edelstein effect for such magnetic material,
showing that the induced spin polarization is quadratic in the supercurrent and
d-wave symmetric.
| [
{
"version": "v1",
"created": "Fri, 23 Feb 2024 17:35:20 GMT"
},
{
"version": "v2",
"created": "Fri, 15 Mar 2024 08:16:55 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 10:31:51 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Zyuzin",
"Alexander A.",
""
]
] |
2403.00094 | Oliver Nagy | Luca Avena, Remco van der Hofstad, Frank den Hollander, Oliver Nagy | Mixing of fast random walks on dynamic random permutations | Probab. Theory Relat. Fields (2025) | null | 10.1007/s00440-025-01375-8 | null | math.PR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We analyse the mixing profile of a random walk on a dynamic random
permutation, focusing on the regime where the walk evolves much faster than the
permutation. Two types of dynamics generated by random transpositions are
considered: one allows for coagulation of permutation cycles only, the other
allows for both coagulation and fragmentation. We show that for both types,
after scaling time by the length of the permutation and letting this length
tend to infinity, the total variation distance between the current distribution
and the uniform distribution converges to a limit process that drops down in a
single jump. This jump is similar to a one-sided cut-off, occurs after a random
time whose law we identify, and goes from the value 1 to a value that is a
strictly decreasing and deterministic function of the time of the jump, related
to the size of the largest component in Erd\H{o}s-R\'enyi random graphs. After
the jump, the total variation distance follows this function down to 0.
| [
{
"version": "v1",
"created": "Thu, 29 Feb 2024 19:47:05 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Avena",
"Luca",
""
],
[
"van der Hofstad",
"Remco",
""
],
[
"Hollander",
"Frank den",
""
],
[
"Nagy",
"Oliver",
""
]
] |
2403.00616 | Philipp J. Vetter | Philipp J. Vetter, Thomas Reisser, Maximilian G. Hirsch, Tommaso
Calarco, Felix Motzoi, Fedor Jelezko and Matthias M. M\"uller | Gate-set evaluation metrics for closed-loop optimal control on
nitrogen-vacancy center ensembles in diamond | null | null | 10.1038/s41534-024-00893-y | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A recurring challenge in quantum science and technology is the precise
control of their underlying dynamics that lead to the desired quantum
operations, often described by a set of quantum gates. These gates can be
subject to application-specific errors, leading to a dependence of their
controls on the chosen circuit, the quality measure and the gate-set itself. A
natural solution would be to apply quantum optimal control in an
application-oriented fashion. In turn, this requires the definition of a
meaningful measure of the contextual gate-set performance. Therefore, we
explore and compare the applicability of quantum process tomography, linear
inversion gate-set tomography, randomized linear gate-set tomography, and
randomized benchmarking as measures for closed-loop quantum optimal control
experiments, using a macroscopic ensemble of nitrogen-vacancy centers in
diamond as a test-bed. Our work demonstrates the relative trade-offs between
those measures and how to significantly enhance the gate-set performance,
leading to an improvement across all investigated methods.
| [
{
"version": "v1",
"created": "Fri, 1 Mar 2024 15:42:56 GMT"
},
{
"version": "v2",
"created": "Sun, 24 Mar 2024 12:35:55 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Vetter",
"Philipp J.",
""
],
[
"Reisser",
"Thomas",
""
],
[
"Hirsch",
"Maximilian G.",
""
],
[
"Calarco",
"Tommaso",
""
],
[
"Motzoi",
"Felix",
""
],
[
"Jelezko",
"Fedor",
""
],
[
"Müller",
"Matthias M.",
""
]
] |
2403.11864 | Franz Hinterleitner | F. Hinterleitner | Symmetry-reduced Loop Quantum Gravity: Plane Waves, Flat Space and the
Hamiltonian Constraint | 34 pages, 1 figure | null | null | null | gr-qc | http://creativecommons.org/publicdomain/zero/1.0/ | Loop quantum gravity methods are applied to a symmetry-reduced model with
homogeneity in two dimensions, derived from a Gowdy model [5,6]. The conditions
for propagation of unidirectional plane gravitational waves at exactly the
speed of light are set up in form of null Killing equations in terms of
Ashtekar variables and imposed as operators on quantum states of the system.
Due to the effective one-dimensionality, holonomies and holonomy operators
appear as simple phase factors. In correspondence, state functions might be
considered as U(1) elements with the usual inner product. Under the assumption
of equal spacing of the eigenvalues of geometrical quantities the solutions are
not normalizable in this sense. With decreasing spacing for growing
eigenvalues, as introduced for example in [11], the situation becomes worse.
Taking over the inner product from the genuine gauge group SU(2) of LQG renders
the obtained states normalizable, nevertheless fluctuations of geometrical
quantities remain divergent. In consequence, the solutions of the Killing
conditions are modified, which means allowing for small fluctuations of the
propagation speed, i. e. dispersion of gravitational waves. Vacuum fluctuations
of Minkowski space are sketched. Finally the same methods are applied to the
Hamiltonian constraint with the same result concerning normalizability. With
such a modification also the constraint is not exactly satisfied any more,
which indicates the necessary presence of some kind of interacting matter.
| [
{
"version": "v1",
"created": "Mon, 18 Mar 2024 15:17:44 GMT"
},
{
"version": "v2",
"created": "Wed, 23 Apr 2025 09:12:55 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Hinterleitner",
"F.",
""
]
] |
2403.13790 | Fan Yang | Fan Yang, Hadi Yarloo, Hua-Chen Zhang, Klaus M{\o}lmer, and Anne E. B.
Nielsen | Probing Hilbert Space Fragmentation with Strongly Interacting Rydberg
Atoms | 14 pages, 12 figures | Phys. Rev. B 111, 144313 (2025) | 10.1103/PhysRevB.111.144313 | null | quant-ph cond-mat.quant-gas physics.atom-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hilbert space fragmentation provides a mechanism to break ergodicity in
closed many-body systems. Here, we propose a feasible scheme to explore this
exotic paradigm on a Rydberg quantum simulator. We show that the Rydberg Ising
model in the large detuning regime can be mapped to a generalized folded XXZ
model featuring a strongly fragmented Hilbert space. The emergent Hamiltonian,
however, displays distinct time scales for the transport of a magnon and a hole
excitation. This interesting property facilitates a continuous tuning of the
Krylov-subspace ergodicity, from the integrable regime, to the
Krylov-restricted thermal phase, and eventually to the statistical bubble
localization region. By further introducing nonlocal interactions, we find that
both the fragmentation behavior and the ergodicity of the Krylov subspace can
be significantly enriched. We also examine the role of atomic position
disorders and identify a symmetry-selective many-body localization transition.
We demonstrate that these phenomena manifest themselves in quench dynamics,
which can be readily probed in state-of-the-art Rydberg array setups.
| [
{
"version": "v1",
"created": "Wed, 20 Mar 2024 17:53:20 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 12:23:37 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Yang",
"Fan",
""
],
[
"Yarloo",
"Hadi",
""
],
[
"Zhang",
"Hua-Chen",
""
],
[
"Mølmer",
"Klaus",
""
],
[
"Nielsen",
"Anne E. B.",
""
]
] |
2403.18055 | Mohamed Camil Belhadjoudja | Mohamed Camil Belhadjoudja, Mohamed Maghenem, Emmanuel Witrant,
Christophe Prieur | Adaptive Boundary Control of the Kuramoto-Sivashinsky Equation Under
Intermittent Sensing | null | null | null | null | eess.SY cs.SY math.AP | http://creativecommons.org/licenses/by/4.0/ | We study in this paper boundary stabilization, in the L2 sense, of the
perturbed Kuramoto-Sivashinsky (KS) equation subject to intermittent sensing.
We assume that we measure the state on a given spatial subdomain during certain
time intervals, while we measure the state on the remaining spatial subdomain
during the remaining time intervals. We assign a feedback law at the boundary
of the spatial domain and force to zero the value of the state at the junction
of the two subdomains. Throughout the study, the equation's destabilizing
coefficient is assumed to be unknown and possibly space dependent but bounded.
As a result, adaptive boundary controllers are designed under different
assumptions on the perturbation. In particular, we guarantee input-to-state
stability (ISS) when an upperbound on the perturbation's size is known.
Otherwise, only global uniform ultimate boundedness (GUUB) is guaranteed. In
contrast, when the state is measured at every spatial point all the time (full
state measurement), convergence to an arbitrarily-small neighborhood of the
origin is guaranteed, even if the perturbation's maximal size is unknown.
Numerical simulations are performed to illustrate our results.
| [
{
"version": "v1",
"created": "Tue, 26 Mar 2024 19:17:03 GMT"
},
{
"version": "v2",
"created": "Mon, 4 Nov 2024 23:20:09 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 10:16:49 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Belhadjoudja",
"Mohamed Camil",
""
],
[
"Maghenem",
"Mohamed",
""
],
[
"Witrant",
"Emmanuel",
""
],
[
"Prieur",
"Christophe",
""
]
] |
2403.18403 | Xiuwei Shang | Xiuwei Shang, Guoqiang Chen, Shaoyin Cheng, Shikai Guo, Yanming Zhang,
Weiming Zhang, Nenghai Yu | FoC: Figure out the Cryptographic Functions in Stripped Binaries with
LLMs | 38 pages, 10 figures | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Analyzing the behavior of cryptographic functions in stripped binaries is a
challenging but essential task. Cryptographic algorithms exhibit greater
logical complexity compared to typical code, yet their analysis is unavoidable
in areas such as virus analysis and legacy code inspection. Existing methods
often rely on data or structural pattern matching, leading to suboptimal
generalizability and suffering from manual work. In this paper, we propose a
novel framework called FoC to Figure out the Cryptographic functions in
stripped binaries. In FoC, we first build a binary large language model
(FoC-BinLLM) to summarize the semantics of cryptographic functions in natural
language. The prediction of FoC-BinLLM is insensitive to minor changes, such as
vulnerability patches. To mitigate it, we further build a binary code
similarity model (FoC-Sim) upon the FoC-BinLLM to create change-sensitive
representations and use it to retrieve similar implementations of unknown
cryptographic functions in a database. In addition, we construct a
cryptographic binary dataset for evaluation and to facilitate further research
in this domain. And an automated method is devised to create semantic labels
for extensive binary functions. Evaluation results demonstrate that FoC-BinLLM
outperforms ChatGPT by 14.61% on the ROUGE-L score. FoC-Sim outperforms the
previous best methods with a 52% higher Recall@1. Furthermore, our method also
shows practical ability in virus analysis and 1-day vulnerability detection.
| [
{
"version": "v1",
"created": "Wed, 27 Mar 2024 09:45:33 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Nov 2024 22:02:48 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 16:39:40 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Shang",
"Xiuwei",
""
],
[
"Chen",
"Guoqiang",
""
],
[
"Cheng",
"Shaoyin",
""
],
[
"Guo",
"Shikai",
""
],
[
"Zhang",
"Yanming",
""
],
[
"Zhang",
"Weiming",
""
],
[
"Yu",
"Nenghai",
""
]
] |
2403.19145 | Alexander Sherman | Alexander Sherman | On the Cartan-Helgason theorem for supersymmetric pairs | Corrected typos and corrected error in root system for
(osp(2r|2n),osp(r|2s)x osp(r|2n-2s)) which led to a simplified result and
exposition. Fixed typo in column 2, row 2 of Table 1. Improved exposition in
Sections 2 and 3 | null | null | null | math.RT math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | Let $(\mathfrak{g},\mathfrak{k})$ be a supersymmetric pair arising from a
finite-dimensional, symmetrizable Kac-Moody superalgebra $\mathfrak{g}$. An
important branching problem is to determine the finite-dimensional
highest-weight $\mathfrak{g}$-modules which admit a $\mathfrak{k}$-coinvariant,
and thus appear as functions in a corresponding supersymmetric space
$\mathcal{G}/\mathcal{K}$. This is the super-analogue of the Cartan-Helgason
theorem. We solve this problem via a rank one reduction and an understanding of
reflections in singular roots, which generalize odd reflections in the theory
of Kac-Moody superalgebras. An explicit presentation of spherical weights is
provided for every pair when $\mathfrak{g}$ is indecomposable.
| [
{
"version": "v1",
"created": "Thu, 28 Mar 2024 04:35:58 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 10:13:20 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Sherman",
"Alexander",
""
]
] |
2404.00146 | Maggie Cheng | Huiyuan Yu, Jia He, Maggie Cheng | Fast Orthogonal Matching Pursuit through Successive Regression | null | ICPR 2024 | null | null | cs.CV math.OC | http://creativecommons.org/licenses/by/4.0/ | Orthogonal Matching Pursuit (OMP) has been a powerful method in sparse signal
recovery and approximation. However, OMP suffers computational issues when the
signal has a large number of non-zeros. This paper advances OMP and its
extension called generalized OMP (gOMP) by offering fast algorithms for the
orthogonal projection of the input signal at each iteration. The proposed
modifications directly reduce the computational complexity of OMP and gOMP.
Experiment results verified the improvement in computation time. This paper
also provides sufficient conditions for exact signal recovery. For general
signals with additive noise, the approximation error is at the same order as
OMP (gOMP), but is obtained within much less time.
| [
{
"version": "v1",
"created": "Fri, 29 Mar 2024 20:39:37 GMT"
},
{
"version": "v2",
"created": "Wed, 23 Apr 2025 21:05:28 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 03:26:13 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Yu",
"Huiyuan",
""
],
[
"He",
"Jia",
""
],
[
"Cheng",
"Maggie",
""
]
] |
2404.03818 | Zhangdie Yuan | Zhangdie Yuan, Eric Chamoun, Rami Aly, Chenxi Whitehouse and Andreas
Vlachos | PRobELM: Plausibility Ranking Evaluation for Language Models | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces PRobELM (Plausibility Ranking Evaluation for Language
Models), a benchmark designed to assess language models' ability to discern
more plausible from less plausible scenarios through their parametric
knowledge. While benchmarks such as TruthfulQA emphasise factual accuracy or
truthfulness, and others such as COPA explore plausible scenarios without
explicitly incorporating world knowledge, PRobELM seeks to bridge this gap by
evaluating models' capabilities to prioritise plausible scenarios that leverage
world knowledge over less plausible alternatives. This design allows us to
assess the potential of language models for downstream use cases such as
literature-based discovery where the focus is on identifying information that
is likely but not yet known. Our benchmark is constructed from a dataset
curated from Wikidata edit histories, tailored to align the temporal bounds of
the training data for the evaluated models. PRobELM facilitates the evaluation
of language models across multiple prompting types, including statement, text
completion, and question-answering. Experiments with 10 models of various sizes
and architectures on the relationship between model scales, training recency,
and plausibility performance, reveal that factual accuracy does not directly
correlate with plausibility performance and that up-to-date training data
enhances plausibility assessment across different model architectures.
| [
{
"version": "v1",
"created": "Thu, 4 Apr 2024 21:57:11 GMT"
},
{
"version": "v2",
"created": "Wed, 7 Aug 2024 12:05:06 GMT"
},
{
"version": "v3",
"created": "Thu, 27 Feb 2025 01:42:18 GMT"
},
{
"version": "v4",
"created": "Thu, 24 Apr 2025 21:12:33 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Yuan",
"Zhangdie",
""
],
[
"Chamoun",
"Eric",
""
],
[
"Aly",
"Rami",
""
],
[
"Whitehouse",
"Chenxi",
""
],
[
"Vlachos",
"Andreas",
""
]
] |
2404.04700 | Shuyang Sheng | Jinyong Hahn, John Ham, Geert Ridder, Shuyang Sheng | Stratifying on Treatment Status | null | null | null | null | econ.EM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the estimation of treatment effects using samples stratified by
treatment status. Standard estimators of the average treatment effect and the
local average treatment effect are inconsistent in this setting. We propose
consistent estimators and characterize their asymptotic distributions.
| [
{
"version": "v1",
"created": "Sat, 6 Apr 2024 18:03:20 GMT"
},
{
"version": "v2",
"created": "Mon, 27 Jan 2025 07:57:08 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 02:03:41 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Hahn",
"Jinyong",
""
],
[
"Ham",
"John",
""
],
[
"Ridder",
"Geert",
""
],
[
"Sheng",
"Shuyang",
""
]
] |
2404.04813 | Kory Stiffler | Tyler Grover, Kory Stiffler, and Patrick Vecera | Covariant and Manifestly Projective Invariant Formulation of
Thomas-Whitehead Gravity | 21 pages (less pages due to compact PRD two column format), added
"Solutions" subsection of section 4, made equations more concise, added
references, removed "A" from title, version to be published in Physical
Review D | Phys.Rev.D 110 (2024) 8, 084058 | 10.1103/PhysRevD.110.084058 | null | hep-th gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Thomas-Whitehead (TW) gravity is a recently formulated projectively invariant
extension of Einstein-Hilbert gravity. Projective geometry was used long ago by
Thomas et. al. to succinctly package equivalent paths encoded by the geodesic
equation. Projective invariance in gravity has further origins in string theory
through a geometric action constructed from the method of coadjoint orbits
using the Virasoro Algebra. A projectively invariant connection arises from
this construction, a part of which is known as the diffeomorphism field. TW
gravity exploits projective Gauss-Bonnet terms in the action functional to
endow the diffeomorphism field with dynamics, while allowing the theory to
collapse to general relativity in the limit that the diffeomorphism field
vanishes and the connection becomes Levi-Civita. In the original formulation of
TW gravity, the diffeomorphism field is projectively invariant but not
tensorial and the connection is projectively invariant but not affine. In this
paper we reformulate TW gravity in terms of projectively invariant tensor
fields and a projectively invariant covariant derivative, derive field
equations respecting these symmetries, and show that the field equations
obtained are classically equivalent across formulations. This provides a
'Rosetta Stone' between this newly constructed covariant and projective
invariant formulation of TW gravity and the original formulation that was
manifestly projective invariant, but not covariant.
| [
{
"version": "v1",
"created": "Sun, 7 Apr 2024 05:44:19 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Jun 2024 01:17:22 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Sep 2024 19:51:40 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Grover",
"Tyler",
""
],
[
"Stiffler",
"Kory",
""
],
[
"Vecera",
"Patrick",
""
]
] |
2404.10263 | Yuning Wang | Yuning Wang, Zhiyuan Liu, Haotian Lin, Junkai Jiang, Shaobing Xu,
Jianqiang Wang | PreGSU-A Generalized Traffic Scene Understanding Model for Autonomous
Driving based on Pre-trained Graph Attention Network | 14 pages | null | null | null | cs.CV cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Scene understanding, defined as learning, extraction, and representation of
interactions among traffic elements, is one of the critical challenges toward
high-level autonomous driving (AD). Current scene understanding methods mainly
focus on one concrete single task, such as trajectory prediction and risk level
evaluation. Although they perform well on specific metrics, the generalization
ability is insufficient to adapt to the real traffic complexity and downstream
demand diversity. In this study, we propose PreGSU, a generalized pre-trained
scene understanding model based on graph attention network to learn the
universal interaction and reasoning of traffic scenes to support various
downstream tasks. After the feature engineering and sub-graph module, all
elements are embedded as nodes to form a dynamic weighted graph. Then, four
graph attention layers are applied to learn the relationships among agents and
lanes. In the pre-train phase, the understanding model is trained on two
self-supervised tasks: Virtual Interaction Force (VIF) modeling and Masked Road
Modeling (MRM). Based on the artificial potential field theory, VIF modeling
enables PreGSU to capture the agent-to-agent interactions while MRM extracts
agent-to-road connections. In the fine-tuning process, the pre-trained
parameters are loaded to derive detailed understanding outputs. We conduct
validation experiments on three datasets and two downstream tasks, i.e.,
trajectory prediction in urban scenario and intention recognition in highway
scenario, to verify the model's generalization and understanding capabilities.
Results show that compared with single-task-driven baselines, PreGSU achieves
competitive performance on all datasets and downstream tasks, indicating its
potential to be generalized to various scenes and targets. Ablation study shows
the effectiveness of pre-train task design.
| [
{
"version": "v1",
"created": "Tue, 16 Apr 2024 03:34:35 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 07:37:11 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Wang",
"Yuning",
""
],
[
"Liu",
"Zhiyuan",
""
],
[
"Lin",
"Haotian",
""
],
[
"Jiang",
"Junkai",
""
],
[
"Xu",
"Shaobing",
""
],
[
"Wang",
"Jianqiang",
""
]
] |
2404.10270 | Jeremy Johnathan Williams | Jeremy J. Williams, Felix Liu, Jordy Trilaksono, David Tskhakaya,
Stefan Costea, Leon Kos, Ales Podolnik, Jakub Hromadka, Pratibha Hegde, Marta
Garcia-Gasulla, Valentin Seitz, Frank Jenko, Erwin Laure, Stefano Markidis | Accelerating Particle-in-Cell Monte Carlo Simulations with MPI,
OpenMP/OpenACC and Asynchronous Multi-GPU Programming | Accepted by the Journal of Computational Science (ICCS 2024 Special
Issue) prepared in English, formatted in Springer LNCS template and consists
of 32 pages, which includes the main text, references, and figures | null | 10.1016/j.jocs.2025.102590 | null | cs.DC cs.PF physics.comp-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | As fusion energy devices advance, plasma simulations are crucial for reactor
design. Our work extends BIT1 hybrid parallelization by integrating MPI with
OpenMP and OpenACC, focusing on asynchronous multi-GPU programming. Results
show significant performance gains: 16 MPI ranks plus OpenMP threads reduced
runtime by 53% on a petascale EuroHPC supercomputer, while OpenACC multicore
achieved a 58% reduction. At 64 MPI ranks, OpenACC outperformed OpenMP,
improving the particle mover function by 24%. On MareNostrum 5, OpenACC
async(n) delivered strong performance, but OpenMP asynchronous multi-GPU
approach proved more effective at extreme scaling, maintaining efficiency up to
400 GPUs. Speedup and parallel efficiency (PE) studies revealed OpenMP
asynchronous multi-GPU achieving 8.77x speedup (54.81% PE), surpassing OpenACC
(8.14x speedup, 50.87% PE). While PE declined at high node counts due to
communication overhead, asynchronous execution mitigated scalability
bottlenecks. OpenMP nowait and depend clauses improved GPU performance via
efficient data transfer and task management. Using NVIDIA Nsight tools, we
confirmed BIT1 efficiency for large-scale plasma simulations. OpenMP
asynchronous multi-GPU implementation delivered exceptional performance in
portability, high throughput, and GPU utilization, positioning BIT1 for
exascale supercomputing and advancing fusion energy research. MareNostrum 5
brings us closer to achieving exascale performance.
| [
{
"version": "v1",
"created": "Tue, 16 Apr 2024 03:55:12 GMT"
},
{
"version": "v2",
"created": "Thu, 5 Sep 2024 17:26:38 GMT"
},
{
"version": "v3",
"created": "Fri, 6 Sep 2024 15:12:32 GMT"
},
{
"version": "v4",
"created": "Thu, 24 Apr 2025 20:06:38 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Williams",
"Jeremy J.",
""
],
[
"Liu",
"Felix",
""
],
[
"Trilaksono",
"Jordy",
""
],
[
"Tskhakaya",
"David",
""
],
[
"Costea",
"Stefan",
""
],
[
"Kos",
"Leon",
""
],
[
"Podolnik",
"Ales",
""
],
[
"Hromadka",
"Jakub",
""
],
[
"Hegde",
"Pratibha",
""
],
[
"Garcia-Gasulla",
"Marta",
""
],
[
"Seitz",
"Valentin",
""
],
[
"Jenko",
"Frank",
""
],
[
"Laure",
"Erwin",
""
],
[
"Markidis",
"Stefano",
""
]
] |
2404.11600 | Roman Mauch | Roman Mauch and Lorenzo Ruggeri | Super Yang-Mills on Branched Covers and Weighted Projective Spaces | 22 pages; corrected statements about S1 action and base space | null | 10.1007/JHEP08(2024)106 | null | hep-th | http://creativecommons.org/licenses/by/4.0/ | In this work we conjecture the Coulomb branch partition function, including
flux and instanton contributions, for the $\mathcal{N}=2$ vector multiplet on
weighted projective space $\mathbb{CP}^2_{\boldsymbol{N}}$ for equivariant
Donaldson-Witten and ``Pestun-like'' theories. We claim that this partition
function agrees with the one obtained from dimensional reduction of the 5d
$\mathcal{N}=1$ vector multiplet on a certain branched cover of $S^5$. More
precisely, the branch locus and indices have to be such that they match the
singular locus and deficit angles in $\mathbb{CP}^2_{\boldsymbol{N}}$. Our
conjecture is substantiated by checking that partition functions on spindles
are similarly obtained from dimensional reduction of the 3d $\mathcal{N}=2$
vector multiplet on branched covers of $S^3$. This work paves the way for
obtaining partition functions on more generic symplectic toric orbifolds.
| [
{
"version": "v1",
"created": "Wed, 17 Apr 2024 17:50:31 GMT"
},
{
"version": "v2",
"created": "Wed, 25 Sep 2024 11:35:11 GMT"
},
{
"version": "v3",
"created": "Thu, 24 Apr 2025 20:14:34 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Mauch",
"Roman",
""
],
[
"Ruggeri",
"Lorenzo",
""
]
] |
2404.11796 | Niall Taggart | Matthew Carr and Niall Taggart | Symplectic Weiss calculi | v2: fixed typos and an omitted assumption in Theorem E and improved
exposition. 30 pages. v1: 30 pages. Comments welcome! | null | null | null | math.AT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We provide two candidates for symplectic Weiss calculus based on two
different, but closely related, collections of groups. In the case of the
non-compact symplectic groups, i.e., automorphism groups of vector spaces with
symplectic forms, we show that the calculus deformation retracts onto unitary
calculus as a corollary of the fact that Weiss calculus only depends on the
homotopy type of the groupoid core of the diagram category. In the case of the
compact symplectic groups, i.e., automorphism groups of quaternion vector
spaces, we provide a comparison with the other known versions of Weiss calculus
analogous to the comparisons of calculi of the second named author, and
classify certain stably trivial quaternion vector bundles over finite cell
complexes in a range, using elementary results on convergence of Weiss calculi.
| [
{
"version": "v1",
"created": "Wed, 17 Apr 2024 23:14:46 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 09:58:53 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Carr",
"Matthew",
""
],
[
"Taggart",
"Niall",
""
]
] |
2404.16468 | Bram De Cooman | Bram De Cooman and Johan Suykens | A Dual Perspective of Reinforcement Learning for Imposing Policy
Constraints | Accepted for publication in IEEE Transactions on Artificial
Intelligence | null | null | null | cs.LG cs.AI cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | Model-free reinforcement learning methods lack an inherent mechanism to
impose behavioural constraints on the trained policies. Although certain
extensions exist, they remain limited to specific types of constraints, such as
value constraints with additional reward signals or visitation density
constraints. In this work we unify these existing techniques and bridge the gap
with classical optimization and control theory, using a generic primal-dual
framework for value-based and actor-critic reinforcement learning methods. The
obtained dual formulations turn out to be especially useful for imposing
additional constraints on the learned policy, as an intrinsic relationship
between such dual constraints (or regularization terms) and reward
modifications in the primal is revealed. Furthermore, using this framework, we
are able to introduce some novel types of constraints, allowing to impose
bounds on the policy's action density or on costs associated with transitions
between consecutive states and actions. From the adjusted primal-dual
optimization problems, a practical algorithm is derived that supports various
combinations of policy constraints that are automatically handled throughout
training using trainable reward modifications. The proposed $\texttt{DualCRL}$
method is examined in more detail and evaluated under different (combinations
of) constraints on two interpretable environments. The results highlight the
efficacy of the method, which ultimately provides the designer of such systems
with a versatile toolbox of possible policy constraints.
| [
{
"version": "v1",
"created": "Thu, 25 Apr 2024 09:50:57 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 15:20:59 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"De Cooman",
"Bram",
""
],
[
"Suykens",
"Johan",
""
]
] |
2404.17848 | Fabiano Simas | Fabiano C. Simas and E. da Hora | Vacuum state perturbed by an oscillon: composite configurations in
$\phi^4$ model | published version | Annals Phys. 478 (2025) 170008 | 10.1016/j.aop.2025.170008 | null | hep-th nlin.PS | http://creativecommons.org/licenses/by/4.0/ | We examine the evolution of a vacuum configuration when perturbed by an
oscillon. We consider the $\phi^4$ scenario with a single scalar field only.
For highly excited oscillons, we find that new composite solutions appear. They
are formed by multiple antikink-kink pairs and a centered reminiscent oscillon.
The overall process gives rise to a resonant structure whose pattern resembles
that of a genuine kink-antikink scattering. We then approximate the initial
oscillon by a lumplike profile constructed as a kink-antikink pair. We map the
numerical results previously obtained. The developments proposed here may
contribute to the understanding of the mechanism through which an oscillon
decay reproduces the structure typically related to a kink-antikink collision.
| [
{
"version": "v1",
"created": "Sat, 27 Apr 2024 09:56:03 GMT"
},
{
"version": "v2",
"created": "Mon, 13 May 2024 21:24:15 GMT"
},
{
"version": "v3",
"created": "Thu, 24 Apr 2025 22:06:06 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Simas",
"Fabiano C.",
""
],
[
"da Hora",
"E.",
""
]
] |
2404.18478 | Junping Li | Junping Li and Mixuan Hou | Long-time behaviour of Galton-Watson systems with circular mechanism | 22 pages | null | null | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper concentrates on the limit behavior of discrete-time branching
process with circular mechanism. Three types of limit behaviour of
discrete-time branching process with circular mechanism are given explicitly
under various moment conditions on branching rates. It is proved that the rate
of the first one is geometric, while the other two are supergeometric.
| [
{
"version": "v1",
"created": "Mon, 29 Apr 2024 07:33:05 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 07:52:39 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Li",
"Junping",
""
],
[
"Hou",
"Mixuan",
""
]
] |
2404.19035 | Kevin Griffin | Kevin Patrick Griffin, Ganesh Vijayakumar, Ashesh Sharma, and Michael
A. Sprague | Improved pressure-gradient sensor for the prediction of separation onset
in RANS models | null | Journal of Turbulence, 24 Apr 2025 | 10.1080/14685248.2025.2468224 | null | physics.flu-dyn | http://creativecommons.org/licenses/by/4.0/ | We improve upon two key aspects of the Menter shear stress transport (SST)
turbulence model: (1) We propose a more robust adverse pressure gradient sensor
based on the strength of the pressure gradient in the direction of the local
mean flow; (2) We propose two alternative eddy viscosity models to be used in
the adverse pressure gradient regions identified by our sensor. Direct
numerical simulations of the Boeing Gaussian bump are used to identify the
terms in the baseline SST model that need correction, and a posteriori
Reynolds-averaged Navier-Stokes calculations are used to calibrate coefficient
values, leading to a model that is both physics driven and data informed. The
two sensor-equipped models are applied to two thick airfoils representative of
modern wind turbine applications, the FFA-W3-301 and the DU00-W-212. The
proposed models improve the prediction of stall (onset of separation) with
respect to the prediction of the baseline SST model.
| [
{
"version": "v1",
"created": "Mon, 29 Apr 2024 18:23:00 GMT"
},
{
"version": "v2",
"created": "Fri, 28 Mar 2025 23:26:13 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Griffin",
"Kevin Patrick",
""
],
[
"Vijayakumar",
"Ganesh",
""
],
[
"Sharma",
"Ashesh",
""
],
[
"Sprague",
"Michael A.",
""
]
] |
2404.19689 | Kehan Shi | Kehan Shi, Martin Burger | Continuum limit of $p$-biharmonic equations on graphs | 21 pages | null | null | null | math.AP cs.LG cs.NA math.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper studies the $p$-biharmonic equation on graphs, which arises in
point cloud processing and can be interpreted as a natural extension of the
graph $p$-Laplacian from the perspective of hypergraph. The asymptotic behavior
of the solution is investigated when the random geometric graph is considered
and the number of data points goes to infinity. We show that the continuum
limit is an appropriately weighted $p$-biharmonic equation with homogeneous
Neumann boundary conditions. The result relies on the uniform $L^p$ estimates
for solutions and gradients of nonlocal and graph Poisson equations. The
$L^\infty$ estimates of solutions are also obtained as a byproduct.
| [
{
"version": "v1",
"created": "Tue, 30 Apr 2024 16:29:44 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 15:20:37 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Shi",
"Kehan",
""
],
[
"Burger",
"Martin",
""
]
] |
2405.00495 | Ion Victor Gosea | Athanasios C. Antoulas, Ion Victor Gosea, Charles Poussot-Vassal | On the Loewner framework, the Kolmogorov superposition theorem, and the
curse of dimensionality | 32 pages, 4 figures | null | null | null | math.NA cs.NA cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Loewner framework is an interpolatory approach for the approximation of
linear and nonlinear systems. The purpose here is to extend this framework to
linear parametric systems with an arbitrary number n of parameters. To achieve
this, a new generalized multivariate rational function realization is proposed.
Then, we introduce the n-dimensional multivariate Loewner matrices and show
that they can be computed by solving a set of coupled Sylvester equations. The
null space of these Loewner matrices allows the construction of the
multivariate barycentric rational function. The principal result of this work
is to show how the null space of the n-dimensional Loewner matrix can be
computed using a sequence of 1-dimensional Loewner matrices, leading to a
drastic reduction of the computational burden. Equally importantly, this burden
is alleviated by avoiding the explicit construction of large-scale
n-dimensional Loewner matrices of size $N \times N$. Instead, the proposed
methodology achieves decoupling of variables, leading to (i) a complexity
reduction from $O(N^3)$ to below $O(N^{1.5})$ when $n > 5$ and (ii) to memory
storage bounded by the largest variable dimension rather than their product,
thus taming the curse of dimensionality and making the solution scalable to
very large data sets. This decoupling of the variables leads to a result
similar to the Kolmogorov superposition theorem for rational functions. Thus,
making use of barycentric representations, every multivariate rational function
can be computed using the composition and superposition of single-variable
functions. Finally, we suggest two algorithms (one direct and one iterative) to
construct, directly from data, multivariate (or parametric) realizations
ensuring (approximate) interpolation. Numerical examples highlight the
effectiveness and scalability of the method.
| [
{
"version": "v1",
"created": "Wed, 1 May 2024 13:03:56 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Jun 2024 07:42:28 GMT"
},
{
"version": "v3",
"created": "Tue, 8 Apr 2025 23:17:52 GMT"
},
{
"version": "v4",
"created": "Thu, 24 Apr 2025 21:37:28 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Antoulas",
"Athanasios C.",
""
],
[
"Gosea",
"Ion Victor",
""
],
[
"Poussot-Vassal",
"Charles",
""
]
] |
2405.02134 | Guillem Ram\'irez | Guillem Ram\'irez, Alexandra Birch, Ivan Titov | Optimising Calls to Large Language Models with Uncertainty-Based
Two-Tier Selection | null | First Conference on Language Modeling. COLM 2024. Philadelphia,
Pennsylvania, United States | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Researchers and practitioners operating on a limited budget face the
cost-performance trade-off dilemma. The challenging decision often centers on
whether to use a large LLM with better performance or a smaller one with
reduced costs. This has motivated recent research in the optimisation of LLM
calls. Either a cascading strategy is used, where a smaller LLM or both are
called sequentially, or a routing strategy is used, where only one model is
ever called. Both scenarios are dependent on a decision criterion which is
typically implemented by an extra neural model. In this work, we propose a
simpler solution; we use only the uncertainty of the generations of the small
LLM as the decision criterion. We compare our approach with both cascading and
routing strategies using three different pairs of pre-trained small and large
LLMs, on nine different tasks and against approaches that require an additional
neural model. Our experiments reveal this simple solution optimally balances
cost and performance, outperforming existing methods on 25 out of 27
experimental setups.
| [
{
"version": "v1",
"created": "Fri, 3 May 2024 14:38:59 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Ramírez",
"Guillem",
""
],
[
"Birch",
"Alexandra",
""
],
[
"Titov",
"Ivan",
""
]
] |
2405.02854 | Su Hu | Wanyi Wang, Su Hu, Min-Soo Kim | On gamma functions with respect to the alternating Hurwitz zeta
functions | 21 pages | null | null | null | math.NT math-ph math.CA math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In 2021, Hu and Kim defined a new type of gamma function
$\widetilde{\Gamma}(x)$ from the alternating Hurwitz zeta function
$\zeta_{E}(z,x)$, and obtained some of its properties. In this paper, we shall
further investigate the function $\widetilde{\Gamma}(x)$, that is, we obtain
several properties in analogy to the classical Gamma function $\Gamma(x)$,
including the integral representation, the limit representation, the recursive
formula, the special values, the log-convexity, the duplication and
distribution formulas, and the reflection equation. Furthermore, we also prove
a Lerch-type formula, which shows that the derivative of $\zeta_{E}(z,x)$ can
be representative by $\widetilde\Gamma(x)$.
| [
{
"version": "v1",
"created": "Sun, 5 May 2024 08:58:41 GMT"
},
{
"version": "v2",
"created": "Sun, 26 May 2024 10:43:38 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Jun 2024 07:36:54 GMT"
},
{
"version": "v4",
"created": "Sun, 27 Oct 2024 13:55:54 GMT"
},
{
"version": "v5",
"created": "Fri, 25 Apr 2025 09:04:45 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Wang",
"Wanyi",
""
],
[
"Hu",
"Su",
""
],
[
"Kim",
"Min-Soo",
""
]
] |
2405.04161 | Ricardo Vinuesa | Ricardo Vinuesa, Paola Cinnella, Jean Rabault, Hossein Azizpour,
Stefan Bauer, Bingni W. Brunton, Arne Elofsson, Elias Jarlebring, Hedvig
Kjellstrom, Stefano Markidis, David Marlevi, Javier Garcia-Martinez, Steven
L. Brunton | Decoding complexity: how machine learning is redefining scientific
discovery | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | As modern scientific instruments generate vast amounts of data and the volume
of information in the scientific literature continues to grow, machine learning
(ML) has become an essential tool for organising, analysing, and interpreting
these complex datasets. This paper explores the transformative role of ML in
accelerating breakthroughs across a range of scientific disciplines. By
presenting key examples -- such as brain mapping and exoplanet detection -- we
demonstrate how ML is reshaping scientific research. We also explore different
scenarios where different levels of knowledge of the underlying phenomenon are
available, identifying strategies to overcome limitations and unlock the full
potential of ML. Despite its advances, the growing reliance on ML poses
challenges for research applications and rigorous validation of discoveries. We
argue that even with these challenges, ML is poised to disrupt traditional
methodologies and advance the boundaries of knowledge by enabling researchers
to tackle increasingly complex problems. Thus, the scientific community can
move beyond the necessary traditional oversimplifications to embrace the full
complexity of natural systems, ultimately paving the way for interdisciplinary
breakthroughs and innovative solutions to humanity's most pressing challenges.
| [
{
"version": "v1",
"created": "Tue, 7 May 2024 09:58:02 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 14:35:04 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Vinuesa",
"Ricardo",
""
],
[
"Cinnella",
"Paola",
""
],
[
"Rabault",
"Jean",
""
],
[
"Azizpour",
"Hossein",
""
],
[
"Bauer",
"Stefan",
""
],
[
"Brunton",
"Bingni W.",
""
],
[
"Elofsson",
"Arne",
""
],
[
"Jarlebring",
"Elias",
""
],
[
"Kjellstrom",
"Hedvig",
""
],
[
"Markidis",
"Stefano",
""
],
[
"Marlevi",
"David",
""
],
[
"Garcia-Martinez",
"Javier",
""
],
[
"Brunton",
"Steven L.",
""
]
] |
2405.04366 | James Bateman | Rafa{\l} Gajewski and James Bateman | Backaction suppression in levitated optomechanics using reflective
boundaries | 9 pages with 5 figures, and 5 page supplementary | null | 10.1103/PhysRevResearch.7.023041 | null | physics.optics quant-ph | http://creativecommons.org/licenses/by/4.0/ | We show theoretically that the noise due to laser induced backaction acting
on a small nanosphere levitated in a standing-wave trap can be considerably
reduced by utilising a suitable reflective boundary. We examine the spherical
mirror geometry as a case study of this backaction suppression effect,
discussing the theoretical and experimental constraints. We study the effects
of laser recoil directly, by analysing optical force fluctuations acting on a
dipolar particle trapped at the centre of a spherical mirror. We also compute
the corresponding measurement imprecision in an interferometric,
shot-noise-limited position measurement, using the formalism of Fisher
information flow. Our results show that the standing-wave trapping field is
necessary for backaction suppression in three dimensions, and they satisfy the
Heisenberg limit of detection.
| [
{
"version": "v1",
"created": "Tue, 7 May 2024 14:46:48 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 11:25:04 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Gajewski",
"Rafał",
""
],
[
"Bateman",
"James",
""
]
] |
2405.05178 | Yoshiki Fukusumi | Yoshiki Fukusumi | Fusion rule in conformal field theories and topological orders: A
unified view of correspondence and (fractional) supersymmetry and their
relation to topological holography | References are added, discussions on the symmetry in lattice models
in recent literature are added, and typos are corrected (v1). Discussions on
category theories and corresponding references are added (v2). Typos are
corrected and comments on the related research directions have been added
(v3) | null | null | null | hep-th cond-mat.stat-mech cond-mat.str-el math-ph math.MP math.QA | http://creativecommons.org/licenses/by/4.0/ | The algebraic or ring structure of anyons called the fusion rule, is one of
the most fundamental research interests in contemporary studies on topological
orders (TOs) and the corresponding conformal field theories (CFTs). Recently,
the algebraic structure realized as generalized symmetry, including
non-invertible and categorical symmetry, captured attention in the fields. Such
non-abelian anyonic objects appear in a bulk CFT or chiral CFT (CCFT), but it
has been known that the construction of a CCFT contains theoretical
difficulties in general. In this work, we study the structure of the fusion
rule in $Z_{N}$ symmetric chiral and bulk conformal field theories and the
corresponding TOs. We propose a nontrivial expression of subalgebra structure
in the fusion rule of a bulk CFT. We name this subalgebra ``bulk semion". This
corresponds to the fusion rule of the CCFTs and categorical symmetry of the TOs
or symmetry topological field theory (SymTFT). This gives a bulk-edge
correspondence based on the symmetry analysis and corresponds to an anyon
algebraic expression of topological holography in the recent literature. The
recent topological holography is expected to apply to systems in general
space-time dimensions. Moreover, we give a concise way of unifying duality (or
fractional supersymmetry), generalized or categorical symmetry, and Lagrangian
subalgebra. Our method is potentially useful to formulate and study general
TOs, fundamentally only from the data of bulk CFTs or vice versa, and gives a
clue in understanding CCFT (or ancillary CFT more generally).
| [
{
"version": "v1",
"created": "Wed, 8 May 2024 16:14:02 GMT"
},
{
"version": "v2",
"created": "Wed, 15 May 2024 05:31:44 GMT"
},
{
"version": "v3",
"created": "Tue, 10 Dec 2024 05:01:20 GMT"
},
{
"version": "v4",
"created": "Fri, 25 Apr 2025 06:33:37 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Fukusumi",
"Yoshiki",
""
]
] |
2405.05316 | Ophelia Evelyn Sommer | Ophelia Evelyn Sommer, Xueda Wen, Ashvin Vishwanath | Higher Berry Curvature from the Wave Function I: Schmidt Decomposition
and Matrix Product States | 9 pages, 2 figures | Phys. Rev. Lett. 134, 146601 (2025) | 10.1103/PhysRevLett.134.146601 | null | cond-mat.str-el hep-th quant-ph | http://creativecommons.org/licenses/by/4.0/ | Higher Berry curvature (HBC) is the proposed generalization of Berry
curvature to infinitely extended systems. Heuristically HBC captures the flow
of local Berry curvature in a system. Here we provide a simple formula for
computing the HBC for extended $d = 1$ systems at the level of wave functions
using the Schmidt decomposition. We also find a corresponding formula for
matrix product states (MPS), and show that for translationally invariant MPS
this gives rise to a quantized invariant. We demonstrate our approach with an
exactly solvable model and numerical calculations for generic models using
iDMRG
| [
{
"version": "v1",
"created": "Wed, 8 May 2024 18:00:02 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Sommer",
"Ophelia Evelyn",
""
],
[
"Wen",
"Xueda",
""
],
[
"Vishwanath",
"Ashvin",
""
]
] |
2405.05323 | Ophelia Evelyn Sommer | Ophelia Evelyn Sommer, Ashvin Vishwanath, Xueda Wen | Higher Berry Curvature from the Wave function II: Locally Parameterized
States Beyond One Dimension | 18 pages, 4 figures | Phys. Rev. B 111, 155110 (2025) | 10.1103/PhysRevB.111.155110 | null | cond-mat.str-el hep-th quant-ph | http://creativecommons.org/licenses/by/4.0/ | We propose a systematic wave function based approach to construct topological
invariants for families of lattice systems that are short-range entangled using
local parameter spaces. This construction is particularly suitable when given a
family of tensor networks that can be viewed as the ground states of $d$
dimensional lattice systems, for which we construct the closed $(d+2)$-form
higher Berry curvature, which is a generalization of the well known 2-form
Berry curvature. Such $(d+2)$-form higher Berry curvature characterizes a flow
of $(d+1)$-form higher Berry curvature in the system. Our construction is
equally suitable for constructing other higher pumps, such as the (higher)
Thouless pump in the presence of a global on-site $U(1)$ symmetry, which
corresponds to a closed $d$-form. The cohomology classes of such higher
differential forms are topological invariants and are expected to be quantized
for short-range entangled states. We illustrate our construction with exactly
solvable lattice models that are in nontrivial higher Berry classes in $d=2$.
| [
{
"version": "v1",
"created": "Wed, 8 May 2024 18:00:08 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Sommer",
"Ophelia Evelyn",
""
],
[
"Vishwanath",
"Ashvin",
""
],
[
"Wen",
"Xueda",
""
]
] |
2405.08186 | Alejandro Bravo-Doddoli Ph.D. | Alejandro Bravo-Doddoli | Metric Lines in Engel-type Groups | null | null | null | null | math.DG math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the framework of sub-Riemannian Manifolds, a relevant question is: what
are the \enquote{metric lines} (i.e., the isometric embedding of the real
line)? This article presents a conjecture classifying the metric lines in
Carnot groups and takes the first steps in answering this question for
\enquote{arbitrary rank} Carnot groups. We classify the metric lines of the
Engel-type groups $\Eng(n)$ (Theorem 1.2), whose sub-Riemannian structure is
defined on a non-integrable distribution of rank $n+1$. Our approach is a new
method, called the sequence method, which we began to develop to study metric
lines in the jet space.
| [
{
"version": "v1",
"created": "Mon, 13 May 2024 21:07:56 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2025 23:44:34 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Bravo-Doddoli",
"Alejandro",
""
]
] |
2405.11820 | Arindam Das | Arindam Das, Puja Das, Nobuchika Okada | Testing neutrino mass hierarchy under type-II seesaw scenario in
$U(1)_X$ from colliders | 14 pages, 7 figures, extended analysis and muon collider corrected | Phys. Lett. B 865 (2025) 139475 | 10.1016/j.physletb.2025.139475 | null | hep-ph | http://creativecommons.org/publicdomain/zero/1.0/ | The origin of tiny neutrino mass is a long standing unsolved puzzle of the
Standard Model (SM), which allows us to consider scenarios beyond the Standard
Model (BSM) in a variety of ways. One of them being a gauge extension of the SM
may be realized as in the form of an anomaly free, general $U(1)_X$ extension
of the SM, where an $SU(2)_L$ triplet scalar with a $U(1)_X$ charge is
introduced to have Dirac Yukawa couplings with the SM lepton doublets. Once the
triplet scalar developes a Vacuum Expectation Value (VEV), light neutrinos
acquire their tiny Majorana masses. Hence, the decay modes of the triplet
scalar has a direct connection to the neutrino oscillation data for different
neutrino mass hierarchies. After the breaking of the $U(1)_X$ gauge symmetry, a
neutral $U(1)_X$ gauge boson $(Z^\prime)$ acquires mass, which interacts
differently with the left and right handed SM fermions. Satisfying the recent
LHC bounds on the triplet scalar and $Z^\prime$ boson productions, we study the
pair production of the triplet scalar at LHC, 100 TeV proton proton collider
FCC, $e^-e^+$ and $\mu^-\mu^+$ colliders followed by its decay into dominant
dilepton modes whose flavor structure depend on the neutrino mass hierarchy.
Generating the SM backgrounds, we study the possible signal significance of
four lepton final states from the triplet scalar pair production. We also
compare our results with the purely SM gauge mediated triplet scalar pair
production followed by four lepton final states, which could be significant
only in $\mu^- \mu^+$ collider.
| [
{
"version": "v1",
"created": "Mon, 20 May 2024 06:29:24 GMT"
},
{
"version": "v2",
"created": "Tue, 16 Jul 2024 12:29:43 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 04:13:43 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Das",
"Arindam",
""
],
[
"Das",
"Puja",
""
],
[
"Okada",
"Nobuchika",
""
]
] |
2405.13625 | Simone Naldi | Vladimir Kolmogorov and Simone Naldi and Jeferson Zapata | Certifying solutions of degenerate semidefinite programs | 20 pages, 1 table; accepted to SIAM J. Optimization (April 2025) | null | null | null | math.OC cs.SC math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper deals with the algorithmic aspects of solving feasibility problems
of semidefinite programming (SDP), aka linear matrix inequalities (LMI). Since
in some SDP instances all feasible solutions have irrational entries, numerical
solvers that work with rational numbers can only find an approximate solution.
We study the following question: is it possible to certify feasibility of a
given SDP using an approximate solution that is sufficiently close to some
exact solution? Existing approaches make the assumption that there exist
rational feasible solutions (and use techniques such as rounding and lattice
reduction algorithms). We propose an alternative approach that does not need
this assumption. More specifically, we show how to construct a system of
polynomial equations whose set of real solutions is guaranteed to have an
isolated correct solution (assuming that the target exact solution is
maximum-rank). This allows, in particular, to use algorithms from real
algebraic geometry for solving systems of polynomial equations, yielding a
hybrid (or symbolic-numerical) method for SDPs. We experimentally compare it
with a pure symbolic method; the hybrid method was able to certify feasibility
of many SDP instances on which the exact method failed. Our approach may have
further applications, such as refining an approximate solution using methods of
numerical algebraic geometry for systems of polynomial equations.
| [
{
"version": "v1",
"created": "Wed, 22 May 2024 13:23:10 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Apr 2025 19:13:21 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 06:08:50 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Kolmogorov",
"Vladimir",
""
],
[
"Naldi",
"Simone",
""
],
[
"Zapata",
"Jeferson",
""
]
] |
2405.13876 | Chakradhar Rangi | Chakradhar Rangi, Herbert F Fotso, Hanna Terletska, Juana Moreno, and
Ka-Ming Tam | Disorder Enhanced Thermalization in Interacting Many-Particle System | 6 pages, 3 figures | Phys. Rev. B 111, L161122 (2025) | 10.1103/PhysRevB.111.L161122 | null | cond-mat.str-el | http://creativecommons.org/licenses/by/4.0/ | We introduce an extension of the non-equilibrium dynamical mean field theory
to incorporate the effects of static random disorder in the dynamics of a
many-particle system by integrating out different disorder configurations
resulting in an effective time-dependent density-density interaction. We use
this method to study the non-equilibrium transient dynamics of a system
described by the Fermi Anderson-Hubbard model following an interaction and
disorder quench. The method recovers the solution of the disorder-free case for
which the system exhibits qualitatively distinct dynamical behaviors in the
weak-coupling (prethermalization) and strong-coupling regimes
(collapse-and-revival oscillations). However, we find that weak random disorder
promotes thermalization. In the weak coupling regime, the jump in the
quasiparticle weight in the prethermal regime is suppressed by random disorder
while in the strong-coupling regime, random disorder reduces the amplitude of
the quasiparticle weight oscillations. These results highlight the importance
of disorder in the dynamics of realistic many-particle systems.
| [
{
"version": "v1",
"created": "Wed, 22 May 2024 17:57:47 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 17:21:37 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Rangi",
"Chakradhar",
""
],
[
"Fotso",
"Herbert F",
""
],
[
"Terletska",
"Hanna",
""
],
[
"Moreno",
"Juana",
""
],
[
"Tam",
"Ka-Ming",
""
]
] |
2405.14548 | Vinicius Luiz Santos Silva | Vinicius L S Silva, Geraldine Regnier, Pablo Salinas, Claire E Heaney,
Matthew D Jackson, Christopher C Pain | Rapid modelling of reactive transport in porous media using machine
learning: limitations and solutions | null | null | null | null | cs.CE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reactive transport in porous media plays a pivotal role in subsurface
reservoir processes, influencing fluid properties and geochemical
characteristics. However, coupling fluid flow and transport with geochemical
reactions is computationally intensive, requiring geochemical calculations at
each grid cell and each time step within a discretized simulation domain.
Although recent advancements have integrated machine learning techniques as
surrogates for geochemical simulations, ensuring computational efficiency and
accuracy remains a challenge. This work investigates machine learning models as
replacements for a geochemical module in a simulation of reactive transport in
porous media. As a proof of concept, we test this approach on a well-documented
cation exchange problem. While the surrogate models excel in isolated
predictions, they fall short in rollout predictions over successive time steps.
By introducing modifications, including physics-based constraints and tailored
dataset generation strategies, we show that machine learning surrogates can
achieve accurate rollout predictions. Our findings emphasize that even for a
simple sorption equilibrium reaction (cation exchange problem), machine
learning surrogates alone fail in predicting over successive time-steps.
Incorporating simple physics-based modifications enables us to overcome this
limitation. A detailed analysis of the limitations and potential mitigation
strategies is presented in this work.
| [
{
"version": "v1",
"created": "Thu, 23 May 2024 13:28:10 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 11:14:56 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Silva",
"Vinicius L S",
""
],
[
"Regnier",
"Geraldine",
""
],
[
"Salinas",
"Pablo",
""
],
[
"Heaney",
"Claire E",
""
],
[
"Jackson",
"Matthew D",
""
],
[
"Pain",
"Christopher C",
""
]
] |
2405.15638 | Hongyu Wang | Hongyu Wang and Jiayu Xu and Senwei Xie and Ruiping Wang and Jialin Li
and Zhaojie Xie and Bin Zhang and Chuyan Xiong and Xilin Chen | M4U: Evaluating Multilingual Understanding and Reasoning for Large
Multimodal Models | Work in progress | null | null | null | cs.CV cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Multilingual capability is an essential aspect for large multimodal models,
since they are usually deployed across various countries and languages.
However, most existing benchmarks for multilingual multimodal reasoning
struggle to differentiate between models of varying performance; even language
models without visual capabilities can easily achieve high scores. This leaves
a comprehensive evaluation of leading multilingual multimodal models largely
unexplored. In this work, we introduce M4U, a novel and challenging benchmark
for assessing the capability of multi-discipline multilingual multimodal
understanding and reasoning. M4U contains 10k samples covering 64 disciplines
across 16 subfields in Science, Engineering, and Healthcare in six languages.
Using M4U, we conduct extensive evaluations of leading Large Multimodal Models
(LMMs) and Large Language Models (LLMs) with external tools. The evaluation
results demonstrate that the state-of-the-art model, GPT-4o, achieves only
47.6% average accuracy on M4U. Additionally, we observe that the leading LMMs
exhibit significant language preferences. Our in-depth analysis indicates that
leading LMMs, including GPT-4o, struggle to perform reasoning using
multilingual information present in both visual and textual context.
Specifically, they suffer performance degradation when prompted with
cross-lingual multimodal questions. Our code and dataset is public available.
| [
{
"version": "v1",
"created": "Fri, 24 May 2024 15:25:28 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 04:37:45 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Wang",
"Hongyu",
""
],
[
"Xu",
"Jiayu",
""
],
[
"Xie",
"Senwei",
""
],
[
"Wang",
"Ruiping",
""
],
[
"Li",
"Jialin",
""
],
[
"Xie",
"Zhaojie",
""
],
[
"Zhang",
"Bin",
""
],
[
"Xiong",
"Chuyan",
""
],
[
"Chen",
"Xilin",
""
]
] |
2405.16670 | Wojciech O\.za\'nski | W. S. O\.za\'nski, W. Zaj\k{a}czkowski | On the regularity of axially-symmetric solutions to the incompressible
Navier-Stokes equations in a cylinder | 31 pages | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the axisymmetric Navier-Stokes equations in a finite cylinder
$\Omega\subset\mathbb{R}^3$. We assume that $v_r$, $v_\varphi$,
$\omega_\varphi$ vanish on the lateral boundary $\partial \Omega$ of the
cylinder, and that $v_z$, $\omega_\varphi$, $\partial_z v_\varphi$ vanish on
the top and bottom parts of the boundary $\partial \Omega$, where we used
standard cylindrical coordinates, and we denoted by $\omega =\mathrm{curl}\, v$
the vorticity field. We use weighted estimates and $H^3$ Sobolev estimate on
the modified stream function to derive three order-reduction estimates. These
enable one to reduce the order of the nonlinear estimates of the equations, and
help observe that the solutions to the equations are ``almost regular''. We use
the order-reduction estimates to show that the solution to the equations
remains regular as long as, for any $p\in (6,\infty)$, $\| v_\varphi
\|_{L^\infty_t L^p_x}/\| v_\varphi \|_{L^\infty_t L^\infty_x}$ remains bounded
below by a positive number.
| [
{
"version": "v1",
"created": "Sun, 26 May 2024 19:21:26 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 00:15:42 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Ożański",
"W. S.",
""
],
[
"Zajączkowski",
"W.",
""
]
] |
2405.16874 | Xingqun Qi | Xingqun Qi, Hengyuan Zhang, Yatian Wang, Jiahao Pan, Chen Liu, Peng
Li, Xiaowei Chi, Mengfei Li, Wei Xue, Shanghang Zhang, Wenhan Luo, Qifeng
Liu, Yike Guo | CoCoGesture: Toward Coherent Co-speech 3D Gesture Generation in the Wild | After the submission of the paper, we realized that the study still
has room for expansion. In order to make the research findings more profound
and comprehensive, we have decided to withdraw the paper so that we can
conduct further research and expansion | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Deriving co-speech 3D gestures has seen tremendous progress in virtual avatar
animation. Yet, the existing methods often produce stiff and unreasonable
gestures with unseen human speech inputs due to the limited 3D speech-gesture
data. In this paper, we propose CoCoGesture, a novel framework enabling vivid
and diverse gesture synthesis from unseen human speech prompts. Our key insight
is built upon the custom-designed pretrain-fintune training paradigm. At the
pretraining stage, we aim to formulate a large generalizable gesture diffusion
model by learning the abundant postures manifold. Therefore, to alleviate the
scarcity of 3D data, we first construct a large-scale co-speech 3D gesture
dataset containing more than 40M meshed posture instances across 4.3K speakers,
dubbed GES-X. Then, we scale up the large unconditional diffusion model to 1B
parameters and pre-train it to be our gesture experts. At the finetune stage,
we present the audio ControlNet that incorporates the human voice as condition
prompts to guide the gesture generation. Here, we construct the audio
ControlNet through a trainable copy of our pre-trained diffusion model.
Moreover, we design a novel Mixture-of-Gesture-Experts (MoGE) block to
adaptively fuse the audio embedding from the human speech and the gesture
features from the pre-trained gesture experts with a routing mechanism. Such an
effective manner ensures audio embedding is temporal coordinated with motion
features while preserving the vivid and diverse gesture generation. Extensive
experiments demonstrate that our proposed CoCoGesture outperforms the
state-of-the-art methods on the zero-shot speech-to-gesture generation. The
dataset will be publicly available at: https://mattie-e.github.io/GES-X/
| [
{
"version": "v1",
"created": "Mon, 27 May 2024 06:47:14 GMT"
},
{
"version": "v2",
"created": "Thu, 28 Nov 2024 07:33:45 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 09:26:36 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Qi",
"Xingqun",
""
],
[
"Zhang",
"Hengyuan",
""
],
[
"Wang",
"Yatian",
""
],
[
"Pan",
"Jiahao",
""
],
[
"Liu",
"Chen",
""
],
[
"Li",
"Peng",
""
],
[
"Chi",
"Xiaowei",
""
],
[
"Li",
"Mengfei",
""
],
[
"Xue",
"Wei",
""
],
[
"Zhang",
"Shanghang",
""
],
[
"Luo",
"Wenhan",
""
],
[
"Liu",
"Qifeng",
""
],
[
"Guo",
"Yike",
""
]
] |
2405.19325 | Minghan Li | Minghan Li, Xilun Chen, Ari Holtzman, Beidi Chen, Jimmy Lin, Wen-tau
Yih, Xi Victoria Lin | Nearest Neighbor Speculative Decoding for LLM Generation and Attribution | null | Advances in Neural Information Processing Systems (2024), vol. 37,
page 80987-81015 | null | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Large language models (LLMs) often hallucinate and lack the ability to
provide attribution for their generations. Semi-parametric LMs, such as kNN-LM,
approach these limitations by refining the output of an LM for a given prompt
using its nearest neighbor matches in a non-parametric data store. However,
these models often exhibit slow inference speeds and produce non-fluent texts.
In this paper, we introduce Nearest Neighbor Speculative Decoding (NEST), a
novel semi-parametric language modeling approach that is capable of
incorporating real-world text spans of arbitrary length into the LM generations
and providing attribution to their sources. NEST performs token-level retrieval
at each inference step to compute a semi-parametric mixture distribution and
identify promising span continuations in a corpus. It then uses an approximate
speculative decoding procedure that accepts a prefix of the retrieved span or
generates a new token. NEST significantly enhances the generation quality and
attribution rate of the base LM across a variety of knowledge-intensive tasks,
surpassing the conventional kNN-LM method and performing competitively with
in-context retrieval augmentation. In addition, NEST substantially improves the
generation speed, achieving a 1.8x speedup in inference time when applied to
Llama-2-Chat 70B. Code will be released at
https://github.com/facebookresearch/NEST/tree/main.
| [
{
"version": "v1",
"created": "Wed, 29 May 2024 17:55:03 GMT"
},
{
"version": "v2",
"created": "Fri, 31 May 2024 01:41:49 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 01:08:35 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Li",
"Minghan",
""
],
[
"Chen",
"Xilun",
""
],
[
"Holtzman",
"Ari",
""
],
[
"Chen",
"Beidi",
""
],
[
"Lin",
"Jimmy",
""
],
[
"Yih",
"Wen-tau",
""
],
[
"Lin",
"Xi Victoria",
""
]
] |
2405.19548 | Mingqi Yuan | Mingqi Yuan, Roger Creus Castanyer, Bo Li, Xin Jin, Wenjun Zeng, Glen
Berseth | RLeXplore: Accelerating Research in Intrinsically-Motivated
Reinforcement Learning | 42 pages, 31 figures. Transactions on Machine Learning Research, 2025 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Extrinsic rewards can effectively guide reinforcement learning (RL) agents in
specific tasks. However, extrinsic rewards frequently fall short in complex
environments due to the significant human effort needed for their design and
annotation. This limitation underscores the necessity for intrinsic rewards,
which offer auxiliary and dense signals and can enable agents to learn in an
unsupervised manner. Although various intrinsic reward formulations have been
proposed, their implementation and optimization details are insufficiently
explored and lack standardization, thereby hindering research progress. To
address this gap, we introduce RLeXplore, a unified, highly modularized, and
plug-and-play framework offering reliable implementations of eight
state-of-the-art intrinsic reward methods. Furthermore, we conduct an in-depth
study that identifies critical implementation details and establishes
well-justified standard practices in intrinsically-motivated RL. Our
documentation, examples, and source code are available at
https://github.com/RLE-Foundation/RLeXplore.
| [
{
"version": "v1",
"created": "Wed, 29 May 2024 22:23:20 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 01:53:37 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Yuan",
"Mingqi",
""
],
[
"Castanyer",
"Roger Creus",
""
],
[
"Li",
"Bo",
""
],
[
"Jin",
"Xin",
""
],
[
"Zeng",
"Wenjun",
""
],
[
"Berseth",
"Glen",
""
]
] |
2405.19912 | Antonin Schrab | Antonin Schrab, Ilmun Kim | Robust Kernel Hypothesis Testing under Data Corruption | 22 pages, 2 figures, 2 algorithms | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We propose a general method for constructing robust permutation tests under
data corruption. The proposed tests effectively control the non-asymptotic type
I error under data corruption, and we prove their consistency in power under
minimal conditions. This contributes to the practical deployment of hypothesis
tests for real-world applications with potential adversarial attacks. For the
two-sample and independence settings, we show that our kernel robust tests are
minimax optimal, in the sense that they are guaranteed to be non-asymptotically
powerful against alternatives uniformly separated from the null in the kernel
MMD and HSIC metrics at some optimal rate (tight with matching lower bound). We
point out that existing differentially private tests can be adapted to be
robust to data corruption, and we demonstrate in experiments that our proposed
tests achieve much higher power than these private tests. Finally, we provide
publicly available implementations and empirically illustrate the practicality
of our robust tests.
| [
{
"version": "v1",
"created": "Thu, 30 May 2024 10:23:16 GMT"
},
{
"version": "v2",
"created": "Sun, 23 Feb 2025 13:27:00 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 09:16:54 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Schrab",
"Antonin",
""
],
[
"Kim",
"Ilmun",
""
]
] |
2405.20374 | James Alvey | Tim Zimmermann and James Alvey and David J. E. Marsh and Malcolm
Fairbairn and Justin I. Read | Dwarf galaxies imply dark matter is heavier than $\mathbf{2.2 \times
10^{-21}} \, \mathbf{eV}$ | 5 pages, 3 figures. The jaxsp library is available at
https://github.com/timzimm/jaxsp. v3: version accepted by PRL, additional
appendix added on dsph modelling, v2: significantly expanded appendix
containing jaxsp numerical checks and more details on statistical approach | Phys. Rev. Lett. 134, 151001 (2025) | 10.1103/PhysRevLett.134.151001 | KCL-TH-PH-2024-30 | astro-ph.CO astro-ph.GA gr-qc hep-ph | http://creativecommons.org/licenses/by/4.0/ | It is widely established that a lower bound on the dark matter particle mass,
$m$, can be obtained by demanding that the de Broglie wavelength in a given
galaxy must be smaller than the virial radius of the galaxy, leading to
$m\gtrsim 10^{-22}\text{ eV}$ when applied to typical dwarf galaxies. This
lower limit has never been derived precisely or rigorously. We use stellar
kinematical data for the Milky Way satellite galaxy Leo II to self-consistently
reconstruct a statistical ensemble of dark matter wavefunctions and
corresponding density profiles. By comparison to a data-driven,
model-independent reconstruction, and using a variant of the maximum mean
discrepancy as a statistical measure, we determine that a self-consistent
description of dark matter in the local Universe requires $m>2.2 \times
10^{-21}\,\mathrm{eV}\;\mathrm{(CL>95\%)}$. This lower limit is free of any
assumptions pertaining to cosmology, microphysics (including spin), or dynamics
of dark matter, and only assumes that it is predominantly composed of a single
bosonic particle species.
| [
{
"version": "v1",
"created": "Thu, 30 May 2024 18:00:01 GMT"
},
{
"version": "v2",
"created": "Wed, 29 Jan 2025 13:08:36 GMT"
},
{
"version": "v3",
"created": "Tue, 4 Mar 2025 12:45:39 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Zimmermann",
"Tim",
""
],
[
"Alvey",
"James",
""
],
[
"Marsh",
"David J. E.",
""
],
[
"Fairbairn",
"Malcolm",
""
],
[
"Read",
"Justin I.",
""
]
] |
2406.00415 | Xuan Wu | Xuan Wu, Di Wang, Lijie Wen, Yubin Xiao, Chunguo Wu, Yuesong Wu,
Chaoyu Yu, Douglas L. Maskell, and You Zhou | Neural Combinatorial Optimization Algorithms for Solving Vehicle Routing
Problems: A Comprehensive Survey with Perspectives | submitted to TNNLS | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although several surveys on Neural Combinatorial Optimization (NCO) solvers
specifically designed to solve Vehicle Routing Problems (VRPs) have been
conducted, they did not cover the state-of-the-art (SOTA) NCO solvers emerged
recently. More importantly, to establish a comprehensive and up-to-date
taxonomy of NCO solvers, we systematically review relevant publications and
preprints, categorizing them into four distinct types, namely Learning to
Construct, Learning to Improve, Learning to Predict-Once, and Learning to
Predict-Multiplicity solvers. Subsequently, we present the inadequacies of the
SOTA solvers, including poor generalization, incapability to solve large-scale
VRPs, inability to address most types of VRP variants simultaneously, and
difficulty in comparing these NCO solvers with the conventional Operations
Research algorithms. Simultaneously, we discuss on-going efforts, identify open
inadequacies, as well as propose promising and viable directions to overcome
these inadequacies. Notably, existing efforts focus on only one or two of these
inadequacies, with none attempting to address all of them concurrently. In
addition, we compare the performance of representative NCO solvers from the
Reinforcement, Supervised, and Unsupervised Learning paradigms across VRPs of
varying scales. Finally, following the proposed taxonomy, we provide an
accompanying web page as a live repository for NCO solvers. Through this survey
and the live repository, we aim to foster further advancements in the NCO
community.
| [
{
"version": "v1",
"created": "Sat, 1 Jun 2024 12:18:39 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Oct 2024 08:42:13 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 06:52:44 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Wu",
"Xuan",
""
],
[
"Wang",
"Di",
""
],
[
"Wen",
"Lijie",
""
],
[
"Xiao",
"Yubin",
""
],
[
"Wu",
"Chunguo",
""
],
[
"Wu",
"Yuesong",
""
],
[
"Yu",
"Chaoyu",
""
],
[
"Maskell",
"Douglas L.",
""
],
[
"Zhou",
"You",
""
]
] |
2406.02105 | Vignesh Kothapalli | Vignesh Kothapalli, Tom Tirer | Can Kernel Methods Explain How the Data Affects Neural Collapse? | Transactions on Machine Learning Research (TMLR) | null | null | null | cs.LG cs.AI cs.IT math.IT stat.ML | http://creativecommons.org/licenses/by/4.0/ | A vast amount of literature has recently focused on the "Neural Collapse"
(NC) phenomenon, which emerges when training neural network (NN) classifiers
beyond the zero training error point. The core component of NC is the decrease
in the within-class variability of the network's deepest features, dubbed as
NC1. The theoretical works that study NC are typically based on simplified
unconstrained features models (UFMs) that mask any effect of the data on the
extent of collapse. To address this limitation of UFMs, this paper explores the
possibility of analyzing NC1 using kernels associated with shallow NNs. We
begin by formulating an NC1 metric as a function of the kernel. Then, we
specialize it to the NN Gaussian Process kernel (NNGP) and the Neural Tangent
Kernel (NTK), associated with wide networks at initialization and during
gradient-based training with a small learning rate, respectively. As a key
result, we show that the NTK does not represent more collapsed features than
the NNGP for Gaussian data of arbitrary dimensions. This showcases the
limitations of data-independent kernels such as NTK in approximating the NC
behavior of NNs. As an alternative to NTK, we then empirically explore a
recently proposed data-aware Gaussian Process kernel, which generalizes NNGP to
model feature learning. We show that this kernel yields lower NC1 than NNGP but
may not follow the trends of the shallow NN. Our study demonstrates that
adaptivity to data may allow kernel-based analysis of NC, though further
advancements in this area are still needed. A nice byproduct of our study is
showing both theoretically and empirically that the choice of nonlinear
activation function affects NC1 (with ERF yielding lower values than ReLU). The
code is available at: https://github.com/kvignesh1420/shallow_nc1
| [
{
"version": "v1",
"created": "Tue, 4 Jun 2024 08:33:56 GMT"
},
{
"version": "v2",
"created": "Fri, 28 Jun 2024 04:05:53 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 06:43:34 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Kothapalli",
"Vignesh",
""
],
[
"Tirer",
"Tom",
""
]
] |
2406.02566 | Ognjen Kundacina PhD | Ognjen Kundacina, Vladimir Vincan, Dragisa Miskovic | Combining X-Vectors and Bayesian Batch Active Learning: Two-Stage Active
Learning Pipeline for Speech Recognition | null | null | null | null | eess.AS cs.AI cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | This paper introduces a novel two-stage active learning (AL) pipeline for
automatic speech recognition (ASR), combining unsupervised and supervised AL
methods. The first stage utilizes unsupervised AL by using x-vectors clustering
for diverse sample selection from unlabeled speech data, thus establishing a
robust initial dataset for the subsequent supervised AL. The second stage
incorporates a supervised AL strategy, with a batch AL method specifically
developed for ASR, aimed at selecting diverse and informative batches of
samples. Here, sample diversity is also achieved using x-vectors clustering,
while the most informative samples are identified using a Bayesian AL method
tailored for ASR with an adaptation of Monte Carlo dropout to approximate
Bayesian inference. This approach enables precise uncertainty estimation,
thereby enhancing ASR model training with significantly reduced data
requirements. Our method has shown superior performance compared to competing
methods on homogeneous, heterogeneous, and OOD test sets, demonstrating that
strategic sample selection and innovative Bayesian modeling can substantially
optimize both labeling effort and data utilization in deep learning-based ASR
applications.
| [
{
"version": "v1",
"created": "Fri, 3 May 2024 19:24:41 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 06:24:10 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Kundacina",
"Ognjen",
""
],
[
"Vincan",
"Vladimir",
""
],
[
"Miskovic",
"Dragisa",
""
]
] |
2406.02700 | Volodymyr Sivak | Volodymyr Sivak, Michael Newman, Paul Klimov | Optimization of decoder priors for accurate quantum error correction | null | Phys. Rev. Lett. 133, 150603 (2024) | 10.1103/physrevlett.133.150603 | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | Accurate decoding of quantum error-correcting codes is a crucial ingredient
in protecting quantum information from decoherence. It requires characterizing
the error channels corrupting the logical quantum state and providing this
information as a prior to the decoder. We introduce a reinforcement learning
inspired method for calibrating these priors that aims to minimize the logical
error rate. Our method significantly improves the decoding accuracy in
repetition and surface code memory experiments executed on Google's Sycamore
processor, outperforming the leading decoder-agnostic method by 16% and 3.3%
respectively. This calibration approach will serve as an important tool for
maximizing the performance of both near-term and future error-corrected quantum
devices.
| [
{
"version": "v1",
"created": "Tue, 4 Jun 2024 18:26:09 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Sivak",
"Volodymyr",
""
],
[
"Newman",
"Michael",
""
],
[
"Klimov",
"Paul",
""
]
] |
2406.03655 | Steven Senger | Robert Coulter and Steven Senger | More on the number of distinct values of a class of functions | We corrected an error, and greatly improved and generalized our
bounds from the previous version | null | null | null | math.CO | http://creativecommons.org/licenses/by/4.0/ | In a previous article, the authors determined the first (and at the time of
writing, the only) non-trivial upper bound for the cardinality of the image set
for several classes of functions, including planar functions. Here, we show
that the upper bound cannot be tight for planar functions over $\mathbb F_{q}$,
with the possible exception of $q=343.$ We further show that if such an
exception exists, then it implies the existence of a projective plane of order
18.
This follows from more general upper bound results, which apply to a wider
class of functions. To determine these bounds, we first resolve a problem
related to Gauss' well-known result on triangular sums. By doing so, we are
then able to approach our problem via a well-studied Diophantine problem, which
eventually yields upper bounds that are related to class numbers of quadratic
number fields.
| [
{
"version": "v1",
"created": "Wed, 5 Jun 2024 23:47:19 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2025 19:27:24 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Coulter",
"Robert",
""
],
[
"Senger",
"Steven",
""
]
] |
2406.04621 | Xu Wen | Jie Xiong, Wen Xu | Mean-field stochastic linear quadratic control problem with random
coefficients | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | In this paper, we first prove that the mean-field stochastic linear quadratic
(MFSLQ for short) control problem with random coefficients has a unique optimal
control and derive a preliminary stochastic maximum principle to characterize
this optimal control by an optimality system. However, because of the term of
the form $\mathbb{E}[A_1(\cdot)^\top Y(\cdot)] $ in the adjoint equation, which
cannot be represented in the form $\mathbb{E}[A_1(\cdot)^\top]\mathbb{E}
[Y(\cdot)] $, we cannot solve this optimality system explicitly. To this end,
we decompose the MFSLQ control problem into two problems without the mean-field
terms, and one of them is a constrained problem. The constrained SLQ control
problem is solved explicitly by an extended LaGrange multiplier method
developed in this article.
| [
{
"version": "v1",
"created": "Fri, 7 Jun 2024 04:05:10 GMT"
},
{
"version": "v2",
"created": "Tue, 30 Jul 2024 07:27:57 GMT"
},
{
"version": "v3",
"created": "Sun, 24 Nov 2024 12:43:25 GMT"
},
{
"version": "v4",
"created": "Mon, 24 Mar 2025 02:25:59 GMT"
},
{
"version": "v5",
"created": "Thu, 24 Apr 2025 23:59:17 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Xiong",
"Jie",
""
],
[
"Xu",
"Wen",
""
]
] |
2406.05556 | Thomas Allard | Thomas Allard, Helmut B\"olcskei | Entropy of Compact Operators with Applications to Landau-Pollak-Slepian
Theory and Sobolev Spaces | null | null | 10.1016/j.acha.2025.101762 | null | math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We derive a precise general relation between the entropy of a compact
operator and its eigenvalues. It is then shown how this result along with the
underlying philosophy can be applied to improve substantially on the best known
characterizations of the entropy of the Landau-Pollak-Slepian operator and the
metric entropy of unit balls in Sobolev spaces.
| [
{
"version": "v1",
"created": "Sat, 8 Jun 2024 19:07:20 GMT"
},
{
"version": "v2",
"created": "Wed, 5 Mar 2025 09:31:57 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Allard",
"Thomas",
""
],
[
"Bölcskei",
"Helmut",
""
]
] |
2406.07529 | Lu Li | Lu Li, Tianyu Zhang, Zhiqi Bu, Suyuchen Wang, Huan He, Jie Fu, Yonghui
Wu, Jiang Bian, Yong Chen, Yoshua Bengio | MAP: Low-compute Model Merging with Amortized Pareto Fronts via
Quadratic Approximation | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Model merging has emerged as an effective approach to combine multiple
single-task models into a multitask model. This process typically involves
computing a weighted average of the model parameters without any additional
training. Existing model-merging methods focus on enhancing average task
accuracy. However, interference and conflicts between the objectives of
different tasks can lead to trade-offs during the merging process. In
real-world applications, a set of solutions with various trade-offs can be more
informative, helping practitioners make decisions based on diverse preferences.
In this paper, we introduce a novel and low-compute algorithm, Model Merging
with Amortized Pareto Front (MAP). MAP efficiently identifies a Pareto set of
scaling coefficients for merging multiple models, reflecting the trade-offs
involved. It amortizes the substantial computational cost of evaluations needed
to estimate the Pareto front by using quadratic approximation surrogate models
derived from a pre-selected set of scaling coefficients. Experimental results
on vision and natural language processing tasks demonstrate that MAP can
accurately identify the Pareto front, providing practitioners with flexible
solutions to balance competing task objectives. We also introduce Bayesian MAP
for scenarios with a relatively low number of tasks and Nested MAP for
situations with a high number of tasks, further reducing the computational cost
of evaluation.
| [
{
"version": "v1",
"created": "Tue, 11 Jun 2024 17:55:25 GMT"
},
{
"version": "v2",
"created": "Tue, 18 Jun 2024 06:24:11 GMT"
},
{
"version": "v3",
"created": "Mon, 2 Sep 2024 20:42:08 GMT"
},
{
"version": "v4",
"created": "Sat, 19 Oct 2024 00:49:52 GMT"
},
{
"version": "v5",
"created": "Fri, 25 Apr 2025 06:15:26 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Li",
"Lu",
""
],
[
"Zhang",
"Tianyu",
""
],
[
"Bu",
"Zhiqi",
""
],
[
"Wang",
"Suyuchen",
""
],
[
"He",
"Huan",
""
],
[
"Fu",
"Jie",
""
],
[
"Wu",
"Yonghui",
""
],
[
"Bian",
"Jiang",
""
],
[
"Chen",
"Yong",
""
],
[
"Bengio",
"Yoshua",
""
]
] |
2406.07872 | Maxim Cheremisin | M.V.Cheremisin and M.A. Zudov | How to accurately measure the mobility and viscosity of two-dimensional
carriers? | 7 pages, 5 figures, additions | null | null | null | cond-mat.mes-hall cond-mat.str-el | http://creativecommons.org/licenses/by/4.0/ | Two different methods of \emph{metrological} accuracy are proposed to
determine the mobility and viscosity of ultra clean two-dimensional electron
liquids. The experimental data analysis is based on the Gurzhi hydrodynamic
model under no-slip boundary conditions without preliminary assumptions about
carrier scattering times. The applicability of no-slip boundary conditions has
been proven. A Hall bar with several conducting channels of different widths in
a zero magnetic field and a sample with a single channel in a perpendicular
field are considered. In both cases, it was possible to accurately isolate the
ohmic part of the total measured resistance and, then find the exact mobility
and viscosity of the charge liquid. The extracted e-e scattering time is
extremely close to that obtained by other experimental group for transport
measurements of superballistic point contact. At low temperatures the e-e
scattering time demonstrates a stronger dependence compare to $1/T^{2}$
behavior predicted by Fermi liquid theory. We propose both methods as powerful
tools for viscometry and finding the mobility of two-dimensional systems.
| [
{
"version": "v1",
"created": "Wed, 12 Jun 2024 04:57:05 GMT"
},
{
"version": "v2",
"created": "Fri, 6 Sep 2024 16:31:54 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 12:39:31 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Cheremisin",
"M. V.",
""
],
[
"Zudov",
"M. A.",
""
]
] |
2406.10060 | Kota Kondo | Kota Kondo, Claudius T. Tewari, Andrea Tagliabue, Jesus Tordesillas,
Parker C. Lusk, Mason B. Peterson, Jonathan P. How | PRIMER: Perception-Aware Robust Learning-based Multiagent Trajectory
Planner | 7 pages, 3 figures | null | 10.13140/RG.2.2.14435.57124 | null | cs.RO cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In decentralized multiagent trajectory planners, agents need to communicate
and exchange their positions to generate collision-free trajectories. However,
due to localization errors/uncertainties, trajectory deconfliction can fail
even if trajectories are perfectly shared between agents. To address this
issue, we first present PARM and PARM*, perception-aware, decentralized,
asynchronous multiagent trajectory planners that enable a team of agents to
navigate uncertain environments while deconflicting trajectories and avoiding
obstacles using perception information. PARM* differs from PARM as it is less
conservative, using more computation to find closer-to-optimal solutions. While
these methods achieve state-of-the-art performance, they suffer from high
computational costs as they need to solve large optimization problems onboard,
making it difficult for agents to replan at high rates. To overcome this
challenge, we present our second key contribution, PRIMER, a learning-based
planner trained with imitation learning (IL) using PARM* as the expert
demonstrator. PRIMER leverages the low computational requirements at deployment
of neural networks and achieves a computation speed up to 5500 times faster
than optimization-based approaches.
| [
{
"version": "v1",
"created": "Fri, 14 Jun 2024 14:16:39 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Oct 2024 13:35:07 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 14:44:51 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Kondo",
"Kota",
""
],
[
"Tewari",
"Claudius T.",
""
],
[
"Tagliabue",
"Andrea",
""
],
[
"Tordesillas",
"Jesus",
""
],
[
"Lusk",
"Parker C.",
""
],
[
"Peterson",
"Mason B.",
""
],
[
"How",
"Jonathan P.",
""
]
] |
2406.10370 | Marissa Radensky | Marissa Radensky, Daniel S. Weld, Joseph Chee Chang, Pao Siangliulue,
Jonathan Bragg | Papers-to-Posts: Supporting Detailed Long-Document Summarization with an
Interactive LLM-Powered Source Outline | Revised for clearer message | null | null | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | Compressing long and technical documents (e.g., >10 pages) into shorter-form
articles (e.g., <2 pages) is critical for communicating information to
different audiences, for example, blog posts of scientific research paper or
legal briefs of dense court proceedings. While large language models (LLMs) are
powerful tools for condensing large amounts of text, current interfaces to
these models lack support for understanding and controlling what content is
included in a detailed summarizing article. Such capability is especially
important for detail- and technical-oriented domains, in which tactical
selection and coherent synthesis of key details is critical for effective
communication to the target audience. For this, we present interactive reverse
source outlines, a novel mechanism for controllable long-form summarization
featuring outline bullet points with automatic point selections that the user
can iteratively adjust to obtain an article with the desired content coverage.
We implement this mechanism in Papers-to-Posts, a new LLM-powered system for
authoring research-paper blog posts. Through a within-subjects lab study (n=20)
and a between-subjects deployment study (n=37 blog posts, 26 participants), we
compare Papers-to-Posts to a strong baseline tool that provides an
LLM-generated draft and access to free-form prompting. Under time constraints,
Papers-to-Posts significantly increases writer satisfaction with blog post
quality, particularly with respect to content coverage. Furthermore,
quantitative results showed an increase in editing power (change in text for an
amount of time or writing actions) while using Papers-to-Posts, and qualitative
results showed that participants found incorporating key research-paper
insights in their blog posts easier while using Papers-to-Posts.
| [
{
"version": "v1",
"created": "Fri, 14 Jun 2024 18:56:40 GMT"
},
{
"version": "v2",
"created": "Tue, 17 Dec 2024 18:35:27 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 00:10:42 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Radensky",
"Marissa",
""
],
[
"Weld",
"Daniel S.",
""
],
[
"Chang",
"Joseph Chee",
""
],
[
"Siangliulue",
"Pao",
""
],
[
"Bragg",
"Jonathan",
""
]
] |
2406.10432 | Lis Pereira | Peitao Han, Lis Kanashiro Pereira, Fei Cheng, Wan Jou She, Eiji
Aramaki | AMR-RE: Abstract Meaning Representations for Retrieval-Based In-Context
Learning in Relation Extraction | Accepted to NAACL 2025 SRW | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Existing in-context learning (ICL) methods for relation extraction (RE) often
prioritize language similarity over structural similarity, which can lead to
overlooking entity relationships. To address this, we propose an AMR-enhanced
retrieval-based ICL method for RE. Our model retrieves in-context examples
based on semantic structure similarity between task inputs and training
samples. Evaluations on four standard English RE datasets show that our model
outperforms baselines in the unsupervised setting across all datasets. In the
supervised setting, it achieves state-of-the-art results on three datasets and
competitive results on the fourth.
| [
{
"version": "v1",
"created": "Fri, 14 Jun 2024 22:36:08 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Sep 2024 03:35:59 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 02:23:21 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Han",
"Peitao",
""
],
[
"Pereira",
"Lis Kanashiro",
""
],
[
"Cheng",
"Fei",
""
],
[
"She",
"Wan Jou",
""
],
[
"Aramaki",
"Eiji",
""
]
] |
2406.10602 | Daniil Gurgurov | Daniil Gurgurov, Tanja B\"aumel, Tatiana Anikina | Multilingual Large Language Models and Curse of Multilinguality | null | null | 10.48550/arXiv.2406.10602 | null | cs.CL cs.CY | http://creativecommons.org/licenses/by-sa/4.0/ | Multilingual Large Language Models (LLMs) have gained large popularity among
Natural Language Processing (NLP) researchers and practitioners. These models,
trained on huge datasets, show proficiency across various languages and
demonstrate effectiveness in numerous downstream tasks. This paper navigates
the landscape of multilingual LLMs, providing an introductory overview of their
technical aspects. It explains underlying architectures, objective functions,
pre-training data sources, and tokenization methods. This work explores the
unique features of different model types: encoder-only (mBERT, XLM-R),
decoder-only (XGLM, PALM, BLOOM, GPT-3), and encoder-decoder models (mT5,
mBART). Additionally, it addresses one of the significant limitations of
multilingual LLMs - the curse of multilinguality - and discusses current
attempts to overcome it.
| [
{
"version": "v1",
"created": "Sat, 15 Jun 2024 11:31:39 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 10:53:27 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Gurgurov",
"Daniil",
""
],
[
"Bäumel",
"Tanja",
""
],
[
"Anikina",
"Tatiana",
""
]
] |
2406.10711 | Simon Lizotte | Simon Lizotte, Jean-Gabriel Young and Antoine Allard | Symmetry-driven embedding of networks in hyperbolic space | null | null | null | null | stat.CO cs.SI stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hyperbolic models are known to produce networks with properties observed
empirically in most network datasets, including heavy-tailed degree
distribution, high clustering, and hierarchical structures. As a result,
several embeddings algorithms have been proposed to invert these models and
assign hyperbolic coordinates to network data. Current algorithms for finding
these coordinates, however, do not quantify uncertainty in the inferred
coordinates. We present BIGUE, a Markov chain Monte Carlo (MCMC) algorithm that
samples the posterior distribution of a Bayesian hyperbolic random graph model.
We show that the samples are consistent with current algorithms while providing
added credible intervals for the coordinates and all network properties. We
also show that some networks admit two or more plausible embeddings, a feature
that an optimization algorithm can easily overlook.
| [
{
"version": "v1",
"created": "Sat, 15 Jun 2024 18:44:02 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2025 18:16:40 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Lizotte",
"Simon",
""
],
[
"Young",
"Jean-Gabriel",
""
],
[
"Allard",
"Antoine",
""
]
] |
2406.10719 | Orson Mengara | Orson Mengara | Trading Devil: Robust backdoor attack via Stochastic investment models
and Bayesian approach | (Last update!, a constructive comment from arxiv led to this latest
update ) Stochastic investment models and a Bayesian approach to better
modeling of uncertainty : adversarial machine learning or Stochastic market.
arXiv admin note: substantial text overlap with arXiv:2402.05967 (see this
link to the paper by : Orson Mengara) | null | null | null | cs.CR cs.LG q-fin.CP q-fin.ST stat.ML | http://creativecommons.org/licenses/by-nc-nd/4.0/ | With the growing use of voice-activated systems and speech recognition
technologies, the danger of backdoor attacks on audio data has grown
significantly. This research looks at a specific type of attack, known as a
Stochastic investment-based backdoor attack (MarketBack), in which adversaries
strategically manipulate the stylistic properties of audio to fool speech
recognition systems. The security and integrity of machine learning models are
seriously threatened by backdoor attacks, in order to maintain the reliability
of audio applications and systems, the identification of such attacks becomes
crucial in the context of audio data. Experimental results demonstrated that
MarketBack is feasible to achieve an average attack success rate close to 100%
in seven victim models when poisoning less than 1% of the training data.
| [
{
"version": "v1",
"created": "Sat, 15 Jun 2024 19:12:00 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Jun 2024 09:32:49 GMT"
},
{
"version": "v3",
"created": "Fri, 21 Jun 2024 17:42:32 GMT"
},
{
"version": "v4",
"created": "Mon, 16 Sep 2024 04:16:35 GMT"
},
{
"version": "v5",
"created": "Wed, 23 Apr 2025 18:40:00 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Mengara",
"Orson",
""
]
] |
2406.10896 | Wojciech S{\l}omian | Wojciech S{\l}omian | Oscillation inequalities for Carleson--Dunkl operator | 15 pages, no figures. Incorporated remarks from the reviewers | null | null | null | math.CA | http://creativecommons.org/licenses/by/4.0/ | In this paper, we establish estimates for the oscillation seminorm for the
so-called Carleson--Dunkl operator on weighted
$L^p(\mathbb{R},w(x)|x|^{2\alpha+1}{\rm d}x)$ spaces with power weights
$w(x)=|x|^\beta$. As a result, we obtain oscillation estimates for the standard
Carleson operator on $L_{\rm rad}^p(\mathbb{R}^n,|x|^\beta{\rm d}x)$. As a
byproduct, we obtain a transference principle for radial multipliers on $L_{\rm
rad}^p$ spaces, in the spirit of the Rubio de Francia transference principle.
| [
{
"version": "v1",
"created": "Sun, 16 Jun 2024 11:10:00 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2025 18:39:57 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Słomian",
"Wojciech",
""
]
] |
2406.12377 | Masahiro Hoshino | Masahiro Hoshino, Masaki Oshikawa, Yuto Ashida | Entanglement swapping in critical quantum spin chains | 17 pages, 13 figures | Physical Review B 111 (2025): 155143 | 10.1103/PhysRevB.111.155143 | null | quant-ph cond-mat.quant-gas cond-mat.stat-mech cond-mat.str-el | http://creativecommons.org/licenses/by/4.0/ | The transfer of quantum information between many-qubit states is a subject of
fundamental importance in quantum science and technology. We consider
entanglement swapping in critical quantum spin chains, where the entanglement
between the two chains is induced solely by the Bell-state measurements. We
employ a boundary conformal field theory (CFT) approach and describe the
measurements as conformal boundary conditions in the replicated field theory.
We show that the swapped entanglement exhibits a logarithmic scaling, whose
coefficient takes a universal value determined by the scaling dimension of the
boundary condition changing operator. We apply our framework to the critical
spin-1/2 XXZ chain and determine the universal coefficient by the boundary CFT
analysis. We also numerically verify these results by the tensor-network
calculations. Possible experimental relevance to Rydberg atom arrays is briefly
discussed.
| [
{
"version": "v1",
"created": "Tue, 18 Jun 2024 08:04:58 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 02:38:16 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Hoshino",
"Masahiro",
""
],
[
"Oshikawa",
"Masaki",
""
],
[
"Ashida",
"Yuto",
""
]
] |
2406.12766 | Fabrizio Andreatta | Fabrizio Andreatta | On a p-adic version of Narasimhan and Seshadri's theorem | Using the paper "Higgs bundles over the good reduction of a
quaternionic Shimura curve" by Mao Sheng, Jiajin Zhang, Kang Zuo, one can
constrct a counterexample to the main claim of the paper | null | null | null | math.AG math.NT | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Consider a smooth projective curve C of genus g over a complete discrete
valuation field of characteristic 0 and residue field \Fbar_p. Motivated by
Narasimhan and Seshadri's theorem, Faltings asked whether all semistable vector
bundles of degree 0 over C_{\C_p} are in the image of the p-adic Simpson
correspondence. Works of Deninger-Werner and Xu show that this is equivalent
for the vector bundle to having potentially strongly semistable reduction. We
prove that if C has good reduction, p>r(r-1) (g-1) and we consider a vector
bundle of rank r with stable reduction, the conditions of having potentially
strongly semistable reduction and of having strongly semistable reduction are
equivalent. In particular, we provide a negative answer to Faltings' question
| [
{
"version": "v1",
"created": "Tue, 18 Jun 2024 16:32:45 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 10:37:43 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Andreatta",
"Fabrizio",
""
]
] |
2406.13122 | Stefan Faridani | Stefan Faridani | Testing for Underpowered Literatures | null | null | null | null | econ.EM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | How many experimental studies would have come to different conclusions had
they been run on larger samples? I show how to estimate the expected number of
statistically significant results that a set of experiments would have reported
had their sample sizes all been counterfactually increased. The proposed
deconvolution estimator is asymptotically normal and adjusts for publication
bias. Unlike related methods, this approach requires no assumptions of any kind
about the distribution of true intervention treatment effects. An application
to randomized trials (RCTs) published in economics journals finds that doubling
every sample would increase the power of t-tests by 7.2 percentage points on
average. This effect is smaller than for non-RCTs and comparable to systematic
replications in laboratory psychology where previous studies enabled more
accurate power calculations. This suggests that RCTs are on average relatively
insensitive to sample size increases. Funders should generally consider
sponsoring more experiments rather than fewer, larger ones.
| [
{
"version": "v1",
"created": "Wed, 19 Jun 2024 00:31:41 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Apr 2025 13:58:13 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Faridani",
"Stefan",
""
]
] |
2406.13649 | Riku Rantanen | Riku Rantanen, Vladimir Eltsov | Competition of vortex core structures in superfluid $^3$He-B | 16 pages, 6 figures; revised after referee suggestions | null | 10.1103/PhysRevResearch.6.043112 | null | cond-mat.other | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Among vortex structures identified so far in superfluid $^3$He-B, the most
common are the A-phase-core vortex and the double-core vortex. According to
earlier numerical calculations, the double-core vortex is energetically favored
nearly everywhere in the $p$-$T$ phase diagram. Nevertheless, in experiments
the A-phase-core vortex has been observed down to temperatures of
$0.6T_{\mathrm{c}}$ at high pressures. We use the Ginzburg-Landau formalism to
calculate the energies of the two vortex structures in the experimentally
relevant magnetic field as well as the energy barrier for the transition
between the two structures. Assigning vanishing barrier as the boundary of the
metastability region of the A-phase-core vortex, we reproduce the
experimentally measured vortex phase diagram and provide an explanation for the
reappearance of the double-core vortex near the critical temperature
$T_{\mathrm{c}}$ at low pressures: The difference in Zeeman energy between the
two vortex structures becomes relatively more important close to
$T_{\mathrm{c}}$, and the A-phase-core vortex becomes unstable. In contrast to
the equilibrium vortex structures, we suggest that the vortex nucleation
process favors the A-phase-core vortex over the double-core vortex. Our
approach can be used to analyze competition between different vortex structures
in other unconventional superfluids and superconductors.
| [
{
"version": "v1",
"created": "Wed, 19 Jun 2024 15:54:12 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Oct 2024 08:17:53 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Rantanen",
"Riku",
""
],
[
"Eltsov",
"Vladimir",
""
]
] |
2406.16386 | Yuxuan Wan | Yuxuan Wan, Chaozheng Wang, Yi Dong, Wenxuan Wang, Shuqing Li, Yintong
Huo, Michael R. Lyu | Automatically Generating UI Code from Screenshot: A
Divide-and-Conquer-Based Approach | Accepted by FSE 2025 | null | 10.1145/3729364 | null | cs.SE cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Websites are critical in today's digital world, with over 1.11 billion
currently active and approximately 252,000 new sites launched daily. Converting
website layout design into functional UI code is a time-consuming yet
indispensable step of website development. Manual methods of converting visual
designs into functional code present significant challenges, especially for
non-experts. To explore automatic design-to-code solutions, we first conduct a
motivating study on GPT-4o and identify three types of issues in generating UI
code: element omission, element distortion, and element misarrangement. We
further reveal that a focus on smaller visual segments can help multimodal
large language models (MLLMs) mitigate these failures in the generation
process.
In this paper, we propose DCGen, a divide-and-conquer-based approach to
automate the translation of webpage design to UI code. DCGen starts by dividing
screenshots into manageable segments, generating code for each segment, and
then reassembling them into complete UI code for the entire screenshot. We
conduct extensive testing with a dataset comprised of real-world websites and
various MLLMs and demonstrate that DCGen achieves up to a 15% improvement in
visual similarity and 8% in code similarity for large input images. Human
evaluations show that DCGen can help developers implement webpages
significantly faster and more similar to the UI designs. To the best of our
knowledge, DCGen is the first segment-aware MLLM-based approach for generating
UI code directly from screenshots.
| [
{
"version": "v1",
"created": "Mon, 24 Jun 2024 07:58:36 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Oct 2024 11:22:53 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 15:12:34 GMT"
}
] | 2025-04-28T00:00:00 | [
[
"Wan",
"Yuxuan",
""
],
[
"Wang",
"Chaozheng",
""
],
[
"Dong",
"Yi",
""
],
[
"Wang",
"Wenxuan",
""
],
[
"Li",
"Shuqing",
""
],
[
"Huo",
"Yintong",
""
],
[
"Lyu",
"Michael R.",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.