id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 6
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-04-25 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2409.17290 | Andrea Tononi | Andrea Tononi, Maciej Lewenstein | Temporal Bell inequalities in non-relativistic many-body physics | 9 pages, 1 figure; | Quantum Sci. Technol. 10, 03LT01 (2025) | 10.1088/2058-9565/adcbce | null | quant-ph cond-mat.quant-gas | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Analyzing the spreading of information in many-body systems is crucial to
understanding their quantum dynamics. At the most fundamental level, this task
is accomplished by Bell inequalities, whose violation by quantum mechanics
implies that information cannot always be stored locally. While Bell-like
inequalities, such as the one of Clauser and Horne, envisage a situation in
which two parties perform measurements on systems at different positions, one
could formulate temporal inequalities, in which the two parties measure at
different times. However, for causally-connected measurement events, these
extensions are compatible with a local description, so that no
intrinsically-quantum information spreading is involved in such temporal
correlations. Here we show that a temporal Clauser-Horne inequality for two
spins is violated for a nonzero time interval between the measurements if the
two measured parties are connected by a spin chain. Since the chain constitutes
the sole medium for the spreading of quantum information, it prevents the
immediate vanishing of Bell correlations after the first measurement and it
induces violation revivals. The dynamics we analyze shows that, as expected in
a non-relativistic setup, the spreading of information is fundamentally limited
by the Lieb-Robinson bound. New insights on many-body quantum dynamics could
emerge through future applications of our temporal Bell inequality to more
general systems.
| [
{
"version": "v1",
"created": "Wed, 25 Sep 2024 19:00:47 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Nov 2024 21:15:23 GMT"
},
{
"version": "v3",
"created": "Fri, 4 Apr 2025 12:08:37 GMT"
},
{
"version": "v4",
"created": "Thu, 17 Apr 2025 10:21:14 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Tononi",
"Andrea",
""
],
[
"Lewenstein",
"Maciej",
""
]
] |
2409.17505 | Diego Martinez-Taboada | Diego Martinez-Taboada, Aaditya Ramdas | Sequential Kernelized Stein Discrepancy | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a sequential version of the kernelized Stein discrepancy
goodness-of-fit test, which allows for conducting goodness-of-fit tests for
unnormalized densities that are continuously monitored and adaptively stopped.
That is, the sample size need not be fixed prior to data collection; the
practitioner can choose whether to stop the test or continue to gather evidence
at any time while controlling the false discovery rate. In stark contrast to
related literature, we do not impose uniform boundedness on the Stein kernel.
Instead, we exploit the potential boundedness of the Stein kernel at arbitrary
point evaluations to define test martingales, that give way to the subsequent
novel sequential tests. We prove the validity of the test, as well as an
asymptotic lower bound for the logarithmic growth of the wealth process under
the alternative. We further illustrate the empirical performance of the test
with a variety of distributions, including restricted Boltzmann machines.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 03:24:59 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 01:52:49 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Martinez-Taboada",
"Diego",
""
],
[
"Ramdas",
"Aaditya",
""
]
] |
2409.17803 | Sanchari Bhattacharyya | Sweta Baradia, Sanchari Bhattacharyya, Anindya Datta, Suchandra Dutta,
Suvankar Roy Chowdhury and Subir Sarkar | Search for Dark Matter in association with a Higgs boson at the LHC: A
model independent study | 24 Pages, 13 Figures, 9 Tables | null | null | null | hep-ph hep-ex | http://creativecommons.org/licenses/by/4.0/ | Astrophysical and cosmological observations strongly suggest the existence of
Dark Matter\,(DM). Experiments at the Large Hadron Collider\,(LHC) have the
potential to probe the particle nature of the DM. In the present work, we
investigate the potential of the mono-Higgs plus Missing Transverse Energy
signature at the LHC to search for a relatively light fermionic dark matter
candidate using the framework of Effective Field Theory. In our study, the DM
interacts with the Standard Model\,(SM) via dimension-6 and dimension-7
effective operators involving the Higgs and the gauge bosons. Although, our
analysis is independent of any Ultra Violet complete dynamics of DM, such
interactions can be realized in an extension of the SM where the gauge group is
extended minimally by adding an extra $U(1)$. Both cut-based and Boosted
Decision Tree\,(BDT) discriminators are used to estimate and optimize the
signal sensitivity over the SM backgrounds, assuming an integrated luminosity
of $3000~fb^{-1}$ at $\sqrt{s}=14$ TeV at the High Luminosity phase of the
LHC\,(HL-LHC). It can be seen that in the best scenario, atleast $4\sigma$
significance can be achieved for relic masses upto 200 GeV, showcasing the
prospects of this search at the HL-LHC. This study provides a foundation for
future explorations in this direction.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 12:55:16 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 14:48:39 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Baradia",
"Sweta",
""
],
[
"Bhattacharyya",
"Sanchari",
""
],
[
"Datta",
"Anindya",
""
],
[
"Dutta",
"Suchandra",
""
],
[
"Chowdhury",
"Suvankar Roy",
""
],
[
"Sarkar",
"Subir",
""
]
] |
2409.17948 | N Furey | N. Furey, M.J. Hughes | Three Generations and a Trio of Trialities | 11 pages, 6 figures | Phys. Lett. B (2025) | 10.1016/j.physletb.2025.139473 | HU-EP-24/26 | hep-ph hep-th | http://creativecommons.org/licenses/by/4.0/ | We identify the Standard Model's $\mathfrak{su}(3)\oplus
\mathfrak{su}(2)\oplus \mathfrak{u}(1)$ internal symmetries within the triality
symmetries $\mathfrak{tri}(\mathbb{C}) \oplus \mathfrak{tri}(\mathbb{H}) \oplus
\mathfrak{tri}(\mathbb{O})$. From here, the corresponding Standard Model group
action is applied to the triality triple $\left( \Psi_+, \Psi_-,V\right)$ for
$\Psi_+, \Psi_-, V \in \mathbb{C}\otimes\mathbb{H}\otimes\mathbb{O}$. Together,
$\Psi_+$ and $\Psi_-$ provide the correct irreducible representations for two
generations. Owing to a certain Cartan Factorization, which we define, $V$
provides the irreducible representations for a third generation. Said more
explicitly in another way, division algebraic multiplication merges a third
generation of spinor representations into a set of scalar bosons. This set of
scalar bosons includes the familiar Standard Model Higgs representation.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 15:22:10 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Oct 2024 17:50:35 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Furey",
"N.",
""
],
[
"Hughes",
"M. J.",
""
]
] |
2409.18105 | Thijs Becker | T. Becker, R. Smet, B. Macharis, K. Vanthournout | Effect of electric vehicles, heat pumps, and solar panels on low-voltage
feeders: Evidence from smart meter profiles | Published version | Sustainable Energy, Grids and Networks, Volume 42, 2025 | 10.1016/j.segan.2025.101705 | null | eess.SY cs.CY cs.SY stat.AP | http://creativecommons.org/licenses/by/4.0/ | Electric vehicles (EVs), heat pumps (HPs) and solar panels are low-carbon
technologies (LCTs) that are being connected to the low-voltage grid (LVG) at a
rapid pace. One of the main hurdles to understand their impact on the LVG is
the lack of recent, large electricity consumption datasets, measured in
real-world conditions. We investigated the contribution of LCTs to the size and
timing of peaks on LV feeders by using a large dataset of 42,089 smart meter
profiles of residential LVG customers. These profiles were measured in 2022 by
Fluvius, the distribution system operator (DSO) of Flanders, Belgium. The
dataset contains customers that proactively requested higher-resolution smart
metering data, and hence is biased towards energy-interested people. LV feeders
of different sizes were statistically modelled with a profile sampling
approach. For feeders with 40 connections, we found a contribution to the
feeder peak of 1.2 kW for a HP, 1.4 kW for an EV and 2.0 kW for an EV charging
faster than 6.5 kW. A visual analysis of the feeder-level loads shows that the
classical duck curve is replaced by a night-camel curve for feeders with only
HPs and a night-dromedary curve for feeders with only EVs charging faster than
6.5 kW. Consumption patterns will continue to change as the energy transition
is carried out, because of e.g. dynamic electricity tariffs or increased
battery capacities. Our introduced methods are simple to implement, making it a
useful tool for DSOs that have access to smart meter data to monitor changing
consumption patterns.
| [
{
"version": "v1",
"created": "Wed, 11 Sep 2024 10:15:56 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 06:40:24 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Becker",
"T.",
""
],
[
"Smet",
"R.",
""
],
[
"Macharis",
"B.",
""
],
[
"Vanthournout",
"K.",
""
]
] |
2409.19859 | Siming He | Mengyang Gu and Siming He | Mixing, Enhanced Dissipation and Phase Transition in the Kinetic Vicsek
Model | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we study the kinetic Vicsek model, which serves as a starting
point for describing the polarization phenomena observed in the experiments of
fibroblasts moving on liquid crystalline substrates. The long-time behavior of
the kinetic equation is analyzed, revealing that, within specific parameter
regimes, the mixing and enhanced dissipation phenomena stabilize the dynamics
and ensure effective information communication among agents. Consequently, the
solution exhibits features similar to those of a spatially-homogeneous system.
As a result, we confirm the phase transition observed in the agent-based Vicsek
model at the kinetic level.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 01:34:28 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 12:58:36 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Gu",
"Mengyang",
""
],
[
"He",
"Siming",
""
]
] |
2409.19969 | Heiko Gimperlein | Heiko Gimperlein, Magnus Goffeng | Riesz energies and the magnitude of manifolds | 12 pages | null | 10.1090/proc/17203 | null | math.DG math.AP math.MG math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the geometric significance of Leinster's magnitude invariant. For
closed manifolds we find a precise relation with Brylinski's beta function and
therefore with classical invariants of knots and submanifolds. In the special
case of compact homogeneous spaces we obtain an elementary proof that the
residues of the beta function contain the same geometric information as the
asymptotic expansion of the magnitude function. For general closed manifolds we
use the recent pseudodifferential analysis of the magnitude operator to relate
these via an interpolating polynomial family. Beyond manifolds, the relation
with the Brylinski beta function allows to deduce unexpected properties of the
magnitude function for the $p$-adic integers.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 05:47:06 GMT"
},
{
"version": "v2",
"created": "Wed, 2 Oct 2024 13:32:40 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Gimperlein",
"Heiko",
""
],
[
"Goffeng",
"Magnus",
""
]
] |
2409.20245 | Yousef Kloob | Yousef Kloob, Mohammad Al-Jarrah, Emad Alsusa | A Framework for Holistic KLD-based Waveform Design for
Multi-User-Multi-Target ISAC Systems | 13 pages | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces a novel framework aimed at designing integrated
waveforms for robust integrated sensing and communication systems. The system
model consists of a multiple-input multiple-output (MIMO) base station that
simultaneously serves communication user equipments and detects multiple
targets using a shared-antenna deployment scenario. By leveraging
Kullback-Leibler divergence to holistically characterise both communication and
sensing subsystems, three optimisation problems are formulated: (i) radar
waveform KLD maximisation under communication constraints, (ii) communication
waveform KLD maximisation subject to radar KLD requirements, and (iii) an
integrated waveform KLD-based optimisation for ISAC that jointly balances both
subsystems. The first two problems are solved using a projected gradient method
with adaptive penalties for the radar waveforms and a gradient assisted
interior point method for the communication waveforms. The third, integrated
waveform optimisation approach adopts an alternating direction method of
multipliers framework to unify radar and communication waveform designs into a
single integrated optimisation, thereby strengthening the trade-off between
sensing and communication objectives and achieving higher overall performance
than the individual radar- or communication-only techniques. Unlike most
existing ISAC waveform designs that regard communication signals solely as
interference for sensing, the proposed framework utilises the holistic ISAC
waveform-that is, the superimposed communication and sensing signals to boost
detection performance in the radar subsystem. Simulation results show
significant improvements in both radar detection and communication reliability
compared with conventional zero-forcing beamforming and identity covariance
radar baselines, demonstrating the promise of KLD-based waveform designs for
next-generation ISAC networks.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 12:34:04 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Oct 2024 09:41:46 GMT"
},
{
"version": "v3",
"created": "Thu, 13 Mar 2025 13:38:25 GMT"
},
{
"version": "v4",
"created": "Thu, 17 Apr 2025 10:56:59 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Kloob",
"Yousef",
""
],
[
"Al-Jarrah",
"Mohammad",
""
],
[
"Alsusa",
"Emad",
""
]
] |
2410.03165 | Yuri Prokhorov | Shigefumi Mori and Yuri Prokhorov | Toward the classification of threefold extremal contractions with
one-dimensional fibers | 32 page , LaTeX, revised version | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | An extremal curve germ is a germ of a threefold $X$ with terminal
singularities along a connected reduced complete curve~$C$ such that there
exists a $K_X$-negative contraction $f : X \to Z$ with~$C$ being a fiber. We
give a rough classification of extremal curve germs with reducible central
curve~$C$.
| [
{
"version": "v1",
"created": "Fri, 4 Oct 2024 05:53:56 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 08:51:24 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Mori",
"Shigefumi",
""
],
[
"Prokhorov",
"Yuri",
""
]
] |
2410.03309 | Jonas Ellert | Gabriel Bathie, Jonas Ellert, Tatiana Starikovskaya | Small Space Encoding and Recognition of $k$-Palindromic Prefixes | null | null | null | null | cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Palindromes are non-empty strings that read the same forward and backward.
The problem of recognizing strings that can be represented as the concatenation
of even-length palindromes, the concatenation of palindromes of length at least
two, and the concatenation of exactly $k$ palindromes was introduced in the
seminal paper of Knuth, Morris, and Pratt [SIAM J. Comput., 1977].
In this work, we study the problem of recognizing so-called $k$-palindromic
strings, which can be represented as the concatenation of exactly $k$
palindromes. We show the following results:
1. First, we show a structural characterization of the set of all
$k$-palindromic prefixes of a string by representing it as a union of a small
number of highly structured string sets, called affine prefix sets.
Representing the lengths of the $k$-palindromic prefixes in this way requires
$O(6^{k^2} \cdot \log^k n)$ space. By constructing a lower bound, we show that
the space complexity is optimal up to polylogarithmic factors for reasonably
small values of $k$.
2. Secondly, we derive a read-only algorithm that, given a string $T$ of
length $n$ and an integer $k$, computes a compact representation of
$i$-palindromic prefixes of $T$, for all $1 \le i \le k$. The algorithm uses
$O(n \cdot 6^{k^2} \cdot \log^k n)$ time and $O(6^{k^2} \cdot \log^k n)$ space.
3. Finally, we also give a read-only algorithm for computing the palindromic
length of $T$, which is the smallest $\ell$ such that $T$ is
$\ell$-palindromic. Here, we achieve $O(n \cdot 6^{\ell^2} \cdot
\log^{\lceil{\ell/2 \rceil}} n)$ time and $O(6^{\ell^2} \cdot
\log^{\lceil{\ell/2\rceil}} n)$ space. For some values of $\ell$, this is the
first algorithm for palindromic length that uses $o(n)$ additional working
space on top of the input.
| [
{
"version": "v1",
"created": "Fri, 4 Oct 2024 10:45:06 GMT"
},
{
"version": "v2",
"created": "Wed, 13 Nov 2024 12:55:24 GMT"
},
{
"version": "v3",
"created": "Sun, 9 Feb 2025 10:14:48 GMT"
},
{
"version": "v4",
"created": "Thu, 17 Apr 2025 14:41:38 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Bathie",
"Gabriel",
""
],
[
"Ellert",
"Jonas",
""
],
[
"Starikovskaya",
"Tatiana",
""
]
] |
2410.03802 | Pasquale Claudio Africa | Giuseppe Alessio D'Inverno, Saeid Moradizadeh, Sajad Salavatidezfouli,
Pasquale Claudio Africa, Gianluigi Rozza | Mesh-Informed Reduced Order Models for Aneurysm Rupture Risk Prediction | null | null | null | null | physics.med-ph cs.LG cs.NA math.NA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The complexity of the cardiovascular system needs to be accurately reproduced
in order to promptly acknowledge health conditions; to this aim, advanced
multifidelity and multiphysics numerical models are crucial. On one side, Full
Order Models (FOMs) deliver accurate hemodynamic assessments, but their high
computational demands hinder their real-time clinical application. In contrast,
Reduced Order Models (ROMs) provide more efficient yet accurate solutions,
essential for personalized healthcare and timely clinical decision-making. In
this work, we explore the application of computational fluid dynamics (CFD) in
cardiovascular medicine by integrating FOMs with ROMs for predicting the risk
of aortic aneurysm growth and rupture. Wall Shear Stress (WSS) and the
Oscillatory Shear Index (OSI), sampled at different growth stages of the
thoracic aortic aneurysm, are predicted by means of Graph Neural Networks
(GNNs). GNNs exploit the natural graph structure of the mesh obtained by the
Finite Volume (FV) discretization, taking into account the spatial local
information, regardless of the dimension of the input graph. Our experimental
validation framework yields promising results, confirming our method as a valid
alternative that overcomes the curse of dimensionality.
| [
{
"version": "v1",
"created": "Fri, 4 Oct 2024 09:39:15 GMT"
},
{
"version": "v2",
"created": "Wed, 29 Jan 2025 08:19:32 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 12:45:16 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"D'Inverno",
"Giuseppe Alessio",
""
],
[
"Moradizadeh",
"Saeid",
""
],
[
"Salavatidezfouli",
"Sajad",
""
],
[
"Africa",
"Pasquale Claudio",
""
],
[
"Rozza",
"Gianluigi",
""
]
] |
2410.03931 | Aled Williams | Aled Williams and Yilun Cai | Insights into Weighted Sum Sampling Approaches for Multi-Criteria
Decision Making Problems | 32 pages, 5 figures | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | In this paper we explore several approaches for sampling weight vectors in
the context of weighted sum scalarisation approaches for solving multi-criteria
decision making (MCDM) problems. This established method converts a
multi-objective problem into a (single) scalar optimisation problem. It does so
by assigning weights to each objective. We outline various methods to select
these weights, with a focus on ensuring computational efficiency and avoiding
redundancy. The challenges and computational complexity of these approaches are
explored and numerical examples are provided. The theoretical results
demonstrate the trade-offs between systematic and randomised weight generation
techniques, highlighting their performance for different problem settings.
These sampling approaches will be tested and compared computationally in an
upcoming paper.
| [
{
"version": "v1",
"created": "Fri, 4 Oct 2024 21:16:11 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 15:25:44 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Williams",
"Aled",
""
],
[
"Cai",
"Yilun",
""
]
] |
2410.05167 | Zachary Novack | Zachary Novack, Ge Zhu, Jonah Casebeer, Julian McAuley, Taylor
Berg-Kirkpatrick, Nicholas J. Bryan | Presto! Distilling Steps and Layers for Accelerating Music Generation | Accepted as Spotlight at ICLR 2025 | null | null | null | cs.SD cs.AI cs.LG eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite advances in diffusion-based text-to-music (TTM) methods, efficient,
high-quality generation remains a challenge. We introduce Presto!, an approach
to inference acceleration for score-based diffusion transformers via reducing
both sampling steps and cost per step. To reduce steps, we develop a new
score-based distribution matching distillation (DMD) method for the EDM-family
of diffusion models, the first GAN-based distillation method for TTM. To reduce
the cost per step, we develop a simple, but powerful improvement to a recent
layer distillation method that improves learning via better preserving hidden
state variance. Finally, we combine our step and layer distillation methods
together for a dual-faceted approach. We evaluate our step and layer
distillation methods independently and show each yield best-in-class
performance. Our combined distillation method can generate high-quality outputs
with improved diversity, accelerating our base model by 10-18x (230/435ms
latency for 32 second mono/stereo 44.1kHz, 15x faster than comparable SOTA) --
the fastest high-quality TTM to our knowledge. Sound examples can be found at
https://presto-music.github.io/web/.
| [
{
"version": "v1",
"created": "Mon, 7 Oct 2024 16:24:18 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 17:37:06 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Novack",
"Zachary",
""
],
[
"Zhu",
"Ge",
""
],
[
"Casebeer",
"Jonah",
""
],
[
"McAuley",
"Julian",
""
],
[
"Berg-Kirkpatrick",
"Taylor",
""
],
[
"Bryan",
"Nicholas J.",
""
]
] |
2410.05576 | David Thorne | David Thorne, Nathan Chan, Yanlong Ma, Christa S. Robison, Philip R.
Osteen, Brett T. Lopez | Submodular Optimization for Keyframe Selection & Usage in SLAM | Accepted to the International Conference on Robotics and Automation
(ICRA) 2025 | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Keyframes are LiDAR scans saved for future reference in Simultaneous
Localization And Mapping (SLAM), but despite their central importance most
algorithms leave choices of which scans to save and how to use them to wasteful
heuristics. This work proposes two novel keyframe selection strategies for
localization and map summarization, as well as a novel approach to submap
generation which selects keyframes that best constrain localization. Our
results show that online keyframe selection and submap generation reduce the
number of saved keyframes and improve per scan computation time without
compromising localization performance. We also present a map summarization
feature for quickly capturing environments under strict map size constraints.
| [
{
"version": "v1",
"created": "Tue, 8 Oct 2024 00:24:13 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 18:17:44 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Thorne",
"David",
""
],
[
"Chan",
"Nathan",
""
],
[
"Ma",
"Yanlong",
""
],
[
"Robison",
"Christa S.",
""
],
[
"Osteen",
"Philip R.",
""
],
[
"Lopez",
"Brett T.",
""
]
] |
2410.05967 | Grigoris Kopsacheilis | Grigoris Kopsacheilis, Wilhelm Winter | Diagonal comparison of ample C*-diagonals | v2: 26 pages; minor revision, added Remarks 1.14 and 4.2 and added
further details in the proof of Proposition 4.2 (now 4.3); final version, to
appear on Int. Math. Res. Not. (IMRN) | null | null | null | math.OA math.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce diagonal comparison, a regularity property of diagonal pairs
where the sub-C*-algebra has totally disconnected spectrum, and establish its
equivalence with the concurrence of strict comparison of the ambient C*-algebra
and dynamical comparison of the underlying dynamics induced by the partial
action of the normalisers. As an application, we show that for diagonal pairs
arising from principal minimal transformation groupoids with totally
disconnected unit space, diagonal comparison is equivalent to tracial
Z-stability of the pair and that it is implied by finite diagonal dimension.
In-between, we show that any projection of the diagonal sub-C*-algebra can be
uniformly tracially divided, and explore a property of conditional expectations
onto abelian sub-C*-algebras, namely containment of every positive element in
the hereditary subalgebra generated by its conditional expectation. We show
that the expectation associated to a C*-pair with finite diagonal dimension is
always hereditary in that sense, and we give an example where this property
does not occur.
| [
{
"version": "v1",
"created": "Tue, 8 Oct 2024 12:16:12 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 09:38:52 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Kopsacheilis",
"Grigoris",
""
],
[
"Winter",
"Wilhelm",
""
]
] |
2410.06066 | Aurelien Buchet | Aur\'elien Buchet and Cristel Pelsser | An Analysis of QUIC Connection Migration in the Wild | 7 pages, 4 figures | ACM SIGCOMM Computer Communication Review, 55(1), 3-9 (2025) | 10.1145/3727063.3727066 | null | cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As QUIC gains attention, more applications that leverage its capabilities are
emerging. These include defenses against on-path IP tracking and traffic
analysis. However, the deployment of the underlying required support for
connection migration remains largely unexplored. This paper provides a
comprehensive examination of the support of the QUIC connection migration
mechanism over the Internet. We perform Internet-wide scans revealing that
despite a rapid evolution in the deployment of QUIC on web servers, some of the
most popular destinations do not support connection migration yet.
| [
{
"version": "v1",
"created": "Tue, 8 Oct 2024 14:13:36 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Buchet",
"Aurélien",
""
],
[
"Pelsser",
"Cristel",
""
]
] |
2410.08244 | Nuria Rodr\'iguez-Barroso | Nuria Rodr\'iguez-Barroso, M. Victoria Luz\'on and Francisco Herrera | RAB$^2$-DEF: Dynamic and explainable defense against adversarial attacks
in Federated Learning to fair poor clients | null | Nuria Rodr\'iguez-Barroso, M. Victoria Luz\'on, Francisco Herrera.
RAB2-DEF: Dynamic and explainable defense against adversarial attacks in
Federated Learning to fair poor clients. (2025) Machine Intelligence Research | 10.1007/s11633-025-1557-1 | null | cs.CR cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | At the same time that artificial intelligence is becoming popular, concern
and the need for regulation is growing, including among other requirements the
data privacy. In this context, Federated Learning is proposed as a solution to
data privacy concerns derived from different source data scenarios due to its
distributed learning. The defense mechanisms proposed in literature are just
focused on defending against adversarial attacks and the performance, leaving
aside other important qualities such as explainability, fairness to poor
quality clients, dynamism in terms of attacks configuration and generality in
terms of being resilient against different kinds of attacks. In this work, we
propose RAB$^2$-DEF, a $\textbf{r}$esilient $\textbf{a}$gainst
$\textbf{b}\text{yzantine}$ and $\textbf{b}$ackdoor attacks which is
$\textbf{d}$ynamic, $\textbf{e}$xplainable and $\textbf{f}$air to poor clients
using local linear explanations. We test the performance of RAB$^2$-DEF in
image datasets and both byzantine and backdoor attacks considering the
state-of-the-art defenses and achieve that RAB$^2$-DEF is a proper defense at
the same time that it boosts the other qualities towards trustworthy artificial
intelligence.
| [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 09:32:59 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 07:54:52 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Rodríguez-Barroso",
"Nuria",
""
],
[
"Luzón",
"M. Victoria",
""
],
[
"Herrera",
"Francisco",
""
]
] |
2410.09708 | Jielong Yang | Jielong Yang, Rui Ding, Feng Ji, Hongbin Wang, Linbo Xie | Control the GNN: Utilizing Neural Controller with Lyapunov Stability for
Test-Time Feature Reconstruction | This work has been submitted to the IEEE for possible publication | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The performance of graph neural networks (GNNs) is susceptible to
discrepancies between training and testing sample distributions. Prior studies
have attempted to mitigating the impact of distribution shift by reconstructing
node features during the testing phase without modifying the model parameters.
However, these approaches lack theoretical analysis of the proximity between
predictions and ground truth at test time. In this paper, we propose a novel
node feature reconstruction method grounded in Lyapunov stability theory.
Specifically, we model the GNN as a control system during the testing phase,
considering node features as control variables. A neural controller that
adheres to the Lyapunov stability criterion is then employed to reconstruct
these node features, ensuring that the predictions progressively approach the
ground truth at test time. We validate the effectiveness of our approach
through extensive experiments across multiple datasets, demonstrating
significant performance improvements.
| [
{
"version": "v1",
"created": "Sun, 13 Oct 2024 03:34:19 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 09:27:02 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Yang",
"Jielong",
""
],
[
"Ding",
"Rui",
""
],
[
"Ji",
"Feng",
""
],
[
"Wang",
"Hongbin",
""
],
[
"Xie",
"Linbo",
""
]
] |
2410.10079 | Ogulcan Orsel | Ogulcan E. Orsel, Jiho Noh, Penghao Zhu, Jieun Yim, Taylor L. Hughes,
Ronny Thomale, and Gaurav Bahl | Giant non-reciprocity and gyration through modulation-induced
Hatano-Nelson coupling in integrated photonics | null | Phys. Rev. Lett. 134, 153801 (2025) | 10.1103/PhysRevLett.134.153801 | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Asymmetric energy exchange interactions, also known as Hatano-Nelson type
couplings, enable the study of non-Hermitian physics and associated phenomena
like the non-Hermitian skin effect and exceptional points (EP). Since these
interactions are by definition non-reciprocal, there have been very few options
for real-space implementations in integrated photonics. In this work, we show
that real-space asymmetric couplings are readily achievable in integrated
photonic systems through time-domain dynamic modulation. We experimentally
study this concept using a two-resonator photonic molecule produced in a
lithium niobate on insulator platform that is electro-optically modulated by rf
stimuli. We demonstrate the dynamic tuning of the Hatano-Nelson coupling
between the resonators, surpassing the asymmetry that has been achieved in
previous work, to reach an EP for the first time. We are additionally able to
flip the relative sign of the couplings for opposite directions by going past
the EP. Using this capability, we show that the through-chain transport can be
configured to exhibit both giant (60 dB) optical contrast as well as photonic
gyration or non-reciprocal pi phase contrast.
| [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 01:48:47 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Orsel",
"Ogulcan E.",
""
],
[
"Noh",
"Jiho",
""
],
[
"Zhu",
"Penghao",
""
],
[
"Yim",
"Jieun",
""
],
[
"Hughes",
"Taylor L.",
""
],
[
"Thomale",
"Ronny",
""
],
[
"Bahl",
"Gaurav",
""
]
] |
2410.10291 | Xiangru Zhu | Xiangru Zhu, Penglei Sun, Yaoxian Song, Yanghua Xiao, Zhixu Li,
Chengyu Wang, Jun Huang, Bei Yang, Xiaoxiao Xu | Evaluating Semantic Variation in Text-to-Image Synthesis: A Causal
Perspective | Accepted by ICLR 2025 | null | null | null | cs.CL cs.AI cs.CV cs.LG cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate interpretation and visualization of human instructions are crucial
for text-to-image (T2I) synthesis. However, current models struggle to capture
semantic variations from word order changes, and existing evaluations, relying
on indirect metrics like text-image similarity, fail to reliably assess these
challenges. This often obscures poor performance on complex or uncommon
linguistic patterns by the focus on frequent word combinations. To address
these deficiencies, we propose a novel metric called SemVarEffect and a
benchmark named SemVarBench, designed to evaluate the causality between
semantic variations in inputs and outputs in T2I synthesis. Semantic variations
are achieved through two types of linguistic permutations, while avoiding
easily predictable literal variations. Experiments reveal that the
CogView-3-Plus and Ideogram 2 performed the best, achieving a score of 0.2/1.
Semantic variations in object relations are less understood than attributes,
scoring 0.07/1 compared to 0.17-0.19/1. We found that cross-modal alignment in
UNet or Transformers plays a crucial role in handling semantic variations, a
factor previously overlooked by a focus on textual encoders. Our work
establishes an effective evaluation framework that advances the T2I synthesis
community's exploration of human instruction understanding. Our benchmark and
code are available at https://github.com/zhuxiangru/SemVarBench .
| [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 08:45:35 GMT"
},
{
"version": "v2",
"created": "Fri, 18 Oct 2024 09:26:46 GMT"
},
{
"version": "v3",
"created": "Tue, 1 Apr 2025 21:43:19 GMT"
},
{
"version": "v4",
"created": "Thu, 17 Apr 2025 08:31:14 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Zhu",
"Xiangru",
""
],
[
"Sun",
"Penglei",
""
],
[
"Song",
"Yaoxian",
""
],
[
"Xiao",
"Yanghua",
""
],
[
"Li",
"Zhixu",
""
],
[
"Wang",
"Chengyu",
""
],
[
"Huang",
"Jun",
""
],
[
"Yang",
"Bei",
""
],
[
"Xu",
"Xiaoxiao",
""
]
] |
2410.11297 | Ran Si | Ran Si, Yanting Li, Kai Wang, Chongyang Chen, Gediminas Gaigalas,
Michel Godefroid, Per J\"onsson | Graspg -- An extension to Grasp2018 based on Configuration State
Function Generators | null | Computer Physics Communications 312 (2025) 109604 | 10.1016/j.cpc.2025.109604 | null | physics.atom-ph | http://creativecommons.org/licenses/by/4.0/ | The Graspg program package is an extension of Grasp2018 [Comput. Phys.
Commun. 237 (2019) 184-187] based on configuration state function generators
(CSFGs). The generators keep spin-angular integrations at a minimum and reduce
substantially the execution time and the memory requirements for large-scale
multiconfiguration Dirac-Hartree-Fock (MCDHF) and relativistic configuration
interaction (CI) atomic structure calculations. The package includes the
improvements reported in [Atoms 11 (2023) 12] in terms of redesigned and
efficient constructions of direct- and exchange potentials, as well as Lagrange
multipliers, and additional parallelization of the diagonalization procedure.
Tools have been developed for predicting configuration state functions (CSFs)
that are unimportant and can be discarded for large MCDHF or CI calculations
based on results from smaller calculations, thus providing efficient methods
for a priori condensation. The package provides a seamless interoperability
with Grasp2018. From extensive test runs and benchmarking, we have demonstrated
reductions in the execution time and disk file sizes with factors of 37 and 98,
respectively, for MCDHF calculations based on large orbital sets compared to
corresponding Grasp2018 calculations. For CI calculations, reductions of the
execution time with factors over 200 have been attained. With a sensible use of
the new possibilities for a priori condensation, CI calculations with nominally
hundreds of millions of CSFs can be handled.
| [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 05:36:43 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Si",
"Ran",
""
],
[
"Li",
"Yanting",
""
],
[
"Wang",
"Kai",
""
],
[
"Chen",
"Chongyang",
""
],
[
"Gaigalas",
"Gediminas",
""
],
[
"Godefroid",
"Michel",
""
],
[
"Jönsson",
"Per",
""
]
] |
2410.11329 | Itay Shomroni | Lina Halef, Itay Shomroni | Route to hyperchaos in quadratic optomechanics | Major revision, in particular identification of mechanism for
hyperchaos. Comments welcome! | null | null | null | nlin.CD quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hyperchaos is a qualitatively stronger form of chaos, in which several
degrees of freedom contribute simultaneously to exponential divergence of small
changes. A hyperchaotic dynamical system is therefore even more unpredictable
than a chaotic one, and has a higher fractal dimension. While hyperchaos has
been studied extensively over the last decades, only a few experimental systems
are known to exhibit hyperchaotic dynamics. Here we introduce hyperchaos in the
context of cavity optomechanics, in which light inside an optical resonator
interacts with a suspended oscillating mass. We show that hyperchaos can arise
in optomechanical systems with quadratic coupling and is well within reach of
current experiments. We compute the two positive Lyapunov exponents,
characteristic of hyperchaos, and independently verify the correlation
dimension. We also identify a possible mechanism for the emergence of
hyperchaos. As systems designed for high-precision measurements, optomechanical
systems enable direct measurement of all four dynamical variables and therefore
the full reconstruction of the hyperchaotic attractor. Our results may
contribute to better understanding of nonlinear systems and the
chaos-hyperchaos transition, and allow the study of hyperchaos in the quantum
regime.
| [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 06:57:07 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 11:58:51 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Halef",
"Lina",
""
],
[
"Shomroni",
"Itay",
""
]
] |
2410.11982 | Alexander Gorokhovsky | Alexander Gorokhovsky, Erik van Erp | A note on traces for the Heisenberg calculus | null | null | null | MPIM-Bonn-2025 | math.FA math.AP math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In previous work, we gave a local formula for the index of Heisenberg
elliptic operators on contact manifolds. We constructed a cocycle in periodic
cyclic cohomology which, when paired with the Connes-Chern character of the
principal Heisenberg symbol, calculates the index. A crucial ingredient of our
index formula was a new trace on the algebra of Heisenberg pseudodifferential
operators. The construction of this trace was rather involved. In the present
paper, we clarify the nature of this trace.
| [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 18:45:14 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 23:11:14 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Gorokhovsky",
"Alexander",
""
],
[
"van Erp",
"Erik",
""
]
] |
2410.12229 | Ziqiang Cui | Ziqiang Cui, Yunpeng Weng, Xing Tang, Fuyuan Lyu, Dugang Liu, Xiuqiang
He, Chen Ma | Comprehending Knowledge Graphs with Large Language Models for
Recommender Systems | Accepted as a full paper by SIGIR'25 | null | 10.1145/3726302.3729932 | null | cs.IR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, the introduction of knowledge graphs (KGs) has significantly
advanced recommender systems by facilitating the discovery of potential
associations between items. However, existing methods still face several
limitations. First, most KGs suffer from missing facts or limited scopes.
Second, existing methods convert textual information in KGs into IDs, resulting
in the loss of natural semantic connections between different items. Third,
existing methods struggle to capture high-order connections in the global KG.
To address these limitations, we propose a novel method called CoLaKG, which
leverages large language models (LLMs) to improve KG-based recommendations. The
extensive knowledge and remarkable reasoning capabilities of LLMs enable our
method to supplement missing facts in KGs, and their powerful text
understanding abilities allow for better utilization of semantic information.
Specifically, CoLaKG extracts useful information from KGs at both local and
global levels. By employing the item-centered subgraph extraction and prompt
engineering, it can accurately understand the local information. In addition,
through the semantic-based retrieval module, each item is enriched by related
items from the entire knowledge graph, effectively harnessing global
information. Furthermore, the local and global information are effectively
integrated into the recommendation model through a representation fusion module
and a retrieval-augmented representation learning module, respectively.
Extensive experiments on four real-world datasets demonstrate the superiority
of our method.
| [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 04:44:34 GMT"
},
{
"version": "v2",
"created": "Fri, 7 Feb 2025 09:08:17 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 11:50:59 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Cui",
"Ziqiang",
""
],
[
"Weng",
"Yunpeng",
""
],
[
"Tang",
"Xing",
""
],
[
"Lyu",
"Fuyuan",
""
],
[
"Liu",
"Dugang",
""
],
[
"He",
"Xiuqiang",
""
],
[
"Ma",
"Chen",
""
]
] |
2410.12544 | Giulio Salizzoni | Giulio Salizzoni, Reda Ouhamma, Maryam Kamgarpour | Nash equilibria in scalar discrete-time linear quadratic games | Updated based on the reviews from ECC25. Camera ready version | null | null | null | cs.GT cs.MA | http://creativecommons.org/licenses/by-nc-sa/4.0/ | An open problem in linear quadratic (LQ) games has been characterizing the
Nash equilibria. This problem has renewed relevance given the surge of work on
understanding the convergence of learning algorithms in dynamic games. This
paper investigates scalar discrete-time infinite-horizon LQ games with two
agents. Even in this arguably simple setting, there are no results for finding
$\textit{all}$ Nash equilibria. By analyzing the best response map, we
formulate a polynomial system of equations characterizing the linear feedback
Nash equilibria. This enables us to bring in tools from algebraic geometry,
particularly the Gr\"obner basis, to study the roots of this polynomial system.
Consequently, we can not only compute all Nash equilibria numerically, but we
can also characterize their number with explicit conditions. For instance, we
prove that the LQ games under consideration admit at most three Nash
equilibria. We further provide sufficient conditions for the existence of at
most two Nash equilibria and sufficient conditions for the uniqueness of the
Nash equilibrium. Our numerical experiments demonstrate the tightness of our
bounds and showcase the increased complexity in settings with more than two
agents.
| [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 13:21:58 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 11:13:30 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Salizzoni",
"Giulio",
""
],
[
"Ouhamma",
"Reda",
""
],
[
"Kamgarpour",
"Maryam",
""
]
] |
2410.12876 | Zihao Zeng | Zihao Zeng, Bokai Lin, Tianqi Hou, Hao Zhang, Zhijie Deng | In-context KV-Cache Eviction for LLMs via Attention-Gate | null | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The KV-Cache technique has become the standard for the inference of large
language models (LLMs). Yet, it is widely criticized that KV-Cache can become a
bottleneck of the LLM inference system. This paper enables a novel dynamic
KV-Cache eviction policy by injecting a lightweight module called
Attention-Gate to the model. It accepts the global context as input and yields
eviction flags for each token. The self-attention modules in the model proceed
according to the flags and cache only a subset of the KV states for next token
prediction. The Attention-Gates can yield various flags for different heads and
layers and be easily tuned on top of a pre-trained LLM via continual
pre-training or supervised fine-tuning. The computational and memory overhead
introduced by Attention-Gates can be minimal. We empirically evaluate the
proposed approach across multiple scenarios, showing that effective eviction of
redundant tokens can not only improve efficiency but also enhance performance.
| [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 05:01:19 GMT"
},
{
"version": "v2",
"created": "Sat, 19 Oct 2024 08:45:11 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 03:51:06 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Zeng",
"Zihao",
""
],
[
"Lin",
"Bokai",
""
],
[
"Hou",
"Tianqi",
""
],
[
"Zhang",
"Hao",
""
],
[
"Deng",
"Zhijie",
""
]
] |
2410.13054 | Moritz Willig | Moritz Willig, Tim Nelson Tobiasch, Florian Peter Busch, Jonas Seng,
Devendra Singh Dhami, Kristian Kersting | Systems with Switching Causal Relations: A Meta-Causal Perspective | 21 pages, 3 figures, 4 tables, ICLR 2025 Camera Ready Version | null | null | null | cs.LG cs.AI stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most work on causality in machine learning assumes that causal relationships
are driven by a constant underlying process. However, the flexibility of
agents' actions or tipping points in the environmental process can change the
qualitative dynamics of the system. As a result, new causal relationships may
emerge, while existing ones change or disappear, resulting in an altered causal
graph. To analyze these qualitative changes on the causal graph, we propose the
concept of meta-causal states, which groups classical causal models into
clusters based on equivalent qualitative behavior and consolidates specific
mechanism parameterizations. We demonstrate how meta-causal states can be
inferred from observed agent behavior, and discuss potential methods for
disentangling these states from unlabeled data. Finally, we direct our analysis
towards the application of a dynamical system, showing that meta-causal states
can also emerge from inherent system dynamics, and thus constitute more than a
context-dependent framework in which mechanisms emerge only as a result of
external factors.
| [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 21:32:31 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 07:48:47 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Willig",
"Moritz",
""
],
[
"Tobiasch",
"Tim Nelson",
""
],
[
"Busch",
"Florian Peter",
""
],
[
"Seng",
"Jonas",
""
],
[
"Dhami",
"Devendra Singh",
""
],
[
"Kersting",
"Kristian",
""
]
] |
2410.14804 | Yongda Zhu | Yongda Zhu, Stacey Alberts, Jianwei Lyu, Jane Morrison, George H.
Rieke, Yang Sun, Jakob M. Helton, Zhiyuan Ji, Rachana Bhatawdekar, Nina
Bonaventura, Andrew J. Bunker, Xiaojing Lin, Marcia J. Rieke, Pierluigi
Rinaldi, Irene Shivaei, Christopher N. A. Willmer, Junyu Zhang | SMILES: Potentially Higher Ionizing Photon Production Efficiency in
Overdense Regions | Resubmitted after addressing the reviewer's comments | null | null | null | astro-ph.GA astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | The topology of reionization and the environments where galaxies efficiently
produce ionizing photons are key open questions. For the first time, we
investigate the trend between ionizing photon production efficiency, $\xi_{\rm
ion}$, and galaxy overdensity, $\log(1+\delta)$. We analyze the ionizing
properties of 79 galaxies between $1.0 < z < 5.2$ using JWST NIRSpec
medium-resolution spectra from the Systematic Mid-infrared Instrument (MIRI)
Legacy Extragalactic Survey (SMILES) program. Among these, 67 galaxies have
H$\alpha$ coverage, spanning $1.0 < z < 3.1$. The galaxy overdensity,
$\log(1+\delta)$, is measured using the JADES photometric catalog, which covers
the SMILES footprint. For the subset with H$\alpha$ coverage, we find that
$\log\xi_{\rm ion}$ is positively correlated with $\log(1+\delta)$, with a
slope of $0.94_{-0.46}^{+0.46}$. Additionally, the mean $\xi_{\rm ion}$ for
galaxies in overdense regions ($\log(1+\delta) > 0.1$) is 2.43 times that of
galaxies in lower density regions ($\log(1+\delta) < 0.1$). This strong trend
is found to be independent of redshift evolution. Furthermore, our results
confirm the robust correlations between $\xi_{\rm ion}$ and the rest-frame
equivalent widths of the [O III] or H$\alpha$ emission lines. Our results
suggest that galaxies in high-density regions are efficient producers of
ionizing photons.
| [
{
"version": "v1",
"created": "Fri, 18 Oct 2024 18:21:11 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 23:42:44 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Zhu",
"Yongda",
""
],
[
"Alberts",
"Stacey",
""
],
[
"Lyu",
"Jianwei",
""
],
[
"Morrison",
"Jane",
""
],
[
"Rieke",
"George H.",
""
],
[
"Sun",
"Yang",
""
],
[
"Helton",
"Jakob M.",
""
],
[
"Ji",
"Zhiyuan",
""
],
[
"Bhatawdekar",
"Rachana",
""
],
[
"Bonaventura",
"Nina",
""
],
[
"Bunker",
"Andrew J.",
""
],
[
"Lin",
"Xiaojing",
""
],
[
"Rieke",
"Marcia J.",
""
],
[
"Rinaldi",
"Pierluigi",
""
],
[
"Shivaei",
"Irene",
""
],
[
"Willmer",
"Christopher N. A.",
""
],
[
"Zhang",
"Junyu",
""
]
] |
2410.15787 | Hanseul Cho | Hanseul Cho and Jaeyoung Cha and Srinadh Bhojanapalli and Chulhee Yun | Arithmetic Transformers Can Length-Generalize in Both Operand Length and
Count | 44 pages, 20 figures, 26 tables, accepted to ICLR 2025 | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Transformers often struggle with length generalization, meaning they fail to
generalize to sequences longer than those encountered during training. While
arithmetic tasks are commonly used to study length generalization, certain
tasks are considered notoriously difficult, e.g., multi-operand addition
(requiring generalization over both the number of operands and their lengths)
and multiplication (requiring generalization over both operand lengths). In
this work, we achieve approximately 2-3x length generalization on both tasks,
which is the first such achievement in arithmetic Transformers. We design
task-specific scratchpads enabling the model to focus on a fixed number of
tokens per each next-token prediction step, and apply multi-level versions of
\Position Coupling (Cho et al., 2024; McLeish et al., 2024) to let Transformers
know the right position to attend to. On the theory side, we prove that a
1-layer Transformer using our method can solve multi-operand addition, up to
operand length and operand count that are exponential in embedding dimension.
| [
{
"version": "v1",
"created": "Mon, 21 Oct 2024 08:49:51 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 08:18:02 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Cho",
"Hanseul",
""
],
[
"Cha",
"Jaeyoung",
""
],
[
"Bhojanapalli",
"Srinadh",
""
],
[
"Yun",
"Chulhee",
""
]
] |
2410.16207 | Kumar Manas | Kumar Manas, Stefan Zwicklbauer and Adrian Paschke | CoT-TL: Low-Resource Temporal Knowledge Representation of Planning
Instructions Using Chain-of-Thought Reasoning | Accepted for publication in Proceedings of the 2024 IEEE/RSJ
International Conference on Intelligent Robots and Systems (IROS 2024), Abu
Dhabi 14-18 October 2024 | null | 10.1109/IROS58592.2024.10801817 | null | cs.RO cs.CL cs.FL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Autonomous agents often face the challenge of interpreting uncertain natural
language instructions for planning tasks. Representing these instructions as
Linear Temporal Logic (LTL) enables planners to synthesize actionable plans. We
introduce CoT-TL, a data-efficient in-context learning framework for
translating natural language specifications into LTL representations. CoT-TL
addresses the limitations of large language models, which typically rely on
extensive fine-tuning data, by extending chain-of-thought reasoning and
semantic roles to align with the requirements of formal logic creation. This
approach enhances the transparency and rationale behind LTL generation,
fostering user trust. CoT-TL achieves state-of-the-art accuracy across three
diverse datasets in low-data scenarios, outperforming existing methods without
fine-tuning or intermediate translations. To improve reliability and minimize
hallucinations, we incorporate model checking to validate the syntax of the
generated LTL output. We further demonstrate CoT-TL's effectiveness through
ablation studies and evaluations on unseen LTL structures and formulas in a new
dataset. Finally, we validate CoT-TL's practicality by integrating it into a
QuadCopter for multi-step drone planning based on natural language
instructions.
| [
{
"version": "v1",
"created": "Mon, 21 Oct 2024 17:10:43 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Manas",
"Kumar",
""
],
[
"Zwicklbauer",
"Stefan",
""
],
[
"Paschke",
"Adrian",
""
]
] |
2410.16226 | Polina Petriakova | Alexander Kamenshchik, Polina Petriakova | IR finite correlation functions in de Sitter space, a smooth massless
limit, and an autonomous equation | Appendix C is expanded and matches the version to appear in JHEP | JHEP 04 (2025) 127 | 10.1007/JHEP04(2025)127 | null | hep-th gr-qc | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We explore two-point and four-point correlation functions of a massive scalar
field on the flat de Sitter background in the long-wavelength approximation. By
employing the Yang-Feldman-type equation, we compute the two-point correlation
function up to the $\lambda^3$ order and the four-point correlation function up
to the $\lambda^2$ one. In contrast to the standard theory of a massive scalar
field based on the de Sitter-invariant vacuum, we develop the
vacuum-independent reasoning that may not possess de Sitter invariance but
results in a smooth massless limit of the correlation function's infrared part.
Our elaboration affords to calculate correlation functions of a free massive
scalar field and to proceed with quantum corrections, relying only on the known
two-point correlation function's infrared part of a free massless one.
Remarkably, the two-point correlation function of a free massive scalar field
coincides with the Ornstein-Uhlenbeck stochastic process's one and has a clear
physical interpretation. We compared our results with those obtained with the
Schwinger-Keldysh diagrammatic technique, Starobinsky's stochastic approach,
and the Hartree-Fock approximation. At last, we have constructed a
renormalization group-inspired autonomous equation for the two-point
correlation function. Integrating its approximate version, one obtains the
non-analytic expression with respect to a self-interaction coupling constant
$\lambda$. That solution reproduces the correct perturbative series up to the
two-loop level. At the late-time limit, it almost coincides with the result of
Starobinsky's stochastic approach in the whole interval of a new dimensionless
parameter $0 \leq \tfrac{\pi^2 m^4}{3\lambda H^4} < \infty$.
| [
{
"version": "v1",
"created": "Mon, 21 Oct 2024 17:30:15 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Oct 2024 21:40:09 GMT"
},
{
"version": "v3",
"created": "Fri, 4 Apr 2025 15:30:02 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Kamenshchik",
"Alexander",
""
],
[
"Petriakova",
"Polina",
""
]
] |
2410.16969 | Maxim Zabzine | Guido Festuccia, Roman Mauch and Maxim Zabzine | The Equivariant B model | 38 pages, refs added, the published version | null | null | null | hep-th | http://creativecommons.org/licenses/by/4.0/ | In this work, we introduce an equivariant deformation of the $B$ model on the
sphere with a $U(1)$-action. We present the deformed supersymmetry
transformations and corresponding Lagrangians and study observables in the
supercharge cohomology. The inclusion of equivariance allows for the
introduction of novel, position-dependent observables on the sphere, which have
no counterparts in the conventional $B$ model. Two specific cases we explore in
detail are position-dependent superpotentials and complex structure
deformations. In both instances, the theory exhibits notable differences from
the standard $B$ model, revealing intriguing new features.
| [
{
"version": "v1",
"created": "Tue, 22 Oct 2024 12:49:36 GMT"
},
{
"version": "v2",
"created": "Tue, 5 Nov 2024 15:31:33 GMT"
},
{
"version": "v3",
"created": "Wed, 16 Apr 2025 19:12:32 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Festuccia",
"Guido",
""
],
[
"Mauch",
"Roman",
""
],
[
"Zabzine",
"Maxim",
""
]
] |
2410.16975 | Fabian Kr\"uger | Fabian P. Kr\"uger, Johan \"Ostman, Lewis Mervin, Igor V. Tetko, Ola
Engkvist | Publishing Neural Networks in Drug Discovery Might Compromise Training
Data Privacy | null | null | 10.1186/s13321-025-00982-w | null | cs.CR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study investigates the risks of exposing confidential chemical
structures when machine learning models trained on these structures are made
publicly available. We use membership inference attacks, a common method to
assess privacy that is largely unexplored in the context of drug discovery, to
examine neural networks for molecular property prediction in a black-box
setting. Our results reveal significant privacy risks across all evaluated
datasets and neural network architectures. Combining multiple attacks increases
these risks. Molecules from minority classes, often the most valuable in drug
discovery, are particularly vulnerable. We also found that representing
molecules as graphs and using message-passing neural networks may mitigate
these risks. We provide a framework to assess privacy risks of classification
models and molecular representations. Our findings highlight the need for
careful consideration when sharing neural networks trained on proprietary
chemical structures, informing organisations and researchers about the
trade-offs between data confidentiality and model openness.
| [
{
"version": "v1",
"created": "Tue, 22 Oct 2024 12:55:02 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Krüger",
"Fabian P.",
""
],
[
"Östman",
"Johan",
""
],
[
"Mervin",
"Lewis",
""
],
[
"Tetko",
"Igor V.",
""
],
[
"Engkvist",
"Ola",
""
]
] |
2410.17139 | MD Abdullah Al Nasim | MD Abdullah Al Nasim, A.S.M Anas Ferdous, Abdur Rashid, Fatema Tuj
Johura Soshi, Parag Biswas, Angona Biswas, Kishor Datta Gupta | Trustworthy XAI and Application | null | null | null | null | cs.AI | http://creativecommons.org/publicdomain/zero/1.0/ | Artificial Intelligence (AI) is an important part of our everyday lives. We
use it in self-driving cars and smartphone assistants. People often call it a
"black box" because its complex systems, especially deep neural networks, are
hard to understand. This complexity raises concerns about accountability, bias,
and fairness, even though AI can be quite accurate. Explainable Artificial
Intelligence (XAI) is important for building trust. It helps ensure that AI
systems work reliably and ethically. This article looks at XAI and its three
main parts: transparency, explainability, and trustworthiness. We will discuss
why these components matter in real-life situations. We will also review recent
studies that show how XAI is used in different fields. Ultimately, gaining
trust in AI systems is crucial for their successful use in society.
| [
{
"version": "v1",
"created": "Tue, 22 Oct 2024 16:10:10 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 18:07:32 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Nasim",
"MD Abdullah Al",
""
],
[
"Ferdous",
"A. S. M Anas",
""
],
[
"Rashid",
"Abdur",
""
],
[
"Soshi",
"Fatema Tuj Johura",
""
],
[
"Biswas",
"Parag",
""
],
[
"Biswas",
"Angona",
""
],
[
"Gupta",
"Kishor Datta",
""
]
] |
2410.17188 | Samarth Kalluraya | Samarth Kalluraya, Beichen Zhou, Yiannis Kantaros | Minimum-Violation Temporal Logic Planning for Heterogeneous Robots under
Robot Skill Failures | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | In this paper, we consider teams of robots with heterogeneous skills (e.g.,
sensing and manipulation) tasked with collaborative missions described by
Linear Temporal Logic (LTL) formulas. These LTL-encoded tasks require robots to
apply their skills to specific regions and objects in a temporal and logical
order. While existing temporal logic planning algorithms can synthesize
correct-by-construction plans, they typically lack reactivity to unexpected
failures of robot skills, which can compromise mission performance. This paper
addresses this challenge by proposing a reactive LTL planning algorithm that
adapts to unexpected failures during deployment. Specifically, the proposed
algorithm reassigns sub-tasks to robots based on their functioning skills and
locally revises team plans to accommodate these new assignments and ensure
mission completion. The main novelty of the proposed algorithm is its ability
to handle cases where mission completion becomes impossible due to limited
functioning robots. Instead of reporting mission failure, the algorithm
strategically prioritizes the most crucial sub-tasks and locally revises the
team's plans, as per user-specified priorities, to minimize mission violations.
We provide theoretical conditions under which the proposed framework computes
the minimum-violation task reassignments and team plans. We provide numerical
and hardware experiments to demonstrate the efficiency of the proposed method.
| [
{
"version": "v1",
"created": "Tue, 22 Oct 2024 17:09:28 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 14:40:34 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Kalluraya",
"Samarth",
""
],
[
"Zhou",
"Beichen",
""
],
[
"Kantaros",
"Yiannis",
""
]
] |
2410.17385 | Ziqiao Ma | Zheyuan Zhang, Fengyuan Hu, Jayjun Lee, Freda Shi, Parisa
Kordjamshidi, Joyce Chai, Ziqiao Ma | Do Vision-Language Models Represent Space and How? Evaluating Spatial
Frame of Reference Under Ambiguities | Accepted to ICLR 2025 (Oral) | Project page:
https://spatial-comfort.github.io/ | null | null | null | cs.CL cs.CV | http://creativecommons.org/licenses/by/4.0/ | Spatial expressions in situated communication can be ambiguous, as their
meanings vary depending on the frames of reference (FoR) adopted by speakers
and listeners. While spatial language understanding and reasoning by
vision-language models (VLMs) have gained increasing attention, potential
ambiguities in these models are still under-explored. To address this issue, we
present the COnsistent Multilingual Frame Of Reference Test (COMFORT), an
evaluation protocol to systematically assess the spatial reasoning capabilities
of VLMs. We evaluate nine state-of-the-art VLMs using COMFORT. Despite showing
some alignment with English conventions in resolving ambiguities, our
experiments reveal significant shortcomings of VLMs: notably, the models (1)
exhibit poor robustness and consistency, (2) lack the flexibility to
accommodate multiple FoRs, and (3) fail to adhere to language-specific or
culture-specific conventions in cross-lingual tests, as English tends to
dominate other languages. With a growing effort to align vision-language models
with human cognitive intuitions, we call for more attention to the ambiguous
nature and cross-cultural diversity of spatial reasoning.
| [
{
"version": "v1",
"created": "Tue, 22 Oct 2024 19:39:15 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 17:59:27 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Zhang",
"Zheyuan",
""
],
[
"Hu",
"Fengyuan",
""
],
[
"Lee",
"Jayjun",
""
],
[
"Shi",
"Freda",
""
],
[
"Kordjamshidi",
"Parisa",
""
],
[
"Chai",
"Joyce",
""
],
[
"Ma",
"Ziqiao",
""
]
] |
2410.18847 | Ji-Chong Yang Mr | Ji-Chong Yang, Shuai Zhang, Chong-Xing Yue | A novel quantum machine learning classifier to search for new physics | 46 pages, 14 figures | null | null | null | hep-ph quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Due to the success of the Standard Model~(SM), it is reasonable to anticipate
that the signal of new physics~(NP) beyond the SM is small. Consequently,
future searches for NP and precision tests of the SM will require high
luminosity collider experiments. Moreover, as precision tests advance, rare
processes with many final-state particles require consideration which demand
the analysis of a vast number of observables. The high luminosity produces a
large amount of experimental data spanning a large observable space, posing a
significant data-processing challenge. In recent years, quantum machine
learning has emerged as a promising approach for processing large amounts of
complex data on a quantum computer. In this study, we propose quantum searching
neighbor~(QSN) and variational QSN~(VQSN) algorithms to search for NP. The QSN
is a classification algorithm. The VQSN introduces variation to the QSN to
process classical data. As applications, we apply the (V)QSN in the
phenomenological study of the NP at the Large Hadron Collider and muon
colliders. The results indicate that the VQSN demonstrates superior efficiency
to a classical counterpart k-nearest neighbor algorithm, even when dealing with
classical data.
| [
{
"version": "v1",
"created": "Thu, 24 Oct 2024 15:27:28 GMT"
},
{
"version": "v2",
"created": "Fri, 31 Jan 2025 09:56:54 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 01:01:10 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Yang",
"Ji-Chong",
""
],
[
"Zhang",
"Shuai",
""
],
[
"Yue",
"Chong-Xing",
""
]
] |
2410.19101 | Silvio Paolo Sorella | M. S. Guimaraes, I. Roditi, S. P. Sorella | Bell's inequality in relativistic Quantum Field Theory | 30 pages, two figures, new material added in Sect.V | null | null | null | quant-ph hep-th math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A concise and self-contained introduction to the Bell inequality in
relativistic Quantum Field Theory is presented. Taking the example of a real
scalar massive field, the violation of the Bell inequality in the vacuum state
and for causal complementary wedges is illustrated.
| [
{
"version": "v1",
"created": "Thu, 24 Oct 2024 19:05:05 GMT"
},
{
"version": "v2",
"created": "Wed, 11 Dec 2024 10:35:06 GMT"
},
{
"version": "v3",
"created": "Wed, 16 Apr 2025 18:37:33 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Guimaraes",
"M. S.",
""
],
[
"Roditi",
"I.",
""
],
[
"Sorella",
"S. P.",
""
]
] |
2410.19903 | Pascal Ba{\ss}ler | Pascal Ba{\ss}ler, Markus Heinrich, Martin Kliesch | General, efficient, and robust Hamiltonian engineering | 29 pages, 8 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Implementing the time evolution under a desired target Hamiltonian is
critical for various applications in quantum science. Due to the exponential
increase of parameters in the system size and due to experimental imperfections
this task can be challenging in quantum many-body settings. We introduce an
efficient and robust scheme to engineer arbitrary local many-body Hamiltonians.
This is achieved by applying simple single-qubit gates simultaneously to an
always-on system Hamiltonian, which we assume to be native to a given platform.
These sequences are constructed by efficiently solving a linear program (LP)
which minimizes the total evolution time. In this way, we can engineer target
Hamiltonians that are only limited by the locality of the Pauli terms in the
system Hamiltonian. Based on average Hamiltonian theory and by using robust
composite pulses, we make our schemes robust against errors including finite
pulse time errors and various calibration errors. To demonstrate the
performance of our scheme, we provide numerical simulations. In particular, we
solve the Hamiltonian engineering problem for arbitrary two-body Hamiltonians
on a 2D square lattice with $225$ qubits in only $60$ seconds. Moreover, we
simulate the time evolution of Heisenberg Hamiltonians for smaller system sizes
with a fidelity larger than $99.9\%$, which is orders of magnitude better than
non-robust implementations.
| [
{
"version": "v1",
"created": "Fri, 25 Oct 2024 18:00:01 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 07:29:46 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Baßler",
"Pascal",
""
],
[
"Heinrich",
"Markus",
""
],
[
"Kliesch",
"Martin",
""
]
] |
2410.20056 | Millicent Li | Millicent Li, Tongfei Chen, Benjamin Van Durme, Patrick Xia | Multi-Field Adaptive Retrieval | ICLR 2025, Spotlight | null | null | null | cs.IR cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Document retrieval for tasks such as search and retrieval-augmented
generation typically involves datasets that are unstructured: free-form text
without explicit internal structure in each document. However, documents can
have a structured form, consisting of fields such as an article title, message
body, or HTML header. To address this gap, we introduce Multi-Field Adaptive
Retrieval (MFAR), a flexible framework that accommodates any number of and any
type of document indices on structured data. Our framework consists of two main
steps: (1) the decomposition of an existing document into fields, each indexed
independently through dense and lexical methods, and (2) learning a model which
adaptively predicts the importance of a field by conditioning on the document
query, allowing on-the-fly weighting of the most likely field(s). We find that
our approach allows for the optimized use of dense versus lexical
representations across field types, significantly improves in document ranking
over a number of existing retrievers, and achieves state-of-the-art performance
for multi-field structured data.
| [
{
"version": "v1",
"created": "Sat, 26 Oct 2024 03:07:22 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 18:17:24 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Li",
"Millicent",
""
],
[
"Chen",
"Tongfei",
""
],
[
"Van Durme",
"Benjamin",
""
],
[
"Xia",
"Patrick",
""
]
] |
2410.21032 | Mario Kieburg Dr. habil. | Gernot Akemann, Noah Ayg\"un, Mario Kieburg, Patricia P\"a{\ss}ler | Complex symmetric, self-dual, and Ginibre random matrices: Analytical
results for three classes of bulk and edge statistics | 47 pages, 2 figures, v2: typos corrected and minor clarifications
added | J. Phys. A: Math. Theor. 58 (2025) 125204 | 10.1088/1751-8121/adbd9d | null | math-ph cond-mat.stat-mech math.MP math.PR | http://creativecommons.org/licenses/by/4.0/ | Recently, a conjecture about the local bulk statistics of complex eigenvalues
has been made based on numerics. It claims that there are only three
universality classes, which have all been observed in open chaotic quantum
systems. Motivated by these new insights, we compute and compare the
expectation values of $k$ pairs of complex conjugate characteristic polynomials
in three ensembles of Gaussian non-Hermitian random matrices representative for
the three classes: the well-known complex Ginibre ensemble, complex symmetric
and complex self-dual matrices. In the Cartan classification scheme of
non-Hermitian random matrices they are labelled as class A, AI$^\dag$ and
AII$^\dag$, respectively. Using the technique of Grassmann variables, we derive
explicit expressions for a single pair of expected characteristic polynomials
for finite as well as infinite matrix dimension. For the latter we consider the
global limit as well as zoom into the edge and the bulk of the spectrum,
providing new analytical results for classes AI$^\dag$ and AII$^\dag$. For
general $k$, we derive the effective Lagrangians corresponding to the
non-linear $\sigma$-models in the respective physical systems. Interestingly,
they agree for all three ensembles, while the corresponding Goldstone
manifolds, over which one has to perform the remaining integrations, are
different and equal the three classical compact groups in the bulk. In
particular, our analytical results show that these three ensembles have indeed
different local bulk and edge spectral statistics, corroborating the conjecture
further.
| [
{
"version": "v1",
"created": "Mon, 28 Oct 2024 13:53:40 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 07:41:22 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Akemann",
"Gernot",
""
],
[
"Aygün",
"Noah",
""
],
[
"Kieburg",
"Mario",
""
],
[
"Päßler",
"Patricia",
""
]
] |
2410.21179 | Kaijing Luo | Kaijing Luo and Ka-Ho Chow | Harmless Backdoor-based Client-side Watermarking in Federated Learning | Accepted to EuroSP 2025 | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Protecting intellectual property (IP) in federated learning (FL) is
increasingly important as clients contribute proprietary data to
collaboratively train models. Model watermarking, particularly through
backdoor-based methods, has emerged as a popular approach for verifying
ownership and contributions in deep neural networks trained via FL. By
manipulating their datasets, clients can embed a secret pattern, resulting in
non-intuitive predictions that serve as proof of participation, useful for
claiming incentives or IP co-ownership. However, this technique faces practical
challenges: (i) client watermarks can collide, leading to ambiguous ownership
claims, and (ii) malicious clients may exploit watermarks to manipulate model
predictions for harmful purposes. To address these issues, we propose
Sanitizer, a server-side method that ensures client-embedded backdoors can only
be activated in harmless environments but not natural queries. It identifies
subnets within client-submitted models, extracts backdoors throughout the FL
process, and confines them to harmless, client-specific input subspaces. This
approach not only enhances Sanitizer's efficiency but also resolves conflicts
when clients use similar triggers with different target labels. Our empirical
results demonstrate that Sanitizer achieves near-perfect success verifying
client contributions while mitigating the risks of malicious watermark use.
Additionally, it reduces GPU memory consumption by 85% and cuts processing time
by at least 5x compared to the baseline. Our code is open-sourced at
https://hku-tasr.github.io/Sanitizer/.
| [
{
"version": "v1",
"created": "Mon, 28 Oct 2024 16:20:01 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 14:59:30 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Luo",
"Kaijing",
""
],
[
"Chow",
"Ka-Ho",
""
]
] |
2410.21817 | Stefano Di Giovacchino | Raffaele D'Ambrosio, Stefano Di Giovacchino | Backward error analysis of stochastic Poisson integrators | null | null | null | null | math.NA cs.NA | http://creativecommons.org/licenses/by/4.0/ | We address our attention to the numerical time discretization of stochastic
Poisson systems via Poisson integrators. The aim of the investigation regards
the backward error analysis of such integrators to reveal their ability of
being structure-preserving, for long times of integration. In particular, we
first provide stochastic modified equations suitable for such integrators and
then we rigorously study them to prove accurate estimates on the long-term
numerical error along the dynamics generated by stochastic Poisson integrators,
with reference to the preservation of the random Hamiltonian conserved along
the exact flow of the approximating Wong-Zakai Poisson system. Finally,
selected numerical experiments confirm the effectiveness of the theoretical
analysis.
| [
{
"version": "v1",
"created": "Tue, 29 Oct 2024 07:39:23 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 13:48:57 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"D'Ambrosio",
"Raffaele",
""
],
[
"Di Giovacchino",
"Stefano",
""
]
] |
2410.22572 | Mohamed Elsayed | Mohamed M. Elsayed, Bruno Uchoa, and Valeri N. Kotov | Coulomb interactions in systems of generalized semi-Dirac fermions | 11 pages, 6 figures | Phys. Rev. B 111, 165127 (2025) | 10.1103/PhysRevB.111.165127 | null | cond-mat.str-el cond-mat.mes-hall cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Interactions have strong effects in systems with flat bands. We examine the
role of Coulomb interactions in two dimensional chiral anisotropic
quasiparticles that disperse linearly in one direction and have relatively flat
bands near the neutrality point in the other direction, dispersing with an
arbitrary positive even power law $2n\geq 2$. As in the conventional semi-Dirac
case ($n = 1$), we show using renormalization group that strong logarithmic
divergences in the self-energy of generalized semi-Dirac fermions resum and
lead to a restoration of the linearity of the spectrum for arbitrary n over a
sizable energy window in the perturbative regime. We discuss those results in
light of previous non-perturbative large $N_f$ results and address the
implications for physical observables.
| [
{
"version": "v1",
"created": "Tue, 29 Oct 2024 22:25:56 GMT"
},
{
"version": "v2",
"created": "Sat, 9 Nov 2024 00:20:52 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Elsayed",
"Mohamed M.",
""
],
[
"Uchoa",
"Bruno",
""
],
[
"Kotov",
"Valeri N.",
""
]
] |
2410.23727 | Yaowei Xie | Yaowei Xie, Huan Yu | Mild ill-posedness in $W^{1,\infty}$ for the incompressible porous media
equation | 31 pages | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we establish the mild ill-posedness of 2D IPM equation in the
critical Sobolev space $W^{1,\infty}$ when the initial data are small
perturbations of stable profile $g(x_2).$ Consequently, instability can be
inferred. Notably, our results are valid for arbitrary vertically stratified
density profiles $g(x_2)$ without imposing any restrictions on the sign of
$g'(x_2).$ From a physical perspective, since gravity acts downward, density
profiles satisfying $g'(x_2) < 0$ typically correspond to stable
configurations, whereas those with $g '(x_2) > 0$ are generally expected to be
unstable. Surprisingly, our analysis uncovers an unexpected instability even
when $g'(x_2) < 0$ and $g'(x_2)\in W^{2,\infty}(\mathbb{R})$. To the best of
our knowledge, this work provides the first rigorous demonstration of IPM
instability for vertically nonlinear density profiles, marking a significant
departure from conventional physical expectations.
| [
{
"version": "v1",
"created": "Thu, 31 Oct 2024 08:26:54 GMT"
},
{
"version": "v2",
"created": "Sun, 3 Nov 2024 07:07:35 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 06:22:01 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Xie",
"Yaowei",
""
],
[
"Yu",
"Huan",
""
]
] |
2410.24134 | David Alonso | David Alonso, Oleksandr Hetmantsev, Giulio Fabbian, Anze Slosar, Kate
Storey-Fisher | Measurement of the power spectrum turnover scale from the
cross-correlation between CMB lensing and Quaia | 18 pages, 14 figures. Accepted in the Open Journal of Astrophysics | null | 10.33232/001c.136891 | null | astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | We use the projected clustering of quasars in the Gaia-unWISE quasar catalog,
Quaia, and its cross-correlation with CMB lensing data from Planck, to measure
the large-scale turnover of the matter power spectrum, associated with the size
of the horizon at the epoch of matter-radiation equality. The turnover is
detected with a significance of between $2.3$ and $3.1\sigma$, depending on the
method used to quantify it. From this measurement, the equality scale is
determined at the $\sim20\%$ level. Using the turnover scale as a standard
ruler alone (suppressing information from the large-scale curvature of the
power spectrum), in combination with supernova data through an inverse distance
ladder approach, we measure the current expansion rate to be
$H_0=62.7\pm17.2\,{\rm km}\,{\rm s}^{-1}\,{\rm Mpc}^{-1}$. The addition of
information coming from the power spectrum curvature approximately halves the
standard ruler uncertainty. Our measurement in combination with calibrated
supernovae from Pantheon$+$ and SH0ES constrains the CMB temperature to be
$T_{\rm CMB}=3.10^{+0.48}_{-0.36}\,{\rm K}$, independently of CMB data.
Alternatively, assuming the value of $T_{\rm CMB}$ from COBE-FIRAS, we can
constrain the effective number of relativistic species in the early Universe to
be $N_{\rm eff}=3.0^{+5.8}_{-2.9}$.
| [
{
"version": "v1",
"created": "Thu, 31 Oct 2024 16:57:36 GMT"
},
{
"version": "v2",
"created": "Wed, 6 Nov 2024 19:06:27 GMT"
},
{
"version": "v3",
"created": "Wed, 16 Apr 2025 14:20:59 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Alonso",
"David",
""
],
[
"Hetmantsev",
"Oleksandr",
""
],
[
"Fabbian",
"Giulio",
""
],
[
"Slosar",
"Anze",
""
],
[
"Storey-Fisher",
"Kate",
""
]
] |
2411.00238 | Taylor Webb | Declan Campbell, Sunayana Rane, Tyler Giallanza, Nicol\`o De Sabbata,
Kia Ghods, Amogh Joshi, Alexander Ku, Steven M. Frankland, Thomas L.
Griffiths, Jonathan D. Cohen, Taylor W. Webb | Understanding the Limits of Vision Language Models Through the Lens of
the Binding Problem | null | null | null | null | cs.AI cs.CV cs.LG q-bio.NC | http://creativecommons.org/licenses/by/4.0/ | Recent work has documented striking heterogeneity in the performance of
state-of-the-art vision language models (VLMs), including both multimodal
language models and text-to-image models. These models are able to describe and
generate a diverse array of complex, naturalistic images, yet they exhibit
surprising failures on basic multi-object reasoning tasks -- such as counting,
localization, and simple forms of visual analogy -- that humans perform with
near perfect accuracy. To better understand this puzzling pattern of successes
and failures, we turn to theoretical accounts of the binding problem in
cognitive science and neuroscience, a fundamental problem that arises when a
shared set of representational resources must be used to represent distinct
entities (e.g., to represent multiple objects in an image), necessitating the
use of serial processing to avoid interference. We find that many of the
puzzling failures of state-of-the-art VLMs can be explained as arising due to
the binding problem, and that these failure modes are strikingly similar to the
limitations exhibited by rapid, feedforward processing in the human brain.
| [
{
"version": "v1",
"created": "Thu, 31 Oct 2024 22:24:47 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 21:59:00 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Campbell",
"Declan",
""
],
[
"Rane",
"Sunayana",
""
],
[
"Giallanza",
"Tyler",
""
],
[
"De Sabbata",
"Nicolò",
""
],
[
"Ghods",
"Kia",
""
],
[
"Joshi",
"Amogh",
""
],
[
"Ku",
"Alexander",
""
],
[
"Frankland",
"Steven M.",
""
],
[
"Griffiths",
"Thomas L.",
""
],
[
"Cohen",
"Jonathan D.",
""
],
[
"Webb",
"Taylor W.",
""
]
] |
2411.00826 | Chun Li | Yan Zhang, Ming Li, Chun Li, Zhaoxia Liu, Ye Zhang, and Fei Richard Yu | Uncertainty Quantification via H\"older Divergence for Multi-View
Representation Learning | NA | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Evidence-based deep learning represents a burgeoning paradigm for uncertainty
estimation, offering reliable predictions with negligible extra computational
overheads. Existing methods usually adopt Kullback-Leibler divergence to
estimate the uncertainty of network predictions, ignoring domain gaps among
various modalities. To tackle this issue, this paper introduces a novel
algorithm based on H\"older Divergence (HD) to enhance the reliability of
multi-view learning by addressing inherent uncertainty challenges from
incomplete or noisy data. Generally, our method extracts the representations of
multiple modalities through parallel network branches, and then employs HD to
estimate the prediction uncertainties. Through the Dempster-Shafer theory,
integration of uncertainty from different modalities, thereby generating a
comprehensive result that considers all available representations.
Mathematically, HD proves to better measure the ``distance'' between real data
distribution and predictive distribution of the model and improve the
performances of multi-class recognition tasks.
Specifically, our method surpass the existing state-of-the-art counterparts
on all evaluating benchmarks.
We further conduct extensive experiments on different backbones to verify our
superior robustness. It is demonstrated that our method successfully pushes the
corresponding performance boundaries. Finally, we perform experiments on more
challenging scenarios, \textit{i.e.}, learning with incomplete or noisy data,
revealing that our method exhibits a high tolerance to such corrupted data.
| [
{
"version": "v1",
"created": "Tue, 29 Oct 2024 04:29:44 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 13:24:31 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Zhang",
"Yan",
""
],
[
"Li",
"Ming",
""
],
[
"Li",
"Chun",
""
],
[
"Liu",
"Zhaoxia",
""
],
[
"Zhang",
"Ye",
""
],
[
"Yu",
"Fei Richard",
""
]
] |
2411.01359 | Mattia Zanella | Giuseppe Toscani, Mattia Zanella | Supercritical Fokker-Planck equations for consensus dynamics: large-time
behaviour and weighted Nash-type inequalities | null | null | null | null | math.AP math-ph math.MP nlin.AO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the main properties of the solution of a Fokker-Planck equation
characterized by a variable diffusion coefficient and a polynomial superlinear
drift, modeling the formation of consensus in a large interacting system of
individuals. The Fokker-Planck equation is derived from the kinetic description
of the dynamics of a quantum particle system, and in presence of a high
nonlinearity in the drift operator, mimicking the effects of the mass in the
alignment forces, allows for steady states similar to a Bose-Einstein
condensate. The main feature of this Fokker-Planck equation is the presence of
a variable diffusion coefficient, a nonlinear drift and boundaries, which
introduce new challenging mathematical problems in the study of its long-time
behavior. In particular, propagation of regularity is shown as a consequence of
new weighted Nash and Gagliardo-Nirenberg inequalities.
| [
{
"version": "v1",
"created": "Sat, 2 Nov 2024 20:32:52 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Nov 2024 13:38:19 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 11:00:22 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Toscani",
"Giuseppe",
""
],
[
"Zanella",
"Mattia",
""
]
] |
2411.01639 | Neel P. Bhatt | Neel P. Bhatt, Yunhao Yang, Rohan Siva, Daniel Milan, Ufuk Topcu,
Zhangyang Wang | Know Where You're Uncertain When Planning with Multimodal Foundation
Models: A Formal Framework | Fine-tuned models, code, and datasets are available at
https://uncertainty-in-planning.github.io/ | null | null | null | cs.RO cs.AI cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Multimodal foundation models offer a promising framework for robotic
perception and planning by processing sensory inputs to generate actionable
plans. However, addressing uncertainty in both perception (sensory
interpretation) and decision-making (plan generation) remains a critical
challenge for ensuring task reliability. We present a comprehensive framework
to disentangle, quantify, and mitigate these two forms of uncertainty. We first
introduce a framework for uncertainty disentanglement, isolating perception
uncertainty arising from limitations in visual understanding and decision
uncertainty relating to the robustness of generated plans.
To quantify each type of uncertainty, we propose methods tailored to the
unique properties of perception and decision-making: we use conformal
prediction to calibrate perception uncertainty and introduce
Formal-Methods-Driven Prediction (FMDP) to quantify decision uncertainty,
leveraging formal verification techniques for theoretical guarantees. Building
on this quantification, we implement two targeted intervention mechanisms: an
active sensing process that dynamically re-observes high-uncertainty scenes to
enhance visual input quality and an automated refinement procedure that
fine-tunes the model on high-certainty data, improving its capability to meet
task specifications. Empirical validation in real-world and simulated robotic
tasks demonstrates that our uncertainty disentanglement framework reduces
variability by up to 40% and enhances task success rates by 5% compared to
baselines. These improvements are attributed to the combined effect of both
interventions and highlight the importance of uncertainty disentanglement,
which facilitates targeted interventions that enhance the robustness and
reliability of autonomous systems. Fine-tuned models, code, and datasets are
available at https://uncertainty-in-planning.github.io/.
| [
{
"version": "v1",
"created": "Sun, 3 Nov 2024 17:32:00 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Apr 2025 22:37:07 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 02:45:08 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Bhatt",
"Neel P.",
""
],
[
"Yang",
"Yunhao",
""
],
[
"Siva",
"Rohan",
""
],
[
"Milan",
"Daniel",
""
],
[
"Topcu",
"Ufuk",
""
],
[
"Wang",
"Zhangyang",
""
]
] |
2411.02625 | Deok-Hyeon Cho | Deok-Hyeon Cho, Hyung-Seok Oh, Seung-Bin Kim, Seong-Whan Lee | EmoSphere++: Emotion-Controllable Zero-Shot Text-to-Speech via
Emotion-Adaptive Spherical Vector | null | Published in IEEE Transactions on Affective Computing 2025 | 10.1109/TAFFC.2025.3561267 | null | cs.SD cs.AI eess.AS | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Emotional text-to-speech (TTS) technology has achieved significant progress
in recent years; however, challenges remain owing to the inherent complexity of
emotions and limitations of the available emotional speech datasets and models.
Previous studies typically relied on limited emotional speech datasets or
required extensive manual annotations, restricting their ability to generalize
across different speakers and emotional styles. In this paper, we present
EmoSphere++, an emotion-controllable zero-shot TTS model that can control
emotional style and intensity to resemble natural human speech. We introduce a
novel emotion-adaptive spherical vector that models emotional style and
intensity without human annotation. Moreover, we propose a multi-level style
encoder that can ensure effective generalization for both seen and unseen
speakers. We also introduce additional loss functions to enhance the emotion
transfer performance for zero-shot scenarios. We employ a conditional flow
matching-based decoder to achieve high-quality and expressive emotional TTS in
a few sampling steps. Experimental results demonstrate the effectiveness of the
proposed framework.
| [
{
"version": "v1",
"created": "Mon, 4 Nov 2024 21:33:56 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 01:50:24 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Cho",
"Deok-Hyeon",
""
],
[
"Oh",
"Hyung-Seok",
""
],
[
"Kim",
"Seung-Bin",
""
],
[
"Lee",
"Seong-Whan",
""
]
] |
2411.02729 | Zijian Zhang | Zijian Zhang, Linhua Jiang, Weiyang Liu, and Luis C. Ho | Analysis of Multi-epoch JWST Images of $\sim 300$ Little Red Dots:
Tentative Detection of Variability in a Minority of Sources | 27 pages, 19 figures, 7 tables. Accepted for publication in ApJ. The
light curves and cut-out images of all LRDs can be found at
https://github.com/Zijian-astro/variable_LRD_candidates | null | null | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | James Webb Space Telescope (JWST) has revealed a population of red and
compact sources at $z \gtrsim 5$ known as ``Little Red Dots'' (LRDs) that are
likely active galactic nuclei (AGNs). Here we present a comprehensive study of
the variability of 314 LRDs with multi-epoch JWST observations in five deep
fields: UDS, GOODS-S, GOODS-N, Abell 2744, and COSMOS. Our analyses use all
publicly available JWST NIRCam imaging data in these fields, together with
multi-epoch JWST MIRI images available. We measure the significance
(signal-to-noise ratio or ${\rm SNR}_{\rm var}$) of the variabilities for all
LRDs and statistically evaluate their variabilities using the ${\rm SNR}_{\rm
var}$ distributions. We pay particular attention to the systematic offsets of
photometric zero points among different epochs that seem to commonly exist. The
derived ${\rm SNR}_{\rm var}$ distributions of the LRDs, including those with
broad H$\alpha$/H$\beta$ emission lines, follow the standard Gaussian
distribution, and are generally consistent with those of the comparison samples
of objects detected in the same images. This finding suggests that the LRD
population on average does not show strong variability, which can be explained
by super-Eddington accretion of the black holes in AGNs. Alternatively, many of
them may be dominated by galaxies. We also find eight strongly variable LRD
candidates with variability amplitudes of 0.24 -- 0.82 mag. The rest-frame
optical SEDs of these variable LRDs should have significant AGN contribution.
Future JWST observations will provide more variability information of LRDs.
| [
{
"version": "v1",
"created": "Tue, 5 Nov 2024 01:55:48 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 07:14:14 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Zhang",
"Zijian",
""
],
[
"Jiang",
"Linhua",
""
],
[
"Liu",
"Weiyang",
""
],
[
"Ho",
"Luis C.",
""
]
] |
2411.02968 | Manish Chaudhary | Manish Chaudhary, Zhiyuan Lin, Shuang Li, Mohan Zhang, Yuping Mao,
Valentin Ivannikov, and Tim Byrnes | Macroscopic quantum teleportation with ensembles of qubits | 18 pages, 10 figures | PRA 111, 042421 (2025) | 10.1103/PhysRevA.111.042421 | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | We develop methods for performing quantum teleportation of the total spin
variables of an unknown state, using quantum nondemolition measurements, spin
projection measurements, and classical communication. While theoretically
teleportation of high-dimensional states can be attained with the assumption of
generalized Bell measurements, this is typically experimentally non-trivial to
implement. We introduce two protocols and show that, on average, the
teleportation succeeds in teleporting the spin variables of a spin coherent
state with average zero angular error in the ideal case, beating classical
strategies based on quantum state estimation. In a single run of the
teleportation, there is an angular error at the level of ~ 0.1 radians for
large ensembles. A potential physical implementation for the scheme is with
atomic ensembles and quantum nondemolition measurements performed with light.
We analyze the decoherence of the protocols and find that the protocol is
robust even in the limit of large ensemble sizes.
| [
{
"version": "v1",
"created": "Tue, 5 Nov 2024 10:12:25 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 08:50:05 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Chaudhary",
"Manish",
""
],
[
"Lin",
"Zhiyuan",
""
],
[
"Li",
"Shuang",
""
],
[
"Zhang",
"Mohan",
""
],
[
"Mao",
"Yuping",
""
],
[
"Ivannikov",
"Valentin",
""
],
[
"Byrnes",
"Tim",
""
]
] |
2411.03728 | Pengfei Lyu | Pengfei Lyu, Pak-Hei Yeung, Xiaosheng Yu, Xiufei Cheng, Chengdong Wu,
Jagath C. Rajapakse | Efficient Fourier Filtering Network with Contrastive Learning for
UAV-based Unaligned Bi-modal Salient Object Detection | Accepted by TGRS 2025 | null | 10.1109/TGRS.2025.3562562 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unmanned aerial vehicle (UAV)-based bi-modal salient object detection (BSOD)
aims to segment salient objects in a scene utilizing complementary cues in
unaligned RGB and thermal image pairs. However, the high computational expense
of existing UAV-based BSOD models limits their applicability to real-world UAV
devices. To address this problem, we propose an efficient Fourier filter
network with contrastive learning that achieves both real-time and accurate
performance. Specifically, we first design a semantic contrastive alignment
loss to align the two modalities at the semantic level, which facilitates
mutual refinement in a parameter-free way. Second, inspired by the fast Fourier
transform that obtains global relevance in linear complexity, we propose
synchronized alignment fusion, which aligns and fuses bi-modal features in the
channel and spatial dimensions by a hierarchical filtering mechanism. Our
proposed model, AlignSal, reduces the number of parameters by 70.0%, decreases
the floating point operations by 49.4%, and increases the inference speed by
152.5% compared to the cutting-edge BSOD model (i.e., MROS). Extensive
experiments on the UAV RGB-T 2400 and seven bi-modal dense prediction datasets
demonstrate that AlignSal achieves both real-time inference speed and better
performance and generalizability compared to nineteen state-of-the-art models
across most evaluation metrics. In addition, our ablation studies further
verify AlignSal's potential in boosting the performance of existing aligned
BSOD models on UAV-based unaligned data. The code is available at:
https://github.com/JoshuaLPF/AlignSal.
| [
{
"version": "v1",
"created": "Wed, 6 Nov 2024 07:46:34 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 07:11:34 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Lyu",
"Pengfei",
""
],
[
"Yeung",
"Pak-Hei",
""
],
[
"Yu",
"Xiaosheng",
""
],
[
"Cheng",
"Xiufei",
""
],
[
"Wu",
"Chengdong",
""
],
[
"Rajapakse",
"Jagath C.",
""
]
] |
2411.04011 | Fabio Pavirani | Fabio Pavirani, Jonas Van Gompel, Seyed Soroush Karimi Madahi, Bert
Claessens, Chris Develder | Predicting and Publishing Accurate Imbalance Prices Using Monte Carlo
Tree Search | null | null | null | null | eess.SY cs.AI cs.LG cs.SY | http://creativecommons.org/licenses/by/4.0/ | The growing reliance on renewable energy sources, particularly solar and
wind, has introduced challenges due to their uncontrollable production. This
complicates maintaining the electrical grid balance, prompting some
transmission system operators in Western Europe to implement imbalance tariffs
that penalize unsustainable power deviations. These tariffs create an implicit
demand response framework to mitigate grid instability. Yet, several challenges
limit active participation. In Belgium, for example, imbalance prices are only
calculated at the end of each 15-minute settlement period, creating high risk
due to price uncertainty. This risk is further amplified by the inherent
volatility of imbalance prices, discouraging participation. Although
transmission system operators provide minute-based price predictions, the
system imbalance volatility makes accurate price predictions challenging to
obtain and requires sophisticated techniques. Moreover, publishing price
estimates can prompt participants to adjust their schedules, potentially
affecting the system balance and the final price, adding further complexity. To
address these challenges, we propose a Monte Carlo Tree Search method that
publishes accurate imbalance prices while accounting for potential response
actions. Our approach models the system dynamics using a neural network
forecaster and a cluster of virtual batteries controlled by reinforcement
learning agents. Compared to Belgium's current publication method, our
technique improves price accuracy by 20.4% under ideal conditions and by 12.8%
in more realistic scenarios. This research addresses an unexplored, yet crucial
problem, positioning this paper as a pioneering work in analyzing the potential
of more advanced imbalance price publishing techniques.
| [
{
"version": "v1",
"created": "Wed, 6 Nov 2024 15:49:28 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 10:30:26 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Pavirani",
"Fabio",
""
],
[
"Van Gompel",
"Jonas",
""
],
[
"Madahi",
"Seyed Soroush Karimi",
""
],
[
"Claessens",
"Bert",
""
],
[
"Develder",
"Chris",
""
]
] |
2411.04228 | Aditya Mittal | Taha Abdullah, Arjun Ashok, Brandon Zarate, Shubhada Martha, Billy
Ouattara, Norman Matloff, Aditya Mittal | dsld: A Socially Relevant Tool for Teaching Statistics | To be submitted to journal | null | null | null | stat.ME cs.IR cs.LG stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The growing power of data science can play a crucial role in addressing
social discrimination, necessitating nuanced understanding and effective
mitigation strategies for biases. "Data Science Looks At Discrimination" (DSLD)
is an R and Python package designed to provide users with a comprehensive
toolkit of statistical and graphical methods for assessing possible
discrimination related to protected groups such as race, gender, and age. The
package addresses critical issues by identifying and mitigating confounders and
reducing bias against protected groups in prediction algorithms.
In educational settings, DSLD offers instructors powerful tools to teach
statistical principles through motivating real world examples of discrimination
analysis. The inclusion of an 80 page Quarto book further supports users from
statistics educators to legal professionals in effectively applying these
analytical tools to real world scenarios.
| [
{
"version": "v1",
"created": "Wed, 6 Nov 2024 19:50:00 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 17:23:08 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Abdullah",
"Taha",
""
],
[
"Ashok",
"Arjun",
""
],
[
"Zarate",
"Brandon",
""
],
[
"Martha",
"Shubhada",
""
],
[
"Ouattara",
"Billy",
""
],
[
"Matloff",
"Norman",
""
],
[
"Mittal",
"Aditya",
""
]
] |
2411.04374 | Lin Yang | Lin Yang, Sri Harsha Turlapati, Chen Lv, Domenico Campolo | Planning for quasi-static manipulation tasks via an intrinsic haptic
metric: a book insertion case study | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Contact-rich manipulation often requires strategic interactions with objects,
such as pushing to accomplish specific tasks. We propose a novel scenario where
a robot inserts a book into a crowded shelf by pushing aside neighboring books
to create space before slotting the new book into place. Classical planning
algorithms fail in this context due to limited space and their tendency to
avoid contact. Additionally, they do not handle indirectly manipulable objects
or consider force interactions. Our key contributions are: i) reframing
quasi-static manipulation as a planning problem on an implicit manifold derived
from equilibrium conditions; ii) utilizing an intrinsic haptic metric instead
of ad-hoc cost functions; and iii) proposing an adaptive algorithm that
simultaneously updates robot states, object positions, contact points, and
haptic distances. We evaluate our method on a crowded bookshelf insertion task,
and it can be generally applied to rigid body manipulation tasks. We propose
proxies to capture contact points and forces, with superellipses to represent
objects. This simplified model guarantees differentiability. Our framework
autonomously discovers strategic wedging-in policies while our simplified
contact model achieves behavior similar to real world scenarios. We also vary
the stiffness and initial positions to analyze our framework comprehensively.
The video can be found at https://youtu.be/eab8umZ3AQ0.
| [
{
"version": "v1",
"created": "Thu, 7 Nov 2024 02:14:52 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Feb 2025 04:09:00 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 03:11:20 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Yang",
"Lin",
""
],
[
"Turlapati",
"Sri Harsha",
""
],
[
"Lv",
"Chen",
""
],
[
"Campolo",
"Domenico",
""
]
] |
2411.05728 | Marcello Calvanese Strinati | Marcello Calvanese Strinati, Claudio Conti | Quantum hyperspins: Highly nonclassical collective behavior in quantum
optical parametric oscillators | 6 pages, 3 figures | Phys. Rev. A 111, 043712 (2025) | 10.1103/PhysRevA.111.043712 | null | quant-ph physics.comp-ph physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We report on the emergence of a highly non-classical collective behavior in
quantum parametric oscillators, which we name quantum hyperspin, induced by a
tailored nonlinear interaction. This is the second quantized version of
classical multidimensional spherical spins, as XY spins in two dimensions, and
Heisenberg spins in three dimensions. In the phase space, the quantum
hyperspins are represented as spherical shells whose radius scales with the
number of particles in a way such that it cannot be factorized even in the
limit of large particle number. We show that the nonlinearly coupled quantum
oscillators form a high-dimensional entangled state that is surprisingly robust
with respect to the coupling with the environment. Such a behavior results from
a properly engineered reservoir. Networks of entangled quantum hyperspins are a
new approach to quantum simulations for applications in computing, Ising
machines, and high-energy physics models. We analyze from first principles
through ab initio numerical simulations the properties of quantum hyperspins,
including the interplay of entanglement and coupling frustration.
| [
{
"version": "v1",
"created": "Fri, 8 Nov 2024 17:38:43 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Strinati",
"Marcello Calvanese",
""
],
[
"Conti",
"Claudio",
""
]
] |
2411.06161 | Gholamreza Davarpanah | Gholamreza Davarpanah, Hossein Shirzad, Jawad Faiz | A New 8/14 Two-Phase Switched Reluctance Motor with Improved Performance | null | 2023, pp. 1-6 | 10.1109/PEDSTC57673.2023.10087145 | 979-8-3503-3664-1 | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite their simple and robust structure, low cost, and simple cooling
system, switched reluctance motors (SRMs) face the challenge of low mean
torque. A possible solution is to change the structure of SRMs. This article
introduces an innovative combination of the number of rotor teeth and stator
teeth of a two-phase switch reluctance motor (TPSRM) with eight teeth for the
stator and fourteen teeth for the rotor. As a result of its unique design,
which has a short path for passing the main flux, it requires less
magnetomotive force. This leads to less core and copper loss, resulting in
increased efficiency. Each tooth of the stator in a phase develops a positive
torque during the rotation of the rotor, which increases the torque and
consequently increases the mean torque of the proposed TPSRM. A current
hysteresis control (CHC) is simulated by 2D FEM for the proposed 8/14 TPSRM and
the conventional 8/12 TPSRM under the same mechanical load on the shaft to get
a current hysteresis reference of 15A at the nominal speed of 600 rpm. To
verify the novelty and advantages of the suggested TPSRM, it is compared with
the conventional 8/12 TPSRM in terms of mean and peak torque, torque density,
and core and copper losses were compared. Lastly, the proposed 8/14 TPSRM is
shown to have better performance than the conventional 8/12 TPSRM.
| [
{
"version": "v1",
"created": "Sat, 9 Nov 2024 12:11:05 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Davarpanah",
"Gholamreza",
""
],
[
"Shirzad",
"Hossein",
""
],
[
"Faiz",
"Jawad",
""
]
] |
2411.06447 | Alex Finkelstein | Alex Finkelstein and Nikita Vladimirov and Moritz Zaiss and Or Perlman | Multi-Parameter Molecular MRI Quantification using Physics-Informed
Self-Supervised Learning | This project was funded by the European Union (ERC, BabyMagnet,
project no. 101115639), the Ministry of Innovation, Science and Technology,
Israel, and a grant from the Tel Aviv University Center for AI and Data
Science (TAD, The Blavatnik AI and Data Science Fund). None of above can be
held responsible for views and opinions expressed, which are those of the
authors alone | Commun Phys 8, 164 (2025) | 10.1038/s42005-025-02063-8 | null | physics.med-ph cs.LG physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Biophysical model fitting plays a key role in obtaining quantitative
parameters from physiological signals and images. However, the model complexity
for molecular magnetic resonance imaging (MRI) often translates into excessive
computation time, which makes clinical use impractical. Here, we present a
generic computational approach for solving the parameter extraction inverse
problem posed by ordinary differential equation (ODE) modeling coupled with
experimental measurement of the system dynamics. This is achieved by
formulating a numerical ODE solver to function as a step-wise analytical one,
thereby making it compatible with automatic differentiation-based optimization.
This enables efficient gradient-based model fitting, and provides a new
approach to parameter quantification based on self-supervised learning from a
single data observation. The neural-network-based train-by-fit pipeline was
used to quantify semisolid magnetization transfer (MT) and chemical exchange
saturation transfer (CEST) amide proton exchange parameters in the human brain,
in an in-vivo molecular MRI study (n = 4). The entire pipeline of the first
whole brain quantification was completed in 18.3 $\pm$ 8.3 minutes. Reusing the
single-subject-trained network for inference in new subjects took 1.0 $\pm$ 0.2
s, to provide results in agreement with literature values and scan-specific fit
results.
| [
{
"version": "v1",
"created": "Sun, 10 Nov 2024 12:40:33 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 05:54:02 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Finkelstein",
"Alex",
""
],
[
"Vladimirov",
"Nikita",
""
],
[
"Zaiss",
"Moritz",
""
],
[
"Perlman",
"Or",
""
]
] |
2411.06748 | Qingtong Wu | Qingtong Wu | Global Well-posedness and Long-time Behavior of the General
Ericksen--Leslie System in 2D under a Magnetic Field | 96 pages | null | null | null | math.AP math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we investigate the global well-posedness and long-time
behavior of the two-dimensional general Ericksen--Leslie system for a nematic
liquid crystal in a constant magnetic field. The PDE system consists of
Navier--Stokes equations and the harmonic heat flow equation for the
orientations of liquid crystal molecules. For incompressible nematic liquid
crystal fluids with either isotropic or anisotropic properties in torus
$\mathbb{T}^2$, we derive the global well-posedness of strong solutions through
higher-order energy estimates combined with compactness methods and acquire the
long-time behavior of the solutions by using the \L ojasiewicz--Simon
inequality after obtaining the boundedness of the nematic liquid crystal
molecules' angle.
| [
{
"version": "v1",
"created": "Mon, 11 Nov 2024 06:57:14 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 11:42:42 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Wu",
"Qingtong",
""
]
] |
2411.07265 | Huang Feiyang | Feiyang Huang | ViTOC: Vision Transformer and Object-aware Captioner | The core idea is too close to what has been published in other
journals | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents ViTOC (Vision Transformer and Object-aware Captioner), a
novel vision-language model for image captioning that addresses the challenges
of accuracy and diversity in generated descriptions. Unlike conventional
approaches, ViTOC employs a dual-path architecture based on Vision Transformer
and object detector, effectively fusing global visual features and local object
information through learnable vectors. The model introduces an innovative
object-aware prompting strategy that significantly enhances its capability in
handling long-tail data. Experiments on the standard COCO dataset demonstrate
that ViTOC outperforms baseline models across all evaluation metrics.
Additionally, we propose a reference-free evaluation method based on CLIP to
further validate the model's effectiveness. By utilizing pretrained visual
model parameters, ViTOC achieves efficient end-to-end training.
| [
{
"version": "v1",
"created": "Sat, 9 Nov 2024 13:13:49 GMT"
},
{
"version": "v2",
"created": "Wed, 13 Nov 2024 07:26:33 GMT"
},
{
"version": "v3",
"created": "Wed, 27 Nov 2024 15:45:23 GMT"
},
{
"version": "v4",
"created": "Thu, 17 Apr 2025 06:59:19 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Huang",
"Feiyang",
""
]
] |
2411.07447 | Kyoungmin Kim | Kyoungmin Kim, Kijae Hong, Caglar Gulcehre, Anastasia Ailamaki | Optimizing LLM Inference for Database Systems: Cost-Aware Scheduling for
Concurrent Requests | null | null | null | null | cs.PF cs.AI | http://creativecommons.org/licenses/by/4.0/ | LLMs are increasingly used inside database systems and in database
applications for better complexity management and decision-making, where LLM
inferences require significant GPU costs. LLM inference systems, however, are
slow compared to database systems, limiting the expansion of the use of LLMs
inside database systems. This paper first analyzes the LLM inference
performance and focuses on a data management issue in LLM inference. We reveal
that the root of the problem is the lack of an adequate resource cost model and
optimization strategy when executing multiple concurrent inference requests. We
adapt classic database multi-query optimization techniques by introducing cost
models for concurrent inference requests and new scheduling strategies to
optimize the use of memory resources by concurrent requests, thereby
substantially improving performance.
| [
{
"version": "v1",
"created": "Tue, 12 Nov 2024 00:10:34 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Nov 2024 21:57:16 GMT"
},
{
"version": "v3",
"created": "Wed, 16 Apr 2025 16:58:32 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Kim",
"Kyoungmin",
""
],
[
"Hong",
"Kijae",
""
],
[
"Gulcehre",
"Caglar",
""
],
[
"Ailamaki",
"Anastasia",
""
]
] |
2411.07466 | Kawshik Manikantan | Kawshik Manikantan, Makarand Tapaswi, Vineet Gandhi, Shubham Toshniwal | IdentifyMe: A Challenging Long-Context Mention Resolution Benchmark for
LLMs | 10 pages, 5 figures | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Recent evaluations of LLMs on coreference resolution have revealed that
traditional output formats and evaluation metrics do not fully capture the
models' referential understanding. To address this, we introduce IdentifyMe, a
new benchmark for mention resolution presented in a multiple-choice question
(MCQ) format, commonly used for evaluating LLMs. IdentifyMe features long
narratives and employs heuristics to exclude easily identifiable mentions,
creating a more challenging task. The benchmark also consists of a curated
mixture of different mention types and corresponding entities, allowing for a
fine-grained analysis of model performance. We evaluate both closed- and open
source LLMs on IdentifyMe and observe a significant performance gap (20-30%)
between the state-of-the-art sub-10B open models vs. closed ones. We observe
that pronominal mentions, which have limited surface information, are typically
much harder for models to resolve than nominal mentions. Additionally, we find
that LLMs often confuse entities when their mentions overlap in nested
structures. The highest-scoring model, GPT-4o, achieves 81.9% accuracy,
highlighting the strong referential capabilities of state-of-the-art LLMs while
also indicating room for further improvement.
| [
{
"version": "v1",
"created": "Tue, 12 Nov 2024 01:05:55 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 19:20:42 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Manikantan",
"Kawshik",
""
],
[
"Tapaswi",
"Makarand",
""
],
[
"Gandhi",
"Vineet",
""
],
[
"Toshniwal",
"Shubham",
""
]
] |
2411.07863 | Zhenkai Wu | Zhenkai Wu, Xiaowen Ma, Rongrong Lian, Kai Zheng and Wei Zhang | CDXLSTM: Boosting Remote Sensing Change Detection with Extended Long
Short-Term Memory | null | null | null | null | cs.CV cs.LG eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In complex scenes and varied conditions, effectively integrating
spatial-temporal context is crucial for accurately identifying changes.
However, current RS-CD methods lack a balanced consideration of performance and
efficiency. CNNs lack global context, Transformers are computationally
expensive, and Mambas face CUDA dependence and local correlation loss. In this
paper, we propose CDXLSTM, with a core component that is a powerful XLSTM-based
feature enhancement layer, integrating the advantages of linear computational
complexity, global context perception, and strong interpret-ability.
Specifically, we introduce a scale-specific Feature Enhancer layer,
incorporating a Cross-Temporal Global Perceptron customized for
semantic-accurate deep features, and a Cross-Temporal Spatial Refiner
customized for detail-rich shallow features. Additionally, we propose a
Cross-Scale Interactive Fusion module to progressively interact global change
representations with spatial responses. Extensive experimental results
demonstrate that CDXLSTM achieves state-of-the-art performance across three
benchmark datasets, offering a compelling balance between efficiency and
accuracy. Code is available at https://github.com/xwmaxwma/rschange.
| [
{
"version": "v1",
"created": "Tue, 12 Nov 2024 15:22:14 GMT"
},
{
"version": "v2",
"created": "Sat, 21 Dec 2024 03:10:42 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 16:05:10 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Wu",
"Zhenkai",
""
],
[
"Ma",
"Xiaowen",
""
],
[
"Lian",
"Rongrong",
""
],
[
"Zheng",
"Kai",
""
],
[
"Zhang",
"Wei",
""
]
] |
2411.08535 | Chalisa Veesommai Sillberg Dr. | Chalisa Veesommai Sillberg, Jose Siqueira De Cerqueira, Pekka
Sillberg, Kai-Kristian Kemell, and Pekka Abrahamsson | The EU AI Act is a good start but falls short | This is a preprint version of an article accepted to the 15th
International Conference on Software Business (ICSOB 2024). The final version
accepted for publishing the Springer Lecture Notes in Business Information
Processing (LNBIP) which contains 19 pages, 4 figures, and 3 tables | null | 10.1007/978-3-031-85849-9_10 | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The EU AI Act was created to ensure ethical and safe Artificial Intelligence
(AI) development and deployment across the EU. This study aims to identify key
challenges and strategies for helping enterprises focus on resources
effectively. To achieve this aim, we conducted a Multivocal Literature Review
(MLR) to explore the sentiments of both the industry and the academia. From 130
articles, 56 met the criteria. Our key findings are three-fold. First,
liability. Second, discrimination. Third, tool adequacy. Additionally, some
negative sentiments were expressed by industry and academia regarding
regulatory interpretations, specific requirements, and transparency issues.
Next, our findings are three essential themes for enterprises. First,
risk-based regulatory compliance. Second, ethical frameworks and principles in
technology development. Third, policies and systems for regulatory risk
management. These results identify the key challenges and strategies and
provide less commonly discussed themes, enabling enterprises to align with the
requirements and minimize their distance from the EU market.
| [
{
"version": "v1",
"created": "Wed, 13 Nov 2024 11:32:37 GMT"
},
{
"version": "v2",
"created": "Thu, 5 Dec 2024 13:15:07 GMT"
},
{
"version": "v3",
"created": "Mon, 9 Dec 2024 08:01:13 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Sillberg",
"Chalisa Veesommai",
""
],
[
"De Cerqueira",
"Jose Siqueira",
""
],
[
"Sillberg",
"Pekka",
""
],
[
"Kemell",
"Kai-Kristian",
""
],
[
"Abrahamsson",
"Pekka",
""
]
] |
2411.08947 | Alex Levchenko | Maxim Dzero, Jaglul Hasan, Alex Levchenko | Resonant second harmonic generation in a two-dimensional electron system | 10 pages, 3 figures | null | null | null | cond-mat.mes-hall cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the nonlinear response of a disordered two-dimensional electronic
system, lacking inversion symmetry, to an external alternating electric field.
The application of an in-plane static magnetic field induces local
contributions to the current density that are quadratic in the electric field
and linear in the magnetic field. This current oscillates at twice the
frequency of the external irradiation and there are two linearly independent
vector combinations that contribute to the current density. This particular
mechanism coexists with the topological Berry-dipole contribution to the second
harmonic of the current density, which can be generated by quantum confinement.
Additional nonlocal terms in the current density are possible in the regime
away from the normal incidence. The total current exhibits a nonreciprocal
character upon reversal of the magnetic field direction. We evaluate the
magnitude of this effect by computing its dependence on the strength of
spin-orbit coupling and the disorder scattering rate. Importantly, we show that
these local second-harmonic contributions can be resonantly excited when the
frequency of the external radiation approaches the energy separation between
the spin-orbit split bands.
| [
{
"version": "v1",
"created": "Wed, 13 Nov 2024 19:00:00 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 02:04:27 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Dzero",
"Maxim",
""
],
[
"Hasan",
"Jaglul",
""
],
[
"Levchenko",
"Alex",
""
]
] |
2411.09290 | Chengzhang Sun | Chengzhang Sun | Scalar curvature rigidity of parabolically convex domains in hyperbolic
spaces | Fixed some typos | null | null | null | math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For a parabolically convex domain $M\subseteq \mathbb{H}^n$, $n\ge 3$, we
prove that if $f:(N,\bar g)\to (M,g)$ has nonzero degree, where $N$ is spin
with scalar curvature $R_N\ge -n(n-1)$, and if $f|_{\partial N}$ does not
increase the distance and the mean curvature, then $N$ is hyperbolic, and
$\partial N$ is isometric to $\partial M$. This is a partial generalization of
Lott's result \cite{lott2021index} to negative lower bounds of scalar
curvature.
| [
{
"version": "v1",
"created": "Thu, 14 Nov 2024 09:07:05 GMT"
},
{
"version": "v2",
"created": "Tue, 3 Dec 2024 19:19:56 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 06:51:17 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Sun",
"Chengzhang",
""
]
] |
2411.09457 | Shyamal Biswas | Bikram Keshari Behera and Shyamal Biswas | Scaling theory for the collapse of a trapped Bose gas in a synthetic
magnetic field: a critical study at the condensation point | 12 pages, 4 figures | J. Stat. Mech. (2025) 043101 | 10.1088/1742-5468/adc711 | null | cond-mat.stat-mech | http://creativecommons.org/licenses/by/4.0/ | We have analytically explored both the zero temperature and the finite
temperature scaling theory for the collapse of an attractively interacting 3-D
harmonically trapped Bose gas in a synthetic magnetic field. We have considered
short-ranged (contact) attractive inter-particle interactions and Hartree-Fock
approximation for the same. We have separately studied the collapse of both the
condensate and the thermal cloud below and above the condensation point,
respectively. We have obtained an anisotropy, artificial magnetic field, and
temperature-dependent critical number of particles for the collapse of the
condensate. We have found a dramatic change in the critical exponent (from
$\alpha=1$ to $0$) of the specific heat ($C_v\propto|T-T_c|^{\alpha}$) when the
thermal cloud is about to collapse with the critical number of particles
($N=N_c$) just below and above the condensation point. All the results obtained
by us below and around the condensation point are experimentally testable
within the present-day experimental set-up for the ultracold systems in the
magneto-optical traps.
| [
{
"version": "v1",
"created": "Thu, 14 Nov 2024 14:06:36 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 11:46:08 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Behera",
"Bikram Keshari",
""
],
[
"Biswas",
"Shyamal",
""
]
] |
2411.09685 | Matteo Dell'Acqua | Michele Del Zotto, Matteo Dell'Acqua, Elias Riedel G{\aa}rding | The Higher Structure of Symmetries of Axion-Maxwell Theory | 62 pages, 17 figures. v2: minor clarifications, corrections and
formatting for submission | null | null | null | hep-th math-ph math.CT math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generalized symmetries of quantum field theories can be characterized by
topological defects/operators organized into a higher category. In this paper
we consider the Axion-Maxwell field theory in four dimensions and, building on
the construction of its topological defects by Choi, Lam, Shao, Hidaka, Nitta
and Yokokura, we discuss field theoretical methods to compute some aspects of
the higher structure of such category. In particular, we determine explicitly
the generalized F-symbols for the non-invertible electric 1-form symmetry of
the theory. Along the way, we clarify various aspects of the bottom-up
worldvolume approach towards the calculus of defects.
| [
{
"version": "v1",
"created": "Thu, 14 Nov 2024 18:52:31 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 19:43:02 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Del Zotto",
"Michele",
""
],
[
"Dell'Acqua",
"Matteo",
""
],
[
"Gårding",
"Elias Riedel",
""
]
] |
2411.10149 | Keiko Nagao | Keiko I. Nagao, Tatsuhiro Naka, and Takaaki Nomura | Directional Direct Detection of MeV Scale Boosted Dark Matter in Two
Component Dark Matter Scenario via Dark Photon Interaction | 19 pages, 11 figures, 3 tables | JCAP04(2025)030 | 10.1088/1475-7516/2025/04/030 | null | hep-ph astro-ph.CO hep-ex | http://creativecommons.org/licenses/by/4.0/ | This study explores a two-component dark matter model in which one component,
heavier dark matter, annihilates into a lighter dark matter. The lighter dark
matter is expected to generate detectable signals in detectors due to its
enhanced momentum, enabling direct detection even for MeV-scale dark matter. We
investigate the effectiveness of directional direct detections, especially the
nuclear emulsion detector NEWSdm, in verifying these boosted dark matter
particles through nuclear recoil. In particular, we focus on light nuclei, such
as protons and carbon, as suitable targets for this detection method due to
their high sensitivity to MeV-scale dark matter. By modeling the interactions
mediated by a dark photon in a hidden U(1)$_D$ gauge symmetry framework, we
calculate the expected dark matter flux and scattering rates for various
detector configurations. Our results show that nuclear emulsions have the
potential to yield distinct, direction-sensitive dark matter signals from the
Galactic center, providing a new way to probe low-mass dark matter parameter
spaces that evade conventional detection methods.
| [
{
"version": "v1",
"created": "Fri, 15 Nov 2024 12:42:55 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 09:43:01 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Nagao",
"Keiko I.",
""
],
[
"Naka",
"Tatsuhiro",
""
],
[
"Nomura",
"Takaaki",
""
]
] |
2411.10444 | Jake Robbennolt | Jake Robbennolt, Meiyi Li, Javad Mohammadi, Stephen D. Boyles | Balancing Passenger Transport and Power Distribution: A Distributed
Dispatch Policy for Shared Autonomous Electric Vehicles | null | null | 10.1109/TIA.2025.3559019 | null | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Shared autonomous electric vehicles can provide on-demand transportation for
passengers while also interacting extensively with the electric distribution
system. This interaction is especially beneficial after a disaster when the
large battery capacity of the fleet can be used to restore critical electric
loads. We develop a dispatch policy that balances the need to continue serving
passengers (especially critical workers) and the ability to transfer energy
across the network. The model predictive control policy tracks both passenger
and energy flows and provides maximum passenger throughput if any policy can.
The resulting mixed integer linear programming problem is difficult to solve
for large-scale problems, so a distributed solution approach is developed to
improve scalability, privacy, and resilience. We demonstrate that the proposed
heuristic, based on the alternating direction method of multipliers, is
effective in achieving near-optimal solutions quickly. The dispatch policy is
examined in simulation to demonstrate the ability of vehicles to balance these
competing objectives with benefits to both systems. Finally, we compare several
dispatch behaviors, demonstrating the importance of including operational
constraints and objectives from both the transportation and electric systems in
the model.
| [
{
"version": "v1",
"created": "Fri, 15 Nov 2024 18:59:45 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 22:12:29 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Robbennolt",
"Jake",
""
],
[
"Li",
"Meiyi",
""
],
[
"Mohammadi",
"Javad",
""
],
[
"Boyles",
"Stephen D.",
""
]
] |
2411.11467 | Amaury Wei | Amaury Wei and Olga Fink | Integrating Physics and Topology in Neural Networks for Learning Rigid
Body Dynamics | 19 pages, 10 figures | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Rigid body interactions are fundamental to numerous scientific disciplines,
but remain challenging to simulate due to their abrupt nonlinear nature and
sensitivity to complex, often unknown environmental factors. These challenges
call for adaptable learning-based methods capable of capturing complex
interactions beyond explicit physical models and simulations. While graph
neural networks can handle simple scenarios, they struggle with complex scenes
and long-term predictions. We introduce a novel framework for modeling rigid
body dynamics and learning collision interactions, addressing key limitations
of existing graph-based methods. Our approach extends the traditional
representation of meshes by incorporating higher-order topology complexes,
offering a physically consistent representation. Additionally, we propose a
physics-informed message-passing neural architecture, embedding physical laws
directly in the model. Our method demonstrates superior accuracy, even during
long rollouts, and exhibits strong generalization to unseen scenarios.
Importantly, this work addresses the challenge of multi-entity dynamic
interactions, with applications spanning diverse scientific and engineering
domains.
| [
{
"version": "v1",
"created": "Mon, 18 Nov 2024 11:03:15 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 13:43:47 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Wei",
"Amaury",
""
],
[
"Fink",
"Olga",
""
]
] |
2411.15954 | Gabriel Nahum | G. S. Nahum | A gradient model for the Bernstein polynomial basis | 16 pages, 3 figures | null | null | null | math.PR math-ph math.MP nlin.CG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We introduce a symmetric, gradient exclusion process within the class of
non-cooperative kinetically constrained lattice gases, modelling a non-linear
diffusivity in which the exchange of occupation values between two neighbouring
sites depends on the local density in specific boxes surrounding the pair. The
existence of such a model satisfying the gradient property is the main novelty
of this work, filling a gap in the literature regarding the types of
diffusivities attainable within this class of models. The resulting dynamics
exhibits similarities with the Bernstein polynomial basis and generalises the
Porous Media Model. We also introduce an auxiliary collection of processes,
which extend the Porous Media Model in a different direction and are related to
the former process via an inversion formula.
| [
{
"version": "v1",
"created": "Sun, 24 Nov 2024 18:52:47 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 21:56:42 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Nahum",
"G. S.",
""
]
] |
2411.16619 | Zhichao Zhang | Zhichao Zhang, Wei Sun, Xinyue Li, Yunhao Li, Qihang Ge, Jun Jia,
Zicheng Zhang, Zhongpeng Ji, Fengyu Sun, Shangling Jui, Xiongkuo Min,
Guangtao Zhai | Human-Activity AGV Quality Assessment: A Benchmark Dataset and an
Objective Evaluation Metric | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | AI-driven video generation techniques have made significant progress in
recent years. However, AI-generated videos (AGVs) involving human activities
often exhibit substantial visual and semantic distortions, hindering the
practical application of video generation technologies in real-world scenarios.
To address this challenge, we conduct a pioneering study on human activity AGV
quality assessment, focusing on visual quality evaluation and the
identification of semantic distortions. First, we construct the AI-Generated
Human activity Video Quality Assessment (Human-AGVQA) dataset, consisting of
6,000 AGVs derived from 15 popular text-to-video (T2V) models using 400 text
prompts that describe diverse human activities. We conduct a subjective study
to evaluate the human appearance quality, action continuity quality, and
overall video quality of AGVs, and identify semantic issues of human body
parts. Based on Human-AGVQA, we benchmark the performance of T2V models and
analyze their strengths and weaknesses in generating different categories of
human activities. Second, we develop an objective evaluation metric, named
AI-Generated Human activity Video Quality metric (GHVQ), to automatically
analyze the quality of human activity AGVs. GHVQ systematically extracts
human-focused quality features, AI-generated content-aware quality features,
and temporal continuity features, making it a comprehensive and explainable
quality metric for human activity AGVs. The extensive experimental results show
that GHVQ outperforms existing quality metrics on the Human-AGVQA dataset by a
large margin, demonstrating its efficacy in assessing the quality of human
activity AGVs. The Human-AGVQA dataset and GHVQ metric will be released
publicly.
| [
{
"version": "v1",
"created": "Mon, 25 Nov 2024 17:58:43 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 07:28:43 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Zhang",
"Zhichao",
""
],
[
"Sun",
"Wei",
""
],
[
"Li",
"Xinyue",
""
],
[
"Li",
"Yunhao",
""
],
[
"Ge",
"Qihang",
""
],
[
"Jia",
"Jun",
""
],
[
"Zhang",
"Zicheng",
""
],
[
"Ji",
"Zhongpeng",
""
],
[
"Sun",
"Fengyu",
""
],
[
"Jui",
"Shangling",
""
],
[
"Min",
"Xiongkuo",
""
],
[
"Zhai",
"Guangtao",
""
]
] |
2411.17180 | Xiaoyu Ma | Sylvain Sardy and Maxime van Cutsem and Xiaoyu Ma | Training a neural netwok for data reduction and better generalization | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | At the time of environmental concerns about artificial intelligence, in
particular its need for greedy storage and computation, sparsity inducing
neural networks offer a promising path towards frugality and solution for less
waste.
Sparse learners compress the inputs (features) by selecting only the ones
needed for good generalization. A human scientist can then give an intelligent
interpretation to the few selected features. If genes are the inputs and cancer
type is the output, then the selected genes give the cancerologist clues on
what genes have an effect on certain cancers. LASSO-type regularization leads
to good input selection for linear associations, but few attempts have been
made for nonlinear associations modeled as an artificial neural network. A
stringent but efficient way of testing whether a feature selection method works
is to check if a phase transition occurs in the probability of retrieving the
relevant features, as observed and mathematically studied for linear models.
Our method achieves just so for artificial neural networks, and, on real data,
it has the best compromise between number of selected features and
generalization performance.
Our method is flexible, applying to complex models ranging from shallow to
deep artificial neural networks and supporting various cost functions and
sparsity-promoting penalties. It does not rely on cross-validation or on a
validation set to select its single regularization parameter making it
user-friendly. Our approach can be seen as a form of compressed sensing for
complex models, allowing to distill high-dimensional data into a compact,
interpretable subset of meaningful features, just the opposite of a black box.
A python package is available at https://github.com/VcMaxouuu/AnnHarderLasso
containing all the simulations and ready-to-use models.
| [
{
"version": "v1",
"created": "Tue, 26 Nov 2024 07:41:15 GMT"
},
{
"version": "v2",
"created": "Sun, 1 Dec 2024 11:10:31 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 07:02:56 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Sardy",
"Sylvain",
""
],
[
"van Cutsem",
"Maxime",
""
],
[
"Ma",
"Xiaoyu",
""
]
] |
2411.17821 | Guglielmo Mazzola | Johannes Christmann, Petr Ivashkov, Mattia Chiurco, Guglielmo Mazzola | From quantum enhanced to quantum inspired Monte Carlo | JC and PI contributed equally | Phys. Rev. A 111, 042615, 2025 | 10.1103/PhysRevA.111.042615 | null | quant-ph cond-mat.stat-mech physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We perform a comprehensive analysis of the quantum-enhanced Monte Carlo
method [Nature, 619, 282-287 (2023)], aimed at identifying the optimal working
point of the algorithm. We observe an optimal mixing Hamiltonian strength and
analyze the scaling of the total evolution time with the size of the system. We
also explore extensions of the circuit, including the use of time-dependent
Hamiltonians and reverse digitized annealing. Additionally, we propose that
classical, approximate quantum simulators can be used for the proposal step
instead of the original real-hardware implementation. We observe that
tensor-network simulators, even with unconverged settings, can maintain a
scaling advantage over standard classical samplers. This may extend the utility
of quantum enhanced Monte Carlo as a quantum-inspired algorithm, even before
the deployment of large-scale quantum hardware.
| [
{
"version": "v1",
"created": "Tue, 26 Nov 2024 19:02:21 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Christmann",
"Johannes",
""
],
[
"Ivashkov",
"Petr",
""
],
[
"Chiurco",
"Mattia",
""
],
[
"Mazzola",
"Guglielmo",
""
]
] |
2411.18368 | Amruta Parulekar | Abhishek Gupta, Amruta Parulekar, Sameep Chattopadhyay, Preethi Jyothi | AMPS: ASR with Multimodal Paraphrase Supervision | null | null | null | null | cs.CL cs.AI cs.LG eess.AS | http://creativecommons.org/licenses/by-sa/4.0/ | Spontaneous or conversational multilingual speech presents many challenges
for state-of-the-art automatic speech recognition (ASR) systems. In this work,
we present a new technique AMPS that augments a multilingual multimodal ASR
system with paraphrase-based supervision for improved conversational ASR in
multiple languages, including Hindi, Marathi, Malayalam, Kannada, and Nyanja.
We use paraphrases of the reference transcriptions as additional supervision
while training the multimodal ASR model and selectively invoke this paraphrase
objective for utterances with poor ASR performance. Using AMPS with a
state-of-the-art multimodal model SeamlessM4T, we obtain significant relative
reductions in word error rates (WERs) of up to 5%. We present detailed analyses
of our system using both objective and human evaluation metrics.
| [
{
"version": "v1",
"created": "Wed, 27 Nov 2024 14:16:51 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 18:54:05 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Gupta",
"Abhishek",
""
],
[
"Parulekar",
"Amruta",
""
],
[
"Chattopadhyay",
"Sameep",
""
],
[
"Jyothi",
"Preethi",
""
]
] |
2411.19893 | Hiroto Yanagisawa | Hiroto Yanagisawa, Masami Ouchi, Kimihiko Nakajima, Yuichi Harikane,
Seiji Fujimoto, Yoshiaki Ono, Hiroya Umeda, Minami Nakane, Hidenobu Yajima,
Hajime Fukushima, and Yi Xu | A Galaxy with an Extremely Blue UV Slope $\beta=-3$ at $z=9.25$
Identified by JWST Spectroscopy: Evidence for a Weak Nebular Continuum and
Efficient Ionizing Photon Escape? | 11 pages, 10 figures, accepted for publication in ApJ | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | We investigate UV continuum slopes $\beta$ of 863 galaxies at $z=4-14$ using
archival JWST/NIRSpec PRISM spectra obtained from major JWST GTO, ERS, and GO
programs, including JADES, CEERS, and UNCOVER. Among these galaxies, we
identify a remarkable galaxy at $z=9.25$, dubbed EBG-1, with a significantly
blue UV slope $\beta=-2.99\pm0.15$, unlike the rest of the galaxies that
exhibit red continua or ambiguous blue continua hindered by large
uncertainties. We confirm that the $\beta$ value negligibly changes by the data
reduction and fitting wavelength ranges for UV emission/absorption line
masking. The extreme blue slope, $\beta=-3.0$, rules out significant
contributions from dust extinction or AGN activity. Comparing with stellar and
nebular emission models, we find that such a blue UV slope cannot be reproduced
solely by stellar models even with very young, metal-poor, or top-heavy
contiguous star formation associated with strong nebular continua making the UV
slopes red, but with a high ionizing photon escape fraction,
$f_\mathrm{esc}^\mathrm{ion} \gtrsim 0.5$, for a weak nebular continuum. While
the H$\beta$ emission line is not detected, likely due to the limited
sensitivity of the spectrum, we find moderately weak [O III]
$\lambda\lambda$4959,5007 emission lines for the given star-formation rate
($3\, \mathrm{M_\odot}$ yr$^{-1}$) and stellar mass ($10^{8.0} \,
\mathrm{M_\odot}$) that are about three times weaker than the average emission
lines, again suggestive of the high ionizing photon escape fraction,
$f_\mathrm{esc}^\mathrm{ion} \sim 0.7$ or more. EBG-1 would provide crucial
insights into stellar and nebular continuum emission in high-redshift galaxies,
serving as an example of the ionizing photon escaping site at the epoch of
reionization.
| [
{
"version": "v1",
"created": "Fri, 29 Nov 2024 17:58:15 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 05:51:50 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 06:28:17 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Yanagisawa",
"Hiroto",
""
],
[
"Ouchi",
"Masami",
""
],
[
"Nakajima",
"Kimihiko",
""
],
[
"Harikane",
"Yuichi",
""
],
[
"Fujimoto",
"Seiji",
""
],
[
"Ono",
"Yoshiaki",
""
],
[
"Umeda",
"Hiroya",
""
],
[
"Nakane",
"Minami",
""
],
[
"Yajima",
"Hidenobu",
""
],
[
"Fukushima",
"Hajime",
""
],
[
"Xu",
"Yi",
""
]
] |
2412.00202 | Miguel Quartin | Jo\~ao Ferri, Ian L. Tashiro, L. Raul Abramo, Isabela Matos, Miguel
Quartin and Riccardo Sturani | A robust cosmic standard ruler from the cross-correlations of galaxies
and dark sirens | v2: clarified the methodology. Accepted for publication in JCAP | JCAP 04(2025)008 | 10.1088/1475-7516/2025/04/008 | ET-0544A-24 | astro-ph.CO gr-qc | http://creativecommons.org/licenses/by/4.0/ | Observations of gravitational waves (GWs) from dark sirens allow us to infer
their locations and distances. Galaxies, on the other hand, have precise
angular positions but no direct measurement of their distances -- only
redshifts. The cross-correlation of GWs, which we limit here to binary black
hole mergers (BBH), in spherical shells of luminosity distance $D_L$, with
galaxies in shells of redshift $z$, leads to a direct measurement of the Hubble
diagram $D_L(z)$. Since this standard ruler relies only on the statistical
proximity of the dark sirens and galaxies (a general property of large-scale
structures), it is essentially model-independent: the correlation is maximal
when both redshift and $D_L$ shells coincide. We forecast the constraining
power of this technique, which we call {\it{Peak Sirens}}, for run 5 (O5) of
LIGO-Virgo-KAGRA (LVK), as well as for the third-generation observatories
Einstein Telescope and Cosmic Explorer. We employ thousands of full-sky light
cone simulations with realistic numbers for the tracers, and include masking by
the Milky Way, lensing and inhomogeneous GW sky coverage. We find that the
method is not expected to suffer from some of the issues present in other dark
siren methods, such as biased constraints due to incompleteness of galaxy
catalogs or dependence on priors for the merger rates of BBH. We show that with
Peak Sirens, given the projected O5 sensitivity, LVK can measure $H_0$ with
$7\%$ precision by itself, assuming $\Lambda$CDM, and $4\%$ precision using
external datasets to constrain $\Omega_m$. We also show that future
third-generation GW detectors can achieve, without external data, sub-percent
uncertainties in $H_0$ assuming $\Lambda$CDM, and 3\% in a more flexible
$w_0w_a$CDM model. The method also shows remarkable robustness against
systematic effects such as the modeling of non-linear structure formation.
| [
{
"version": "v1",
"created": "Fri, 29 Nov 2024 19:01:11 GMT"
},
{
"version": "v2",
"created": "Thu, 6 Mar 2025 04:01:35 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Ferri",
"João",
""
],
[
"Tashiro",
"Ian L.",
""
],
[
"Abramo",
"L. Raul",
""
],
[
"Matos",
"Isabela",
""
],
[
"Quartin",
"Miguel",
""
],
[
"Sturani",
"Riccardo",
""
]
] |
2412.00986 | Tiziano De Angelis Dr | Tiziano De Angelis, Caio C\'esar Graciani Rodrigues, Peter Tankov | A model of strategic sustainable investment | 44 pages; 9 figures; improved exposition and expanded numerical
analysis | null | null | null | q-fin.MF math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study a problem of optimal irreversible investment and emission reduction
formulated as a nonzero-sum dynamic game between an investor with environmental
preferences and a firm. The game is set in continuous time on an infinite-time
horizon. The firm generates profits with a stochastic dynamics and may spend
part of its revenues towards emission reduction (e.g., renovating the
infrastructure). The firm's objective is to maximize the discounted expectation
of a function of its profits. The investor participates in the profits, may
decide to invest to support the firm's production capacity and uses a profit
function which accounts for both financial and environmental factors. Nash
equilibria of the game are obtained via a system of variational inequalities.
We formulate a general verification theorem for this system in a diffusive
setup and construct an explicit solution in the zero-noise limit. Our explicit
results and numerical approximations show that both the investor's and the
firm's optimal actions are triggered by moving boundaries that increase with
the total amount of emission abatement.
| [
{
"version": "v1",
"created": "Sun, 1 Dec 2024 22:27:00 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 12:08:48 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"De Angelis",
"Tiziano",
""
],
[
"Rodrigues",
"Caio César Graciani",
""
],
[
"Tankov",
"Peter",
""
]
] |
2412.01113 | Keito Kudo | Keito Kudo, Yoichi Aoki, Tatsuki Kuribayashi, Shusaku Sone, Masaya
Taniguchi, Ana Brassard, Keisuke Sakaguchi, Kentaro Inui | Think-to-Talk or Talk-to-Think? When LLMs Come Up with an Answer in
Multi-Step Arithmetic Reasoning | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This study investigates the internal reasoning process of language models
during arithmetic multi-step reasoning, motivated by the question of when they
internally form their answers during reasoning. Particularly, we inspect
whether the answer is determined before or after chain-of-thought (CoT) begins
to determine whether models follow a post-hoc Think-to-Talk mode or a
step-by-step Talk-to-Think mode of explanation. Through causal probing
experiments in controlled arithmetic reasoning tasks, we found systematic
internal reasoning patterns across models in our case study; for example,
single-step subproblems are solved before CoT begins, and more complicated
multi-step calculations are performed during CoT.
| [
{
"version": "v1",
"created": "Mon, 2 Dec 2024 04:35:54 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 06:57:50 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Kudo",
"Keito",
""
],
[
"Aoki",
"Yoichi",
""
],
[
"Kuribayashi",
"Tatsuki",
""
],
[
"Sone",
"Shusaku",
""
],
[
"Taniguchi",
"Masaya",
""
],
[
"Brassard",
"Ana",
""
],
[
"Sakaguchi",
"Keisuke",
""
],
[
"Inui",
"Kentaro",
""
]
] |
2412.01506 | Jianfeng Xiang | Jianfeng Xiang, Zelong Lv, Sicheng Xu, Yu Deng, Ruicheng Wang, Bowen
Zhang, Dong Chen, Xin Tong, Jiaolong Yang | Structured 3D Latents for Scalable and Versatile 3D Generation | Project Page: https://github.com/Microsoft/TRELLIS | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a novel 3D generation method for versatile and high-quality 3D
asset creation. The cornerstone is a unified Structured LATent (SLAT)
representation which allows decoding to different output formats, such as
Radiance Fields, 3D Gaussians, and meshes. This is achieved by integrating a
sparsely-populated 3D grid with dense multiview visual features extracted from
a powerful vision foundation model, comprehensively capturing both structural
(geometry) and textural (appearance) information while maintaining flexibility
during decoding. We employ rectified flow transformers tailored for SLAT as our
3D generation models and train models with up to 2 billion parameters on a
large 3D asset dataset of 500K diverse objects. Our model generates
high-quality results with text or image conditions, significantly surpassing
existing methods, including recent ones at similar scales. We showcase flexible
output format selection and local 3D editing capabilities which were not
offered by previous models. Code, model, and data will be released.
| [
{
"version": "v1",
"created": "Mon, 2 Dec 2024 13:58:38 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 03:14:38 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Xiang",
"Jianfeng",
""
],
[
"Lv",
"Zelong",
""
],
[
"Xu",
"Sicheng",
""
],
[
"Deng",
"Yu",
""
],
[
"Wang",
"Ruicheng",
""
],
[
"Zhang",
"Bowen",
""
],
[
"Chen",
"Dong",
""
],
[
"Tong",
"Xin",
""
],
[
"Yang",
"Jiaolong",
""
]
] |
2412.01798 | Lan Wang | Lan Wang, Yujia Chen, Du Tran, Vishnu Naresh Boddeti, Wen-Sheng Chu | SEAL: Semantic Attention Learning for Long Video Representation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Long video understanding presents challenges due to the inherent high
computational complexity and redundant temporal information. An effective
representation for long videos must efficiently process such redundancy while
preserving essential contents for downstream tasks. This paper introduces
SEmantic Attention Learning (SEAL), a novel unified representation for long
videos. To reduce computational complexity, long videos are decomposed into
three distinct types of semantic entities: scenes, objects, and actions,
allowing models to operate on a compact set of entities rather than a large
number of frames or pixels. To further address redundancy, we propose an
attention learning module that balances token relevance with diversity,
formulated as a subset selection optimization problem. Our representation is
versatile and applicable across various long video understanding tasks.
Extensive experiments demonstrate that SEAL significantly outperforms
state-of-the-art methods in video question answering and temporal grounding
tasks across diverse benchmarks, including LVBench, MovieChat-1K, and Ego4D.
| [
{
"version": "v1",
"created": "Mon, 2 Dec 2024 18:46:12 GMT"
},
{
"version": "v2",
"created": "Thu, 5 Dec 2024 13:39:06 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 02:23:52 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Wang",
"Lan",
""
],
[
"Chen",
"Yujia",
""
],
[
"Tran",
"Du",
""
],
[
"Boddeti",
"Vishnu Naresh",
""
],
[
"Chu",
"Wen-Sheng",
""
]
] |
2412.02361 | Xiyang Sun | Matteo Baggioli, Yanyan Bu, Xiyang Sun | Chiral Anomalous Magnetohydrodynamics in action: effective field theory
and holography | 21 pages, 1 figure | JHEP04(2025)126 | 10.1007/JHEP04(2025)126 | null | hep-th cond-mat.mes-hall nucl-th | http://creativecommons.org/licenses/by/4.0/ | Chiral Anomalous Magnetohydrodynamics (CAMHD) provides a low-energy effective
framework for describing chiral fluids in the presence of dynamical
electromagnetic fields and axial anomaly. This theory finds applications across
diverse physical systems, including heavy-ion collisions, the early universe,
and Weyl/Dirac semimetals. Along with Schwinger-Keldysh (SK) effective
theories, holographic models serve as a complementary tool to provide a
systematic formulation of CAMHD that goes beyond the weak coupling regime. In
this work, we explore holographic models with $U(1)_A \times U(1)$ symmetry,
where the electromagnetic $U(1)$ field is rendered dynamical through mixed
boundary conditions applied to the bulk gauge field and the axial anomaly is
introduced via a Chern-Simons bulk term. Through a detailed holographic SK
analysis, we demonstrate that the low-energy effective action derived from this
model aligns precisely with the SK field theory proposed by Landry and Liu and,
in fact, it generalizes it to scenarios with finite background axial field.
This alignment not only validates the holographic model but also paves the way
for its use in exploring unresolved aspects of CAMHD, such as the recently
proposed chiral magnetic electric separation wave and nonlinear chiral
instabilities.
| [
{
"version": "v1",
"created": "Tue, 3 Dec 2024 10:40:22 GMT"
},
{
"version": "v2",
"created": "Tue, 10 Dec 2024 02:36:04 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 08:16:09 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Baggioli",
"Matteo",
""
],
[
"Bu",
"Yanyan",
""
],
[
"Sun",
"Xiyang",
""
]
] |
2412.03421 | Laurent H\'ebert-Dufresne | Laurent H\'ebert-Dufresne, Nicholas W. Landry, Juniper Lovato,
Jonathan St-Onge, Jean-Gabriel Young, Marie-\`Eve Couture-M\'enard,
St\'ephane Bernatchez, Catherine Choquette and Alan A. Cohen | Governance as a complex, networked, democratic, satisfiability problem | null | null | null | null | physics.soc-ph cs.SI nlin.AO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Democratic governments comprise a subset of a population whose goal is to
produce coherent decisions, solving societal challenges while respecting the
will of the people. New governance frameworks represent this as a social
network rather than as a hierarchical pyramid with centralized authority. But
how should this network be structured? We model the decisions a population must
make as a satisfiability problem and the structure of information flow involved
in decision-making as a social hypergraph. This framework allows to consider
different governance structures, from dictatorships to direct democracy.
Between these extremes, we find a regime of effective governance where small
overlapping decision groups make specific decisions and share information.
Effective governance allows even incoherent or polarized populations to make
coherent decisions at low coordination costs. Beyond simulations, our
conceptual framework can explore a wide range of governance strategies and
their ability to tackle decision problems that challenge standard governments.
| [
{
"version": "v1",
"created": "Wed, 4 Dec 2024 16:01:34 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 11:57:21 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Hébert-Dufresne",
"Laurent",
""
],
[
"Landry",
"Nicholas W.",
""
],
[
"Lovato",
"Juniper",
""
],
[
"St-Onge",
"Jonathan",
""
],
[
"Young",
"Jean-Gabriel",
""
],
[
"Couture-Ménard",
"Marie-Ève",
""
],
[
"Bernatchez",
"Stéphane",
""
],
[
"Choquette",
"Catherine",
""
],
[
"Cohen",
"Alan A.",
""
]
] |
2412.03696 | Pierre Bousseyroux | Pierre Bousseyroux and Jean-Philippe Bouchaud | Free Convolution and Generalized Dyson Brownian Motion | null | null | null | null | cond-mat.dis-nn math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The eigenvalue spectrum of the sum of large random matrices that are mutually
"free", i.e., randomly rotated, can be obtained using the formalism of
R-transforms, with many applications in different fields. We provide a direct
interpretation of the otherwise abstract additivity property of R-transforms
for the sum in terms of a dynamical evolution of "particles" (the eigenvalues),
interacting through two-body and higher-body forces and subject to a Gaussian
noise, generalizing the usual Dyson Brownian motion with Coulomb interaction.
Interestingly, the appearance of an outlier outside of the bulk of the spectrum
is signalled by a divergence of the "velocity" of the generalized Dyson motion.
We extend our result to products of free matrices.
| [
{
"version": "v1",
"created": "Wed, 4 Dec 2024 20:20:19 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 13:32:02 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Bousseyroux",
"Pierre",
""
],
[
"Bouchaud",
"Jean-Philippe",
""
]
] |
2412.04161 | Gabriele Fissore | Riccardo Cristoferi, Gabriele Fissore, Marco Morandotti | Geometrically constrained walls in three dimensions | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study geometrically constrained magnetic walls in a three dimensional
geometry where two bulks are connected by a thin neck. Without imposing any
symmetry assumption on the domain, we investigate the scaling of the energy as
the size of the neck vanishes. We identify five significant scaling regimes,
for all of which we characterise the energy scaling and identify the asymptotic
behaviour of the domain wall. Finally, we notice the emergence of sub-regimes
that are not present in the previous works due to restrictive symmetry
assumptions.
| [
{
"version": "v1",
"created": "Thu, 5 Dec 2024 13:52:57 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 10:43:09 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Cristoferi",
"Riccardo",
""
],
[
"Fissore",
"Gabriele",
""
],
[
"Morandotti",
"Marco",
""
]
] |
2412.04833 | Peiyan Hu | Peiyan Hu, Rui Wang, Xiang Zheng, Tao Zhang, Haodong Feng, Ruiqi Feng,
Long Wei, Yue Wang, Zhi-Ming Ma, Tailin Wu | Wavelet Diffusion Neural Operator | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Simulating and controlling physical systems described by partial differential
equations (PDEs) are crucial tasks across science and engineering. Recently,
diffusion generative models have emerged as a competitive class of methods for
these tasks due to their ability to capture long-term dependencies and model
high-dimensional states. However, diffusion models typically struggle with
handling system states with abrupt changes and generalizing to higher
resolutions. In this work, we propose Wavelet Diffusion Neural Operator (WDNO),
a novel PDE simulation and control framework that enhances the handling of
these complexities. WDNO comprises two key innovations. Firstly, WDNO performs
diffusion-based generative modeling in the wavelet domain for the entire
trajectory to handle abrupt changes and long-term dependencies effectively.
Secondly, to address the issue of poor generalization across different
resolutions, which is one of the fundamental tasks in modeling physical
systems, we introduce multi-resolution training. We validate WDNO on five
physical systems, including 1D advection equation, three challenging physical
systems with abrupt changes (1D Burgers' equation, 1D compressible
Navier-Stokes equation and 2D incompressible fluid), and a real-world dataset
ERA5, which demonstrates superior performance on both simulation and control
tasks over state-of-the-art methods, with significant improvements in long-term
and detail prediction accuracy. Remarkably, in the challenging context of the
2D high-dimensional and indirect control task aimed at reducing smoke leakage,
WDNO reduces the leakage by 33.2% compared to the second-best baseline. The
code can be found at https://github.com/AI4Science-WestlakeU/wdno.git.
| [
{
"version": "v1",
"created": "Fri, 6 Dec 2024 07:56:25 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 12:34:20 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Hu",
"Peiyan",
""
],
[
"Wang",
"Rui",
""
],
[
"Zheng",
"Xiang",
""
],
[
"Zhang",
"Tao",
""
],
[
"Feng",
"Haodong",
""
],
[
"Feng",
"Ruiqi",
""
],
[
"Wei",
"Long",
""
],
[
"Wang",
"Yue",
""
],
[
"Ma",
"Zhi-Ming",
""
],
[
"Wu",
"Tailin",
""
]
] |
2412.05486 | Lan Wu | Lan Wu, Craig Jin, Monisha Mushtary Uttsha and Teresa Vidal-Calleja | Listen to Your Map: An Online Representation for Spatial Sonification | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Robotic perception is becoming a key technology for navigation aids,
especially helping individuals with visual impairments through spatial
sonification. This paper introduces a mapping representation that accurately
captures scene geometry for sonification, turning physical spaces into auditory
experiences. Using depth sensors, we encode an incrementally built 3D scene
into a compact 360-degree representation with angular and distance information,
aligning this way with human auditory spatial perception. The proposed
framework performs localisation and mapping via VDB-Gaussian Process Distance
Fields for efficient online scene reconstruction. The key aspect is a
sensor-centric structure that maintains either a 2D-circular or 3D-cylindrical
raster-based projection. This spatial representation is then converted into
binaural auditory signals using simple pre-recorded responses from a
representative room. Quantitative and qualitative evaluations show improvements
in accuracy, coverage, timing and suitability for sonification compared to
other approaches, with effective handling of dynamic objects as well. An
accompanying video demonstrates spatial sonification in room-like environments.
https://tinyurl.com/ListenToYourMap
| [
{
"version": "v1",
"created": "Sat, 7 Dec 2024 01:15:07 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 07:23:01 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Wu",
"Lan",
""
],
[
"Jin",
"Craig",
""
],
[
"Uttsha",
"Monisha Mushtary",
""
],
[
"Vidal-Calleja",
"Teresa",
""
]
] |
2412.05584 | Lintong Du | Lintong Du, Huazhen Liu, Yijia Zhang, ShuXin Liu, Yuan Qu, Zenghui
Zhang, and Jiamiao Yang | UMSPU: Universal Multi-Size Phase Unwrapping via Mutual
Self-Distillation and Adaptive Boosting Ensemble Segmenters | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spatial phase unwrapping is a key technique for extracting phase information
to obtain 3D morphology and other features. Modern industrial measurement
scenarios demand high precision, large image sizes, and high speed. However,
conventional methods struggle with noise resistance and processing speed.
Current deep learning methods are limited by the receptive field size and
sparse semantic information, making them ineffective for large size images. To
address this issue, we propose a mutual self-distillation (MSD) mechanism and
adaptive boosting ensemble segmenters to construct a universal multi-size phase
unwrapping network (UMSPU). MSD performs hierarchical attention refinement and
achieves cross-layer collaborative learning through bidirectional distillation,
ensuring fine-grained semantic representation across image sizes. The adaptive
boosting ensemble segmenters combine weak segmenters with different receptive
fields into a strong one, ensuring stable segmentation across spatial
frequencies. Experimental results show that UMSPU overcomes image size
limitations, achieving high precision across image sizes ranging from 256*256
to 2048*2048 (an 8 times increase). It also outperforms existing methods in
speed, robustness, and generalization. Its practicality is further validated in
structured light imaging and InSAR. We believe that UMSPU offers a universal
solution for phase unwrapping, with broad potential for industrial
applications.
| [
{
"version": "v1",
"created": "Sat, 7 Dec 2024 08:38:29 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 03:41:57 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Du",
"Lintong",
""
],
[
"Liu",
"Huazhen",
""
],
[
"Zhang",
"Yijia",
""
],
[
"Liu",
"ShuXin",
""
],
[
"Qu",
"Yuan",
""
],
[
"Zhang",
"Zenghui",
""
],
[
"Yang",
"Jiamiao",
""
]
] |
2412.06031 | Srivatsav Kunnawalkam Elayavalli | Tattwamasi Amrutam, David Gao, Srivatsav Kunnawalkam Elayavalli and
Gregory Patchell | Strict comparison in reduced group $C^*$-algebras | In memory of Dr S Balachander. 14 pages. Comments welcome. In v2:
added a new family of examples groups fitting into our result, heirarchically
hyperbolic groups. Thanks to J. Behrstock for pointing this out to us. In v3:
fixed some typographical errors and made minor edits, also added alternative
argument in free group case shared by M. Magee | null | null | null | math.OA math.FA math.GR math.LO | http://creativecommons.org/publicdomain/zero/1.0/ | We prove that for every $n\geq 2$, the reduced group $C^*$-algebras of the
countable free groups $C^*_r(\mathbb{F}_n)$ have strict comparison. Our method
works in a general setting: for $G$ in a large family of non-amenable groups,
including hyperbolic groups, free products, mapping class groups, right-angled
Artin groups etc., we have $C^*_r(G)$ have strict comparison. This work also
has several applications in the theory of $C^*$-algebras including: resolving
Leonel Robert's selflessness problem for $C^*_r(G)$; uniqueness of embeddings
of the Jiang-Su algebra $\mathcal{Z}$ up to approximate unitary equivalence
into $C^*_r(G)$; full computations of the Cuntz semigroup of $C^*_r(G)$ and
future directions in the $C^*$-classification program.
| [
{
"version": "v1",
"created": "Sun, 8 Dec 2024 19:00:56 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Mar 2025 18:31:26 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 17:41:14 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Amrutam",
"Tattwamasi",
""
],
[
"Gao",
"David",
""
],
[
"Elayavalli",
"Srivatsav Kunnawalkam",
""
],
[
"Patchell",
"Gregory",
""
]
] |
2412.06235 | Michael Yeung | Michael Yeung, Toya Teramoto, Songtao Wu, Tatsuo Fujiwara, Kenji
Suzuki, Tamaki Kojima | VariFace: Fair and Diverse Synthetic Dataset Generation for Face
Recognition | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | The use of large-scale, web-scraped datasets to train face recognition models
has raised significant privacy and bias concerns. Synthetic methods mitigate
these concerns and provide scalable and controllable face generation to enable
fair and accurate face recognition. However, existing synthetic datasets
display limited intraclass and interclass diversity and do not match the face
recognition performance obtained using real datasets. Here, we propose
VariFace, a two-stage diffusion-based pipeline to create fair and diverse
synthetic face datasets to train face recognition models. Specifically, we
introduce three methods: Face Recognition Consistency to refine demographic
labels, Face Vendi Score Guidance to improve interclass diversity, and
Divergence Score Conditioning to balance the identity preservation-intraclass
diversity trade-off. When constrained to the same dataset size, VariFace
considerably outperforms previous synthetic datasets (0.9200 $\rightarrow$
0.9405) and achieves comparable performance to face recognition models trained
with real data (Real Gap = -0.0065). In an unconstrained setting, VariFace not
only consistently achieves better performance compared to previous synthetic
methods across dataset sizes but also, for the first time, outperforms the real
dataset (CASIA-WebFace) across six evaluation datasets. This sets a new
state-of-the-art performance with an average face verification accuracy of
0.9567 (Real Gap = +0.0097) across LFW, CFP-FP, CPLFW, AgeDB, and CALFW
datasets and 0.9366 (Real Gap = +0.0380) on the RFW dataset.
| [
{
"version": "v1",
"created": "Mon, 9 Dec 2024 06:21:11 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 12:12:38 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Yeung",
"Michael",
""
],
[
"Teramoto",
"Toya",
""
],
[
"Wu",
"Songtao",
""
],
[
"Fujiwara",
"Tatsuo",
""
],
[
"Suzuki",
"Kenji",
""
],
[
"Kojima",
"Tamaki",
""
]
] |
2412.06279 | Qian Li | Qian Li, Ziang Yang, Dou Li, Hongliang Zhang | Reconfigurable Holographic Surface-aided Distributed MIMO Radar Systems | null | null | null | null | eess.SP cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Distributed phased Multiple-Input Multiple-Output (phased-MIMO) radar systems
have attracted wide attention in target detection and tracking. However, the
phase-shifting circuits in phased subarrays contribute to high power
consumption and hardware cost. To address this issue, an energy-efficient and
cost-efficient metamaterial antenna array, i.e., reconfigurable holographic
surface (RHS), has been developed. In this letter, we propose RHS-aided
distributed MIMO radar systems to achieve more accurate multi-target detection
under equivalent power consumption and hardware cost as that of distributed
phased-MIMO radar systems. Different from phased arrays, the RHS achieves beam
steering by regulating the radiation amplitude of its elements, and thus
conventional beamforming schemes designed for phased arrays are no longer
applicable. Aiming to maximize detection accuracy, we design an
amplitude-controlled beamforming scheme for multiple RHS transceiver subarrays.
The simulations validate the superiority of the proposed scheme over the
distributed phased-MIMO radar scheme and reveal the optimal allocation of
spatial diversity and coherent processing gain that leads to the best system
performance when hardware resources are fixed.
| [
{
"version": "v1",
"created": "Mon, 9 Dec 2024 07:55:49 GMT"
},
{
"version": "v2",
"created": "Wed, 5 Feb 2025 09:27:00 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 13:11:56 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Li",
"Qian",
""
],
[
"Yang",
"Ziang",
""
],
[
"Li",
"Dou",
""
],
[
"Zhang",
"Hongliang",
""
]
] |
2412.07142 | Guozhen Shen | Ruihuan Mao and Guozhen Shen | A note on dual Dedekind finiteness | 6 pages | null | null | null | math.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A set $A$ is dually Dedekind finite if every surjection from $A$ onto $A$ is
injective; otherwise, $A$ is dually Dedekind infinite. It is proved consistent
with $\mathsf{ZF}$ (i.e., the Zermelo--Fraenkel set theory without the axiom of
choice) that there exists a family $\langle A_n\rangle_{n\in\omega}$ of sets
such that, for all $n\in\omega$, $A_n^n$ is dually Dedekind finite whereas
$A_n^{n+1}$ is dually Dedekind infinite. This resolves a question that was left
open in [J. Truss, Fund. Math. 84, 187--208 (1974)].
| [
{
"version": "v1",
"created": "Tue, 10 Dec 2024 03:06:03 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 03:09:37 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Mao",
"Ruihuan",
""
],
[
"Shen",
"Guozhen",
""
]
] |
2412.07383 | Gesualdo Delfino | Gesualdo Delfino | Critical exponents at the Nishimori point | footnote added, published version | J. Stat. Mech. (2025) 043203 | 10.1088/1742-5468/adc4ce | null | cond-mat.stat-mech cond-mat.dis-nn hep-th math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Nishimori point of the random bond Ising model is a prototype of
renormalization group fixed points with strong disorder. We show that the exact
correlation length and crossover critical exponents at this point can be
identified in two and three spatial dimensions starting from properties of the
Nishimori line. These are the first exact exponents for frustrated random
magnets, a circumstance to be also contrasted with the fact that the exact
exponents of the Ising model without disorder are not known in three
dimensions. Our considerations extend to higher dimensions and models other
than Ising.
| [
{
"version": "v1",
"created": "Tue, 10 Dec 2024 10:29:08 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Apr 2025 14:27:38 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Delfino",
"Gesualdo",
""
]
] |
2412.07468 | Kai Yuan | Kai Yuan, Jiahao Zhang, Yidi Wang, Xiaobing Pei | AHSG: Adversarial Attack on High-level Semantics in Graph Neural
Networks | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial attacks on Graph Neural Networks aim to perturb the performance
of the learner by carefully modifying the graph topology and node attributes.
Existing methods achieve attack stealthiness by constraining the modification
budget and differences in graph properties. However, these methods typically
disrupt task-relevant primary semantics directly, which results in low
defensibility and detectability of the attack. In this paper, we propose an
Adversarial Attack on High-level Semantics for Graph Neural Networks (AHSG),
which is a graph structure attack model that ensures the retention of primary
semantics. By combining latent representations with shared primary semantics,
our model retains detectable attributes and relational patterns of the original
graph while leveraging more subtle changes to carry out the attack. Then we use
the Projected Gradient Descent algorithm to map the latent representations with
attack effects to the adversarial graph. Through experiments on robust graph
deep learning models equipped with defense strategies, we demonstrate that AHSG
outperforms other state-of-the-art methods in attack effectiveness.
Additionally, using Contextual Stochastic Block Models to detect the attacked
graph further validates that our method preserves the primary semantics of the
graph.
| [
{
"version": "v1",
"created": "Tue, 10 Dec 2024 12:35:37 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 16:13:40 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Yuan",
"Kai",
""
],
[
"Zhang",
"Jiahao",
""
],
[
"Wang",
"Yidi",
""
],
[
"Pei",
"Xiaobing",
""
]
] |
2412.08205 | Cosimo Bambi | Yuhao Guo, Swarnim Shashank, Cosimo Bambi | Quasi-normal modes of slowly-rotating Johannsen black holes | 17 pages, 7 figures. v2: refereed version | Eur. Phys. J. C (2025) 85:425 | 10.1140/epjc/s10052-025-14118-9 | null | gr-qc astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The detection of gravitational waves with ground-based laser interferometers
has opened a new window to test and constrain General Relativity (GR) in the
strong, dynamical, and non-linear regime. In this paper, we follow an agnostic
approach and we study the quasi-normal modes of gravitational perturbations of
Johannsen black holes under the assumptions of the validity of the Einstein
Equations and of low values of the black hole spin parameter and deformation
parameters. We find that the deformation parameter $\alpha_{13}$ has a stronger
impact on the quasi-normal modes than the other leading order deformation
parameters ($\alpha_{22}$, $\alpha_{52}$, and $\epsilon_{3}$). We derive a
fitting formula for the fundamental modes with $l=2$ and $l=3$ for the
deformation parameter $\alpha_{13}$ valid in the slow rotation approximation
($a_* < 0.4$). Finally, we constrain $\alpha_{13}$ from the event GW170104;
within our analysis, we find that the data of GW170104 are consistent with the
predictions of GR.
| [
{
"version": "v1",
"created": "Wed, 11 Dec 2024 08:53:12 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 11:28:01 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Guo",
"Yuhao",
""
],
[
"Shashank",
"Swarnim",
""
],
[
"Bambi",
"Cosimo",
""
]
] |
2412.08222 | Hanzhe Yang | Hanzhe Yang, Youlong Wu, Dingzhu Wen, Yong Zhou, Yuanming Shi | Structured IB: Improving Information Bottleneck with Structured Feature
Learning | null | Proceedings of the AAAI Conference on Artificial Intelligence,
2025, Vol. 39, No. 20 | 10.1609/aaai.v39i20.35499 | null | cs.IT cs.LG math.IT | http://creativecommons.org/licenses/by/4.0/ | The Information Bottleneck (IB) principle has emerged as a promising approach
for enhancing the generalization, robustness, and interpretability of deep
neural networks, demonstrating efficacy across image segmentation, document
clustering, and semantic communication. Among IB implementations, the IB
Lagrangian method, employing Lagrangian multipliers, is widely adopted. While
numerous methods for the optimizations of IB Lagrangian based on variational
bounds and neural estimators are feasible, their performance is highly
dependent on the quality of their design, which is inherently prone to errors.
To address this limitation, we introduce Structured IB, a framework for
investigating potential structured features. By incorporating auxiliary
encoders to extract missing informative features, we generate more informative
representations. Our experiments demonstrate superior prediction accuracy and
task-relevant information preservation compared to the original IB Lagrangian
method, even with reduced network size.
| [
{
"version": "v1",
"created": "Wed, 11 Dec 2024 09:17:45 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Mar 2025 11:42:25 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Yang",
"Hanzhe",
""
],
[
"Wu",
"Youlong",
""
],
[
"Wen",
"Dingzhu",
""
],
[
"Zhou",
"Yong",
""
],
[
"Shi",
"Yuanming",
""
]
] |
2412.08534 | Dong Chen | Dong Chen, Alice Dethise, Istemi Ekin Akkus, Ivica Rimac, Klaus
Satzke, Antti Koskela, Marco Canini, Wei Wang, Ruichuan Chen | Protecting Confidentiality, Privacy and Integrity in Collaborative
Learning | null | null | null | null | cs.DC cs.CR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A collaboration between dataset owners and model owners is needed to
facilitate effective machine learning (ML) training. During this collaboration,
however, dataset owners and model owners want to protect the confidentiality of
their respective assets (i.e., datasets, models and training code), with the
dataset owners also caring about the privacy of individual users whose data is
in their datasets. Existing solutions either provide limited confidentiality
for models and training code, or suffer from privacy issues due to collusion.
We present Citadel++, a collaborative ML training system designed to
simultaneously protect the confidentiality of datasets, models and training
code as well as the privacy of individual users. Citadel++ enhances
differential privacy mechanisms to safeguard the privacy of individual user
data while maintaining model utility. By employing Virtual Machine-level
Trusted Execution Environments (TEEs) as well as the improved sandboxing and
integrity mechanisms through OS-level techniques, Citadel++ effectively
preserves the confidentiality of datasets, models and training code, and
enforces our privacy mechanisms even when the models and training code have
been maliciously designed. Our experiments show that Citadel++ provides model
utility and performance while adhering to the confidentiality and privacy
requirements of dataset owners and model owners, outperforming the
state-of-the-art privacy-preserving training systems by up to 543x on CPU and
113x on GPU TEEs.
| [
{
"version": "v1",
"created": "Wed, 11 Dec 2024 16:48:18 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Apr 2025 11:38:29 GMT"
}
] | 2025-04-18T00:00:00 | [
[
"Chen",
"Dong",
""
],
[
"Dethise",
"Alice",
""
],
[
"Akkus",
"Istemi Ekin",
""
],
[
"Rimac",
"Ivica",
""
],
[
"Satzke",
"Klaus",
""
],
[
"Koskela",
"Antti",
""
],
[
"Canini",
"Marco",
""
],
[
"Wang",
"Wei",
""
],
[
"Chen",
"Ruichuan",
""
]
] |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.