id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 6
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-04-25 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.11301 | Wenjia Jiang | Yangyang Zhuang and Wenjia Jiang and Jiayu Zhang and Ze Yang and Joey
Tianyi Zhou and Chi Zhang | Learning to Be A Doctor: Searching for Effective Medical Agent
Architectures | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Model (LLM)-based agents have demonstrated strong capabilities
across a wide range of tasks, and their application in the medical domain holds
particular promise due to the demand for high generalizability and reliance on
interdisciplinary knowledge. However, existing medical agent systems often rely
on static, manually crafted workflows that lack the flexibility to accommodate
diverse diagnostic requirements and adapt to emerging clinical scenarios.
Motivated by the success of automated machine learning (AutoML), this paper
introduces a novel framework for the automated design of medical agent
architectures. Specifically, we define a hierarchical and expressive agent
search space that enables dynamic workflow adaptation through structured
modifications at the node, structural, and framework levels. Our framework
conceptualizes medical agents as graph-based architectures composed of diverse,
functional node types and supports iterative self-improvement guided by
diagnostic feedback. Experimental results on skin disease diagnosis tasks
demonstrate that the proposed method effectively evolves workflow structures
and significantly enhances diagnostic accuracy over time. This work represents
the first fully automated framework for medical agent architecture design and
offers a scalable, adaptable foundation for deploying intelligent agents in
real-world clinical environments.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:44:21 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Zhuang",
"Yangyang",
""
],
[
"Jiang",
"Wenjia",
""
],
[
"Zhang",
"Jiayu",
""
],
[
"Yang",
"Ze",
""
],
[
"Zhou",
"Joey Tianyi",
""
],
[
"Zhang",
"Chi",
""
]
] |
2504.11302 | Hari Nathan | Hari Nathan | Limits of Discrete Energy of Families of Increasing Sets | null | null | null | null | math.CA cs.LG math.MG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Hausdorff dimension of a set can be detected using the Riesz energy.
Here, we consider situations where a sequence of points, $\{x_n\}$, ``fills
in'' a set $E \subset \mathbb{R}^d$ in an appropriate sense and investigate the
degree to which the discrete analog to the Riesz energy of these sets can be
used to bound the Hausdorff dimension of $E$. We also discuss applications to
data science and Erd\H{o}s/Falconer type problems.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:45:14 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Nathan",
"Hari",
""
]
] |
2504.11303 | Krisztian Palotas | Ivan Abilio, Kriszti\'an Palot\'as | Energy-resolved tip-orbital fingerprint in scanning tunneling
spectroscopy based on the revised Chen's derivative rule | 12 pages, 5 figures, 4 tables | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The revised Chen's derivative rule for electron tunneling is implemented to
enable computationally efficient first-principles-based calculations of the
differential conductance dI/dV for scanning tunneling spectroscopy (STS)
simulations. The probing tip is included through a single tip apex atom, and
its electronic structure can be modeled as a linear combination of electron
orbitals of various symmetries, or can be directly transferred from
first-principles electronic structure calculations. By taking pristine and
boron- or nitrogen-doped graphene sheets as sample surfaces, the reliability of
our implementation is demonstrated by comparing its results to those obtained
by the Tersoff-Hamann and Bardeen's electron tunneling models. It is
highlighted that the energy-resolved direct and interference contributions to
dI/dV arising from the tip's electron orbitals result in a fingerprint of the
particular combined surface-tip system. The significant difference between the
electron acceptor boron and donor nitrogen dopants in graphene is reflected in
their dI/dV fingerprints. The presented theoretical method allows for an
unprecedented physical understanding of the electron tunneling process in terms
of tip-orbital-resolved energy-dependent dI/dV maps, that is anticipated to be
extremely useful for investigating the local electronic properties of novel
material surfaces in the future.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:45:41 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Abilio",
"Ivan",
""
],
[
"Palotás",
"Krisztián",
""
]
] |
2504.11304 | Carlos Soto | Aditya Kulkarni and Carlos Soto | Differentially Private Geodesic and Linear Regression | 16 pages, 7 figures | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | In statistical applications it has become increasingly common to encounter
data structures that live on non-linear spaces such as manifolds. Classical
linear regression, one of the most fundamental methodologies of statistical
learning, captures the relationship between an independent variable and a
response variable which both are assumed to live in Euclidean space. Thus,
geodesic regression emerged as an extension where the response variable lives
on a Riemannian manifold. The parameters of geodesic regression, as with linear
regression, capture the relationship of sensitive data and hence one should
consider the privacy protection practices of said parameters. We consider
releasing Differentially Private (DP) parameters of geodesic regression via the
K-Norm Gradient (KNG) mechanism for Riemannian manifolds. We derive theoretical
bounds for the sensitivity of the parameters showing they are tied to their
respective Jacobi fields and hence the curvature of the space. This
corroborates recent findings of differential privacy for the Fr\'echet mean. We
demonstrate the efficacy of our methodology on the sphere,
$\mbS^2\subset\mbR^3$ and, since it is general to Riemannian manifolds, the
manifold of Euclidean space which simplifies geodesic regression to a case of
linear regression. Our methodology is general to any Riemannian manifold and
thus it is suitable for data in domains such as medical imaging and computer
vision.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:45:48 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Kulkarni",
"Aditya",
""
],
[
"Soto",
"Carlos",
""
]
] |
2504.11305 | Yuhan Liu | Jincheng Kang and Yi Cen and Yigang Cen and Ke Wang and Yuhan Liu | CFIS-YOLO: A Lightweight Multi-Scale Fusion Network for Edge-Deployable
Wood Defect Detection | 10 pages, 11 figures | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Wood defect detection is critical for ensuring quality control in the wood
processing industry. However, current industrial applications face two major
challenges: traditional methods are costly, subjective, and labor-intensive,
while mainstream deep learning models often struggle to balance detection
accuracy and computational efficiency for edge deployment. To address these
issues, this study proposes CFIS-YOLO, a lightweight object detection model
optimized for edge devices. The model introduces an enhanced C2f structure, a
dynamic feature recombination module, and a novel loss function that
incorporates auxiliary bounding boxes and angular constraints. These
innovations improve multi-scale feature fusion and small object localization
while significantly reducing computational overhead. Evaluated on a public wood
defect dataset, CFIS-YOLO achieves a mean Average Precision ([email protected]) of
77.5\%, outperforming the baseline YOLOv10s by 4 percentage points. On SOPHON
BM1684X edge devices, CFIS-YOLO delivers 135 FPS, reduces power consumption to
17.3\% of the original implementation, and incurs only a 0.5 percentage point
drop in mAP. These results demonstrate that CFIS-YOLO is a practical and
effective solution for real-world wood defect detection in resource-constrained
environments.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:45:59 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Kang",
"Jincheng",
""
],
[
"Cen",
"Yi",
""
],
[
"Cen",
"Yigang",
""
],
[
"Wang",
"Ke",
""
],
[
"Liu",
"Yuhan",
""
]
] |
2504.11306 | Trinnhallen Brisley | Trinnhallen Brisley, Aryan Gandhi, Joseph Magen | Context-Aware Palmprint Recognition via a Relative Similarity Metric | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We propose a new approach to matching mechanism for palmprint recognition by
introducing a Relative Similarity Metric (RSM) that enhances the robustness and
discriminability of existing matching frameworks. While conventional systems
rely on direct pairwise similarity measures, such as cosine or Euclidean
distances, these metrics fail to capture how a pairwise similarity compares
within the context of the entire dataset. Our method addresses this by
evaluating the relative consistency of similarity scores across up to all
identities, allowing for better suppression of false positives and negatives.
Applied atop the CCNet architecture, our method achieves a new state-of-the-art
0.000036% Equal Error Rate (EER) on the Tongji dataset, outperforming previous
methods and demonstrating the efficacy of incorporating relational structure
into the palmprint matching process.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:46:17 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Brisley",
"Trinnhallen",
""
],
[
"Gandhi",
"Aryan",
""
],
[
"Magen",
"Joseph",
""
]
] |
2504.11307 | Sonia Laguna | Sonia Laguna, Lin Zhang, Can Deniz Bezek, Monika Farkas, Dieter
Schweizer, Rahel A. Kubik-Huch, Orcun Goksel | Uncertainty Estimation for Trust Attribution to Speed-of-Sound
Reconstruction with Variational Networks | Published at the International Journal of Computer Assisted Radiology
and Surgery. Presented at the 16th International Conference on Information
Processing in Computer-Assisted Interventions 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Speed-of-sound (SoS) is a biomechanical characteristic of tissue, and its
imaging can provide a promising biomarker for diagnosis. Reconstructing SoS
images from ultrasound acquisitions can be cast as a limited-angle
computed-tomography problem, with Variational Networks being a promising
model-based deep learning solution. Some acquired data frames may, however, get
corrupted by noise due to, e.g., motion, lack of contact, and acoustic shadows,
which in turn negatively affects the resulting SoS reconstructions. We propose
to use the uncertainty in SoS reconstructions to attribute trust to each
individual acquired frame. Given multiple acquisitions, we then use an
uncertainty based automatic selection among these retrospectively, to improve
diagnostic decisions. We investigate uncertainty estimation based on Monte
Carlo Dropout and Bayesian Variational Inference. We assess our automatic frame
selection method for differential diagnosis of breast cancer, distinguishing
between benign fibroadenoma and malignant carcinoma. We evaluate 21 lesions
classified as BI-RADS~4, which represents suspicious cases for probable
malignancy. The most trustworthy frame among four acquisitions of each lesion
was identified using uncertainty based criteria. Selecting a frame informed by
uncertainty achieved an area under curve of 76% and 80% for Monte Carlo Dropout
and Bayesian Variational Inference, respectively, superior to any
uncertainty-uninformed baselines with the best one achieving 64%. A novel use
of uncertainty estimation is proposed for selecting one of multiple data
acquisitions for further processing and decision making.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:48:51 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Laguna",
"Sonia",
""
],
[
"Zhang",
"Lin",
""
],
[
"Bezek",
"Can Deniz",
""
],
[
"Farkas",
"Monika",
""
],
[
"Schweizer",
"Dieter",
""
],
[
"Kubik-Huch",
"Rahel A.",
""
],
[
"Goksel",
"Orcun",
""
]
] |
2504.11308 | Udit Kumar Tyagi | Udit K. Tyagi, Sandeep Haridasu, Soumen Basak | Constraints on Generalized Gravity-Thermodynamic Cosmology from DESI DR2 | 13 pages, 4 figures, comments are welcome | null | null | null | astro-ph.CO gr-qc hep-ph | http://creativecommons.org/licenses/by/4.0/ | We explore the cosmological implications of generalized entropic models
within the framework of Gravity-Thermodynamics (GT) approaches. These models,
characterized by three or four additional free parameters, are designed to
capture deviations from the standard Bekenstein-Hawking entropy and can
reproduce well-known entropic formulations, including Tsallis, R\'enyi,
Sharma-Mittal, Barrow, Kaniadakis, and Loop Quantum Gravity entropies in
various analytical limits. We implement the corresponding cosmological models
using a fully numerical GT approach to constrain the model parameters and to
study the evolution of the dark energy equation of state as a function of the
scale factor. Our Bayesian analysis, which incorporates the Pantheon+ and DESy5
supernovae data alongside the recently released DESI-DR2/DR1 Baryon Acoustic
Oscillation (BAO) measurements, shows that the data favor the standard
Bekenstein-Hawking entropy, leading to a $\Lambda$CDM-like late-time behavior.
In this context, the three-parameter ($\mathcal{S}_3$) entropic model appears
to be sufficient to capture the observed dark energy phenomenology.
Furthermore, a direct comparison of the Bayesian evidence indicates that the
three-parameter model is preferred over the four-parameter ($\mathcal{S}_4$)
variant by a factor of $\Delta\log\mathcal{B} \sim -6$, while the GT approach
as a whole is significantly disfavored relative to the $\Lambda$CDM model with
at least $\Delta\log\mathcal{B} \sim -8$ ($\mathcal{S}_3$) to
$\Delta\log\mathcal{B} \sim -13$ ($\mathcal{S}_4$), when using the DESy5 and
DESI-DR2 datasets.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:49:10 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Tyagi",
"Udit K.",
""
],
[
"Haridasu",
"Sandeep",
""
],
[
"Basak",
"Soumen",
""
]
] |
2504.11309 | Shangchao Yang | Hongbo Li, Shangchao Yang, Ruiyang Xia, Lin Yuan, Xinbo Gao | Big Brother is Watching: Proactive Deepfake Detection via Learnable
Hidden Face | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As deepfake technologies continue to advance, passive detection methods
struggle to generalize with various forgery manipulations and datasets.
Proactive defense techniques have been actively studied with the primary aim of
preventing deepfake operation effectively working. In this paper, we aim to
bridge the gap between passive detection and proactive defense, and seek to
solve the detection problem utilizing a proactive methodology. Inspired by
several watermarking-based forensic methods, we explore a novel detection
framework based on the concept of ``hiding a learnable face within a face''.
Specifically, relying on a semi-fragile invertible steganography network, a
secret template image is embedded into a host image imperceptibly, acting as an
indicator monitoring for any malicious image forgery when being restored by the
inverse steganography process. Instead of being manually specified, the secret
template is optimized during training to resemble a neutral facial appearance,
just like a ``big brother'' hidden in the image to be protected. By
incorporating a self-blending mechanism and robustness learning strategy with a
simulative transmission channel, a robust detector is built to accurately
distinguish if the steganographic image is maliciously tampered or benignly
processed. Finally, extensive experiments conducted on multiple datasets
demonstrate the superiority of the proposed approach over competing passive and
proactive detection methods.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:50:54 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Li",
"Hongbo",
""
],
[
"Yang",
"Shangchao",
""
],
[
"Xia",
"Ruiyang",
""
],
[
"Yuan",
"Lin",
""
],
[
"Gao",
"Xinbo",
""
]
] |
2504.11310 | Qingrui Zhang | Dayong Liu and Qingrui Zhang and Zeyang Meng | Intelligent driving vehicle front multi-target tracking and detection
based on YOLOv5 and point cloud 3D projection | in Chinese language | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In multi-target tracking and detection tasks, it is necessary to continuously
track multiple targets, such as vehicles, pedestrians, etc. To achieve this
goal, the system must be able to continuously acquire and process image frames
containing these targets. These consecutive frame images enable the algorithm
to update the position and state of the target in real-time in each frame of
the image. How to accurately associate the detected target with the target in
the previous or next frame to form a stable trajectory is a complex problem.
Therefore, a multi object tracking and detection method for intelligent driving
vehicles based on YOLOv5 and point cloud 3D projection is proposed. Using
Retinex algorithm to enhance the image of the environment in front of the
vehicle, remove light interference in the image, and build an intelligent
detection model based on YOLOv5 network structure. The enhanced image is input
into the model, and multiple targets in front of the vehicle are identified
through feature extraction and target localization. By combining point cloud 3D
projection technology, the correlation between the position changes of adjacent
frame images in the projection coordinate system can be inferred. By
sequentially projecting the multi-target recognition results of multiple
consecutive frame images into the 3D laser point cloud environment, effective
tracking of the motion trajectories of all targets in front of the vehicle can
be achieved. The experimental results show that the application of this method
for intelligent driving vehicle front multi-target tracking and detection
yields a MOTA (Tracking Accuracy) value greater than 30, demonstrating its
superior tracking and detection performance.
| [
{
"version": "v1",
"created": "Sun, 13 Apr 2025 17:50:02 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Liu",
"Dayong",
""
],
[
"Zhang",
"Qingrui",
""
],
[
"Meng",
"Zeyang",
""
]
] |
2504.11311 | Elena Castellani | Elena Castellani | From Symmetry to Supersymmetry to Supergravity | Invited contribution to the volume "Half a Century of Supergravity",
A. Ceresole and G. Dall'Agata (eds.) | null | null | null | physics.hist-ph hep-th | http://creativecommons.org/licenses/by/4.0/ | The theoretical developments that led to supersymmetry, first global and then
local, over roughly six years (1970-1976) arose from a convergence of physical
insights and mathematical methods stemming from diverse, and sometimes
independent, research directions. This contribution aims to illustrate the
interplay of ideas, methods, and motivations that informed the entire process
and concludes with a reflection on the scientific modality of these
developments.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:51:54 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Castellani",
"Elena",
""
]
] |
2504.11312 | Nathan Wagner | Bingyang Hu, Ji Li, and Nathan A. Wagner | Boundedness and compactness of Bergman projection commutators in
two-weight setting | 26 pages with references | null | null | null | math.CA math.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The goal of this paper is to study the boundedness and compactness of the
Bergman projection commutators in two weighted settings via the weighted BMO
and VMO spaces, respectively. The novelty of our work lies in the distinct
treatment of the symbol b in the commutator, depending on whether it is
analytic or not, which turns out to be quite different. In particular, we show
that an additional weight condition due to Aleman, Pott, and Reguera is
necessary to study the commutators when b is not analytic, while it can be
relaxed when b is analytic. In the analytic setting, we completely characterize
boundedness and compactness, while in the non-analytic setting, we provide a
sufficient condition which generalizes the Euclidean case and is also necessary
in many cases of interest. Our work initiates a study of the commutators acting
on complex function spaces with different symbols.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:52:36 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Hu",
"Bingyang",
""
],
[
"Li",
"Ji",
""
],
[
"Wagner",
"Nathan A.",
""
]
] |
2504.11313 | Yuting Shen | Yuting Shen and Jingwei Liang | A Composed Alternating Relaxed Projection Algorithm for Feasibility
Problem | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | Feasibility problem aims to find a common point of two or more closed
(convex) sets whose intersection is nonempty. In the literature, projection
based algorithms are widely adopted to solve the problem, such as the method of
alternating projection (MAP), and Douglas--Rachford splitting method (DR). The
performance of the methods are governed by the geometric properties of the
underlying sets. For example, the fixed-point sequence of the Douglas--Rachford
splitting method exhibits a spiraling behavior when solving the feasibility
problem of two subspaces, leading to a slow convergence speed and slower than
MAP. However, when the problem at hand is non-polyhedral, DR can demonstrate
significant faster performance. Motivated by the behaviors of the DR method, in
this paper we propose a new algorithm for solving convex feasibility problems.
The method is designed based on DR method by further incorporating a
composition of projection and reflection. A non-stationary version of the
method is also designed, aiming to achieve faster practical performance.
Theoretical guarantees of the proposed schemes are provided and supported by
numerical experiments.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:53:10 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Shen",
"Yuting",
""
],
[
"Liang",
"Jingwei",
""
]
] |
2504.11314 | Jie Jiang | Xin-Yang Wang and Jie Jiang | Second Law of Black Holes Thermodynamics in Non-Quasi-Static Processes | 154 pages | null | null | null | gr-qc hep-th | http://creativecommons.org/publicdomain/zero/1.0/ | The Bekenstein-Hawking entropy satisfies the generalized second law of black
hole thermodynamics for arbitrary thermodynamic evolution within
Einstein-Maxwell theory. In contrast, the black hole entropy that satisfies the
second law in low-energy effective modified gravity theories related to quantum
gravity is derived solely by considering linear perturbations within
quasi-static processes. Since astrophysical processes are typically
non-quasi-static, deriving a rigorous expression for entropy within the
quasi-static framework is not feasible. By treating the effective quantum
corrections as first-order perturbations, the black hole entropy in
Einstein-Maxwell gravity with generalized quadratic corrections for arbitrary
dynamical processes is derived. This entropy is distinct from the Iyer-Wald and
Dong-Wald entropies and is applicable to non-quasi-static processes.
Furthermore, it is shown that the black hole entropy is consistent with the
generalized covariant entropy boundary. This work establishes a framework for
examining the generalized second law of black hole thermodynamics in non-static
processes within modified gravity theories.
| [
{
"version": "v1",
"created": "Mon, 14 Apr 2025 11:27:19 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Wang",
"Xin-Yang",
""
],
[
"Jiang",
"Jie",
""
]
] |
2504.11315 | Walter Krawec | Trevor N. Thomas and Walter O. Krawec | New Key Rate Bound for High-Dimensional BB84 with Multiple Basis
Measurements | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we derive a new bound on the secret key-rate of the High
Dimensional BB84 protocol operating with multiple mutually unbiased bases
(MUBs). To our knowledge, our proof is the first for this protocol that is both
general (in that it can handle arbitrary, asymmetric channels), and also the
first that derives a bound on the quantum min entropy for general attacks,
without relying on post selection techniques or the asymptotic equipartition
property. Because of this, our new result shows that far more optimistic
key-rates are possible for a low number of signals, even in general channels.
Furthermore, our proof methods may be broadly applicable to other protocols
relying on multiple measurement bases and we prove several technical lemmas
that may have independent interest. We evaluate our new bound and compare to
prior work, showing that higher key-rates are possible in several operating
scenarios. We also show some interesting behavior of the protocol when faced
with asymmetric noise.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:55:24 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Thomas",
"Trevor N.",
""
],
[
"Krawec",
"Walter O.",
""
]
] |
2504.11316 | Soumya Bhattacharyya | Soumya Bhattacharyya and Meng-Ru Wu and Zewei Xiong | Role of Matter Inhomogeneity on Fast Flavor Conversion of Supernova
Neutrinos | 7 pages, 3 figures (Submitted to Physical Review Letters) | null | null | null | astro-ph.HE astro-ph.CO astro-ph.SR hep-ph nucl-th | http://creativecommons.org/licenses/by/4.0/ | We investigated the impact of a spatially varying matter potential $\lambda$,
coming from neutrino-electron forward scattering, on the emergence of fast
neutrino flavor conversion (FFC) triggered by the presence of zero crossings in
the angular distribution of the neutrino electron lepton number (ELN). We find
that FFC can be significantly affected as the spatial variation rate of
$\lambda$ increases, and strong spatial variations can completely stabilize
initially unstable systems. Using stability analysis based solely on initial
conditions, we identified for the first time a critical variation rate above
which no FFC occurs even if the flavor instability exists. By analyzing several
representative matter profiles based on an 18 $M_{\odot}$ SN model, we show
that spatially inhomogeneous $\lambda$ can suppress the occurrence of FFC
associated with shallow ELN zero crossings in most of the SN's radial region,
especially during the accretion phase. Our finding highlights the need to
consider the impact of matter inhomogeneity in the development of improved SN
models that aim to include the effect of neutrino flavor conversions.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:56:26 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Bhattacharyya",
"Soumya",
""
],
[
"Wu",
"Meng-Ru",
""
],
[
"Xiong",
"Zewei",
""
]
] |
2504.11317 | Baptiste Debecker | Baptiste Debecker, Lukas Pausch, Jonathan Louvet, Thierry Bastin, John
Martin, and Fran\c{c}ois Damanet | The role of non-Markovian dissipation in quantum phase transitions:
tricriticality, spin squeezing, and directional symmetry breaking | 16 pages and 7 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding how to control phase transitions in quantum systems is at the
forefront of research for the development of new quantum materials and
technologies. Here, we study how the coupling of a quantum system to a
non-Markovian environment, i.e., an environment with a frequency-dependent
spectral density inducing memory effects, can be used to generate and reshape
phase transitions and squeezing in matter phases. Focusing on a
Lipkin-Meshkov-Glick model, we demonstrate that non-Markovian dissipation can
be leveraged to engineer tricriticality via the fusion of
$2^{\mathrm{nd}}$-order and $1^{\mathrm{st}}$-order critical points. We
identify phases that arise from different ways of breaking the single weak
symmetry of our model, which led us to introduce the concept of
\textit{directional spontaneous symmetry breaking} (DSSB) as a general
framework to understand this phenomenon. We show that signatures of DSSB can be
seen in the emergence of spin squeezing along different directions, and that
the latter is controllable via non-Markovian effects, opening up possibilities
for applications in quantum metrology. Finally, we propose an experimental
implementation of our non-Markovian model in cavity QED. Our work features
non-Markovianity as a resource for controlling phase transitions in general
systems, and highlights shortcomings of the Markovian limit in this context.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:58:15 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Debecker",
"Baptiste",
""
],
[
"Pausch",
"Lukas",
""
],
[
"Louvet",
"Jonathan",
""
],
[
"Bastin",
"Thierry",
""
],
[
"Martin",
"John",
""
],
[
"Damanet",
"François",
""
]
] |
2504.11318 | Vishnu Iyer | Vishnu Iyer | Mildly-Interacting Fermionic Unitaries are Efficiently Learnable | 30 pages, 4 figures | null | null | null | quant-ph cs.DS cs.LG | http://creativecommons.org/licenses/by/4.0/ | Recent work has shown that one can efficiently learn fermionic Gaussian
unitaries, also commonly known as nearest-neighbor matchcircuits or
non-interacting fermionic unitaries. However, one could ask a similar question
about unitaries that are near Gaussian: for example, unitaries prepared with a
small number of non-Gaussian circuit elements. These operators find
significance in quantum chemistry and many-body physics, yet no algorithm
exists to learn them.
We give the first such result by devising an algorithm which makes queries to
a $n$-mode fermionic unitary $U$ prepared by at most $O(t)$ non-Gaussian gates
and returns a circuit approximating $U$ to diamond distance $\varepsilon$ in
time $\textrm{poly}(n,2^t,1/\varepsilon)$. This resolves a central open
question of Mele and Herasymenko under the strongest distance metric. In fact,
our algorithm is much more general: we define a property of unitary Gaussianity
known as unitary Gaussian dimension and show that our algorithm can learn
$n$-mode unitaries of Gaussian dimension at least $2n - O(t)$ in time
$\textrm{poly}(n,2^t,1/\varepsilon)$. Indeed, this class subsumes unitaries
prepared by at most $O(t)$ non-Gaussian gates but also includes several
unitaries that require up to $2^{O(t)}$ non-Gaussian gates to construct.
In addition, we give a $\textrm{poly}(n,1/\varepsilon)$-time algorithm to
distinguish whether an $n$-mode unitary is of Gaussian dimension at least $k$
or $\varepsilon$-far from all such unitaries in Frobenius distance, promised
that one is the case. Along the way, we prove structural results about
near-Gaussian fermionic unitaries that are likely to be of independent
interest.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:59:32 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Iyer",
"Vishnu",
""
]
] |
2504.11319 | Yiqing Zhou | Yiqing Zhou, Karsten Naert, Dirk Nuyens | Sensitivity Analysis of State Space Models for Scrap Composition
Estimation in EAF and BOF | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This study develops and analyzes linear and nonlinear state space models for
estimating the elemental composition of scrap steel used in steelmaking, with
applications to Electric Arc Furnace (EAF) and Basic Oxygen Furnace (BOF)
processes. The models incorporate mass balance equations and are fitted using a
modified Kalman filter for linear cases and the Unscented Kalman Filter (UKF)
for nonlinear cases. Using Cu and Cr as representative elements, we assess the
sensitivity of model predictions to measurement noise in key process variables,
including steel mass, steel composition, scrap input mass, slag mass, and iron
oxide fraction in slag. Results show that the models are robust to moderate
noise levels in most variables, particularly when errors are below $10\%$.
However, accuracy significantly deteriorates with noise in slag mass
estimation. These findings highlight the practical feasibility and limitations
of applying state space models for real-time scrap composition estimation in
industrial settings.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 15:59:42 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Zhou",
"Yiqing",
""
],
[
"Naert",
"Karsten",
""
],
[
"Nuyens",
"Dirk",
""
]
] |
2504.11320 | Ruicheng Ao | Ruicheng Ao, Gan Luo, David Simchi-Levi, Xinshang Wang | Optimizing LLM Inference: Fluid-Guided Online Scheduling with Memory
Constraints | 42 pages, 18 figures | null | null | null | cs.LG cs.AI cs.DC math.OC stat.ML | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Large Language Models (LLMs) are indispensable in today's applications, but
their inference procedure -- generating responses by processing text in
segments and using a memory-heavy Key-Value (KV) cache -- demands significant
computational resources, particularly under memory constraints. This paper
formulates LLM inference optimization as a multi-stage online scheduling
problem where sequential prompt arrivals and KV cache growth render
conventional scheduling ineffective. We develop a fluid dynamics approximation
to provide a tractable benchmark that guides algorithm design. Building on
this, we propose the Waiting for Accumulated Inference Threshold (WAIT)
algorithm, which uses multiple thresholds to schedule incoming prompts
optimally when output lengths are known, and extend it to Nested WAIT for cases
with unknown output lengths. Theoretical analysis shows that both algorithms
achieve near-optimal performance against the fluid benchmark in heavy traffic
conditions, balancing throughput, latency, and Time to First Token (TTFT).
Experiments with the Llama-7B model on an A100 GPU using both synthetic and
real-world datasets demonstrate improved throughput and reduced latency
relative to established baselines like vLLM and Sarathi. This work bridges
operations research and machine learning, offering a rigorous framework for the
efficient deployment of LLMs under memory constraints.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:00:21 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Ao",
"Ruicheng",
""
],
[
"Luo",
"Gan",
""
],
[
"Simchi-Levi",
"David",
""
],
[
"Wang",
"Xinshang",
""
]
] |
2504.11321 | Pedro Henrique da Costa Avelar | Pedro Henrique da Costa Avelar, Min Wu, Sophia Tsoka | Subset-Contrastive Multi-Omics Network Embedding | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Motivation: Network-based analyses of omics data are widely used, and while
many of these methods have been adapted to single-cell scenarios, they often
remain memory- and space-intensive. As a result, they are better suited to
batch data or smaller datasets. Furthermore, the application of network-based
methods in multi-omics often relies on similarity-based networks, which lack
structurally-discrete topologies. This limitation may reduce the effectiveness
of graph-based methods that were initially designed for topologies with better
defined structures. Results: We propose Subset-Contrastive multi-Omics Network
Embedding (SCONE), a method that employs contrastive learning techniques on
large datasets through a scalable subgraph contrastive approach. By exploiting
the pairwise similarity basis of many network-based omics methods, we
transformed this characteristic into a strength, developing an approach that
aims to achieve scalable and effective analysis. Our method demonstrates
synergistic omics integration for cell type clustering in single-cell data.
Additionally, we evaluate its performance in a bulk multi-omics integration
scenario, where SCONE performs comparable to the state-of-the-art despite
utilising limited views of the original data. We anticipate that our findings
will motivate further research into the use of subset contrastive methods for
omics data.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:01:39 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Avelar",
"Pedro Henrique da Costa",
""
],
[
"Wu",
"Min",
""
],
[
"Tsoka",
"Sophia",
""
]
] |
2504.11322 | Senwen Deng | Senwen Deng, Stanislav Babak, Sylvain Marsat | Fast detection and reconstruction of merging Massive Black Hole Binary
signals | 13 pages, 8 figures | null | null | null | gr-qc astro-ph.CO astro-ph.IM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Laser Interferometer Space Antenna (LISA) will detect gravitational waves
from the population of merging massive black holes binaries (MBHBs) throughout
the Universe. The LISA data stream will feature many superposed signals from
different astrophysical sources, requiring a global fit procedure. Most of the
MBHB signals will be loud enough to be detected days or even weeks before the
merger; and for those sources LISA will be able to predict the time of the
merger well in advance of the coalescence, as well as an approximate position
in the sky. In this paper, we present a fast detection and signal
reconstruction scheme for massive black hole binaries in the LISA observation
band. We propose: (i) a detection scheme for MBHB mergers allowing a first
subtraction of these signals for the purpose of a global fit, and (ii) an
efficient early detection scheme providing a time-of-merger estimate for a
pre-merger signal, that will allow to trigger a protection period, placing LISA
in ``do not disturb'' mode and enabling more detailed analysis that will
facilitate multi-messenger observations. We highlight the effect of confusion
of several overlapping in time MBHB signals in the pre-merger detection.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:01:55 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Deng",
"Senwen",
""
],
[
"Babak",
"Stanislav",
""
],
[
"Marsat",
"Sylvain",
""
]
] |
2504.11323 | C. J. A. P. Martins | R. Heilemann, M. C. Rosa, J. R. C. C. C. Correia, C. J. A. P. Martins | Domain wall evolution beyond quartic potentials: The Sine-Gordon and
Christ-Lee potentials | 29 pages, 25 figures, 8 tables | null | null | null | hep-ph astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Domain walls are the simplest type of topological defects formed at
cosmological phase transitions, and one of the most constrained. Their studies
typically assume a quartic double well potential, but this model is not fully
representative of the range of known or plausible particle physics models. Here
we study the cosmological evolution of domain walls in two other classes of
potentials. The Sine-Gordon potential allows several types of walls,
interpolating between different pairs of minima (which demands specific
numerical algorithms to separately measure the relevant properties of each
type). The Christ-Lee potential parametrically interpolates between sextic and
quartic behavior. We use multiple sets of simulations in two and three spatial
dimensions, for various cosmological epochs and under various choices of
initial conditions, to discuss the scaling properties of these networks. In the
Sine-Gordon case, we identify and quantify deviations from the usual scaling
behavior. In the Christ-Lee case, we discuss conditions under which walls form
(or not), and quantify how these outcomes depend on parameters such as the
energy difference between the false and true vacua and the expansion rate of
the Universe. Various biased initial conditions are also addressed in
appendices. Finally, we briefly comment on the possible cosmological
implications of our results.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:02:35 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Heilemann",
"R.",
""
],
[
"Rosa",
"M. C.",
""
],
[
"Correia",
"J. R. C. C. C.",
""
],
[
"Martins",
"C. J. A. P.",
""
]
] |
2504.11324 | Ander Sim\'on Est\'evez | L. Larizgoitia, A. Sim\'on, E. Oblak, C. Echeverria, P. Dietz, A.
Castillo, L. Donneger, J.J. G\'omez-Cadenas, F. Monrabal | The Gaseous Prototype (GaP): a GanESS demonstrator | 20 pages, 14 figures | null | null | null | physics.ins-det hep-ex nucl-ex | http://creativecommons.org/licenses/by/4.0/ | The GanESS experiment will exploit the high-pressure noble gas time
projection chamber technology to detect coherent elastic neutrino-nucleus
scattering (CE$\nu$NS) at the European Spallation Source (ESS). The detector,
able to operate at pressures up to 50 bar with different noble gases (Xe, Ar
and Kr), will employ electroluminescence to amplify the ionization signal with
the objective of reaching a threshold as low as 1-2 e$^-$, equivalent to $<$
100 eV$_{\text{ee}}$. The Gaseous Prototype (GaP) has been built to
characterize the technique at the few-keV energy regime and to understand
various aspects related to the technology. Concretely, it will be used to
measure the quenching factor of the different mediums as well as to
characterize the electroluminescence yield and detection threshold under
different operational conditions. The present paper describes the Gaseous
Prototype and its first results operating with gaseous argon at moderate
pressures (up to 10 bar). A potential detection threshold lower than 2.9 keV
has been observed following operation with a $^{55}$Fe calibration source.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:02:38 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Larizgoitia",
"L.",
""
],
[
"Simón",
"A.",
""
],
[
"Oblak",
"E.",
""
],
[
"Echeverria",
"C.",
""
],
[
"Dietz",
"P.",
""
],
[
"Castillo",
"A.",
""
],
[
"Donneger",
"L.",
""
],
[
"Gómez-Cadenas",
"J. J.",
""
],
[
"Monrabal",
"F.",
""
]
] |
2504.11325 | Jianwei Miao | Yakun Yuan, Saman Moniri, Yao Yang, Jihan Zhou, Andrew Yuan, Dennis S.
Kim, Yongsoo Yang, Chenyang Li, Wei Chen, Peter Ercius, Jianwei Miao | Crystal nucleation and growth in high-entropy alloys revealed by atomic
electron tomography | null | null | null | null | cond-mat.mtrl-sci cond-mat.dis-nn cond-mat.soft | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | High-entropy alloys (HEAs) balance mixing entropy and intermetallic phase
formation enthalpy, creating a vast compositional space for structural and
functional materials (1-6). They exhibit exceptional strength-ductility
trade-offs in metallurgy (4-10) and near-continuum adsorbate binding energies
in catalysis (11-16). A deep understanding of crystal nucleation and growth in
HEAs is essential for controlling their formation and optimizing their
structural and functional properties. However, atomic-scale nucleation in HEAs
challenges traditional theories based on one or two principal elements (17-23).
The intricate interplay of structural and chemical orders among multiple
principal elements further obscures our understanding of nucleation pathways
(5,24-27). Due to the lack of direct three-dimensional (3D) atomic-scale
observations, previous studies have relied on simulations and indirect
measurements (28-32), leaving HEA nucleation and growth fundamentally elusive.
Here, we advance atomic electron tomography (33,34) to resolve the 3D atomic
structure and chemical composition of 7,662 HEA and 498 medium-entropy alloy
nuclei at different nucleation stages. We observe local structural order that
decreases from core to boundary, correlating with local chemical order. As
nuclei grow, structural order improves. At later stages, most nuclei coalesce
without misorientation, while some form coherent twin boundaries. To explain
these experimental observations, we propose the gradient nucleation pathways
model, in which the nucleation energy barrier progressively increases through
multiple evolving intermediate states. We expect these findings to not only
provide fundamental insights into crystal nucleation and growth in HEAs, but
also offer a general framework for understanding nucleation mechanisms in other
materials.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:02:40 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Yuan",
"Yakun",
""
],
[
"Moniri",
"Saman",
""
],
[
"Yang",
"Yao",
""
],
[
"Zhou",
"Jihan",
""
],
[
"Yuan",
"Andrew",
""
],
[
"Kim",
"Dennis S.",
""
],
[
"Yang",
"Yongsoo",
""
],
[
"Li",
"Chenyang",
""
],
[
"Chen",
"Wei",
""
],
[
"Ercius",
"Peter",
""
],
[
"Miao",
"Jianwei",
""
]
] |
2504.11327 | Patrizio Neff | Patrizio Neff and Nina J. Husemann and Sebastian Holthausen and Franz
Gmeineder and Thomas Blesgen | Rate-form equilibrium for an isotropic Cauchy-elastic formulation: Part
I: modeling | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We derive the rate-form spatial equilibrium system for a nonlinear Cauchy
elastic formulation in isotropic finite-strain elasticity. For a given explicit
Cauchy stress-strain constitutive equation, we determine those properties that
pertain to the appearing fourth-order stiffness tensor. Notably, we show that
this stiffness tensor $\mathbb{H}^{\text{ZJ}}(\sigma)$ acting on the
Zaremba-Jaumann stress rate is uniformly positive definite. We suggest a
mathematical treatment of the ensuing spatial PDE-system which may ultimately
lead to a local existence result, to be presented in part II of this work. As a
preparatory step, we show existence and uniqueness of a subproblem based on
Korn's first inequality and the positive definiteness of this stiffness tensor.
The procedure is not confined to Cauchy elasticity, however in the Cauchy
elastic case, most theoretical statements can be made explicit.
Our development suggests that looking at the rate-form equations of given
Cauchy-elastic models may provide additional insight to the modeling of
nonlinear isotropic elasticity. This especially concerns constitutive
requirements emanating from the rate-formulation, here being reflected by the
positive definiteness of $\mathbb{H}^{\text{ZJ}}(\sigma)$.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:02:59 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Neff",
"Patrizio",
""
],
[
"Husemann",
"Nina J.",
""
],
[
"Holthausen",
"Sebastian",
""
],
[
"Gmeineder",
"Franz",
""
],
[
"Blesgen",
"Thomas",
""
]
] |
2504.11328 | Nicolas Crouzet | N. Crouzet, M. Mueller, B. Sargent, F. Lahuis, D. Kester, G. Yang, I.
Argyriou, D. Gasman, P. J. Kavanagh, A. Labiano, K. Larson, D. R. Law, J.
\'Alvarez-M\'arquez, B. R. Brandl, A. Glasse, P. Patapis, P. R. Roelfsema,
{\L}. Tychoniec, E. F. van Dishoeck, and G. S. Wright | Extended source fringe flats for the JWST MIRI Medium Resolution
Spectrometer | 17 pages, 21 figures, A&A, in production | null | null | null | astro-ph.IM | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The detectors of the JWST Mid-Infrared Instrument (MIRI) Medium Resolution
Spectrometer (MRS) form low-finesse resonating cavities that cause periodic
count rate modulations (fringes) with peak amplitudes of up to 15% for sources
external to MIRI. To detect weak features on a strong continuum and reliably
measure line fluxes and line-flux ratios, fringe correction is crucial. This
paper describes the first of two steps implemented in the JWST Science
Calibration Pipeline, which is the division by a static fringe flat that
removes the bulk of the fringes for extended sources. Fringe flats were derived
by fitting a numerical model to observations of spatially extended sources. The
model includes fringes that originate from two resonating cavities in the
detector substrate (a third fringe component that originates from the dichroic
filters is not included). The model, numerical implementation, and resulting
fringe flats are described, and the efficiency of the calibration was evaluated
for sources of various spatial extents on the detector. Flight fringe flats are
obtained from observations of the planetary nebula NGC 7027. The two fringe
components are well recovered and fitted by the model. The derived parameters
are used to build a fringe flat for each MRS spectral band, except for 1A and
1B due to the low signal-to-noise ratio of NGC 7027 in these bands. When
applied to extended sources, fringe amplitudes are reduced to the sub-percent
level on individual spaxels. For point sources, they are reduced to amplitudes
of 1 to 5% considering individual spaxels and a single dither position, and
decrease to the 1 to 2% level after two-dimensional residual fringe correction.
The fringe flats derived from this work are the reference files currently in
use by the JWST Science Calibration Pipeline. They provide an efficient
calibration for extended sources, and are less efficient for point sources.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:02:59 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Crouzet",
"N.",
""
],
[
"Mueller",
"M.",
""
],
[
"Sargent",
"B.",
""
],
[
"Lahuis",
"F.",
""
],
[
"Kester",
"D.",
""
],
[
"Yang",
"G.",
""
],
[
"Argyriou",
"I.",
""
],
[
"Gasman",
"D.",
""
],
[
"Kavanagh",
"P. J.",
""
],
[
"Labiano",
"A.",
""
],
[
"Larson",
"K.",
""
],
[
"Law",
"D. R.",
""
],
[
"Álvarez-Márquez",
"J.",
""
],
[
"Brandl",
"B. R.",
""
],
[
"Glasse",
"A.",
""
],
[
"Patapis",
"P.",
""
],
[
"Roelfsema",
"P. R.",
""
],
[
"Tychoniec",
"Ł.",
""
],
[
"van Dishoeck",
"E. F.",
""
],
[
"Wright",
"G. S.",
""
]
] |
2504.11329 | Matthew Leighton | Johan du Buisson, Jannik Ehrich, Matthew P. Leighton, Avijit Kundu,
Tushar K. Saha, John Bechhoefer, David A. Sivak | Hunting for Maxwell's Demon in the Wild | 12 pages, 5 figures | null | null | null | cond-mat.stat-mech physics.bio-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The apparent paradox of Maxwell's demon motivated the development of
information thermodynamics and, more recently, engineering advances enabling
the creation of nanoscale information engines. From these advances, it is now
understood that nanoscale machines like the molecular motors within cells can
in principle operate as Maxwell demons. This motivates the question: does
information help power molecular motors? Answering this would seemingly require
simultaneous measurement of all system degrees of freedom, which is generally
intractable in single-molecule experiments. To overcome this limitation, we
derive a statistical estimator to infer both the direction and magnitude of
subsystem heat flows, and thus to determine whether -- and how strongly -- a
motor operates as a Maxwell demon. The estimator uses only trajectory
measurements for a single degree of freedom. We demonstrate the estimator by
applying it to simulations of an experimental realization of an information
engine and a kinesin molecular motor. Our results show that kinesin transitions
to a Maxwell-demon mechanism in the presence of nonequilibrium noise, with a
corresponding increase in velocity consistent with experiments. These findings
suggest that molecular motors may have evolved to leverage active fluctuations
within cells.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:03:10 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Buisson",
"Johan du",
""
],
[
"Ehrich",
"Jannik",
""
],
[
"Leighton",
"Matthew P.",
""
],
[
"Kundu",
"Avijit",
""
],
[
"Saha",
"Tushar K.",
""
],
[
"Bechhoefer",
"John",
""
],
[
"Sivak",
"David A.",
""
]
] |
2504.11330 | Qihang Zhang | Qihang Zhang, Haoyu Yue, Ninghe Liu, Danlin Xu, Renjie Zhou, Liangcai
Cao, George Barbastathis | Decorrelation in Complex Wave Scattering | 4 figures | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Phenomena involving multiple scattering, despite having attracted
considerable attention in physics for decades, continue to generate unexpected
and counterintuitive behaviours prompting further studies. For optical
scattering, the memory effect well predicts fourth order statistics, i.e. the
intensity correlation, as long as the scattering strength and depth are within
certain bounds. The memory effect has found a wide range of applications, where
its limitations also become apparent: for example, in imaging through turbid
media, decorrelation due to multiscattering in thick samples has been shown to
restrict the field of view. However, to our knowledge, no comprehensive
mechanism exists to date that can account for decorrelation precisely. In this
paper, we quantify how the scatterer's own statistics determine such
limitations. We show that the ensemble statistics of the backscattered field
may be decomposed into two terms: one expresses surface scattering, where
statistical distributions of multiscale structure features may be inferred from
our previous works; while the second term originates from the underlying
scattering volume and is diffusive. The new framework agrees well with
experiments, including the prediction of a new quasipower law for fluctuations
induced by the single realization.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:04:02 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Zhang",
"Qihang",
""
],
[
"Yue",
"Haoyu",
""
],
[
"Liu",
"Ninghe",
""
],
[
"Xu",
"Danlin",
""
],
[
"Zhou",
"Renjie",
""
],
[
"Cao",
"Liangcai",
""
],
[
"Barbastathis",
"George",
""
]
] |
2504.11331 | Hao Liu | Hao Liu, Lijun He, Jiaxi Liang, Zhihan Ren, Fan Li | Dependency Structure Augmented Contextual Scoping Framework for
Multimodal Aspect-Based Sentiment Analysis | submitted to ACM MM2025 | null | null | null | cs.CL cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimodal Aspect-Based Sentiment Analysis (MABSA) seeks to extract
fine-grained information from image-text pairs to identify aspect terms and
determine their sentiment polarity. However, existing approaches often fall
short in simultaneously addressing three core challenges: Sentiment Cue
Perception (SCP), Multimodal Information Misalignment (MIM), and Semantic Noise
Elimination (SNE). To overcome these limitations, we propose DASCO
(\textbf{D}ependency Structure \textbf{A}ugmented \textbf{Sco}ping Framework),
a fine-grained scope-oriented framework that enhances aspect-level sentiment
reasoning by leveraging dependency parsing trees. First, we designed a
multi-task pretraining strategy for MABSA on our base model, combining
aspect-oriented enhancement, image-text matching, and aspect-level
sentiment-sensitive cognition. This improved the model's perception of aspect
terms and sentiment cues while achieving effective image-text alignment,
addressing key challenges like SCP and MIM. Furthermore, we incorporate
dependency trees as syntactic branch combining with semantic branch, guiding
the model to selectively attend to critical contextual elements within a
target-specific scope while effectively filtering out irrelevant noise for
addressing SNE problem. Extensive experiments on two benchmark datasets across
three subtasks demonstrate that DASCO achieves state-of-the-art performance in
MABSA, with notable gains in JMASA (+3.1\% F1 and +5.4\% precision on
Twitter2015).
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:05:09 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Liu",
"Hao",
""
],
[
"He",
"Lijun",
""
],
[
"Liang",
"Jiaxi",
""
],
[
"Ren",
"Zhihan",
""
],
[
"Li",
"Fan",
""
]
] |
2504.11332 | Cristian Sierra C.F.Sierra | J. S. Alvarado, R. Martinez, Cristian Sierra | Constraints from muon $g-2$ on a gauged non-universal $U(1)_{X}$ model
with inverse see-saw neutrinos | 9 pages, 6 figures, 3 tables | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | We study the effects on a non-universal $U(1)_{X}$ extension of the Standard
Model given the alternative value obtained by the Budapest-Marseille-Wuppertal
(BMW) group for the anomalous magnetic moment of the muon $g-2$. The model
explains the fermion mass hierarchy through the non-universality of the extra
gauge symmetry and by an additional $\mathbb{Z}_{2}$ discrete symmetry, where
the heaviest fermions acquire their masses from two different scales determined
by two Higgs doublets and one singlet, whereas the lightest fermions obtain
their masses from radiative corrections. From cancellation of chiral anomalies,
the model also includes heavy extra fermions, both charged and neutral. The
latter are right-handed neutrinos that acquire masses via an inverse see-saw
mechanism, reproducing the observed squared mass differences for the active
neutrinos. Using the latest lattice calculation of the leading hadronic vacuum
polarization (HVP) contribution to the muon $g-2$, we compute the dominant
one-loop diagrams mediated by the $W$ and charged Higgs bosons, both with a
heavy Majorana neutrino in the loop, setting bounds for masses of the new
particles. We also provide predictions for observables that can probe our model
in the future such as charged lepton flavor violating searches at Belle II like
$\tau\to \mu\gamma$, $\tau\to e\gamma$ and at MEG II for $\mu\to e\gamma$.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:05:23 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Alvarado",
"J. S.",
""
],
[
"Martinez",
"R.",
""
],
[
"Sierra",
"Cristian",
""
]
] |
2504.11333 | Mohammed Sayyari | Mohammed Sayyari and Nail K. Yamaleev | Implicit dual time-stepping positivity-preserving entropy-stable schemes
for the compressible Navier-Stokes equations | null | null | null | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We generalize the explicit high-order positivity-preserving entropy-stable
spectral collocation schemes developed in [30, 34] for the three-dimensional
(3D) compressible Navier Stokes equations to a time implicit formulation. The
time derivative terms are discretized by using the first- and second-order
implicit backward difference formulas (BDF1 and BDF2) that are well suited for
solving steady-state and time-dependent viscous flows at high Reynolds numbers,
respectively. The nonlinear system of discrete equations at each physical
timestep is solved by using a dual time-stepping technique. The proposed scheme
is provably entropy-stable and positivity-preserving and provides unconditional
stability properties in the physical time. Numerical results demonstrating
accuracy and positivity-preserving properties of the new dual time-stepping
scheme are presented for supersonic viscous flows with strong shock waves and
contact discontinuities.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:07:05 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Sayyari",
"Mohammed",
""
],
[
"Yamaleev",
"Nail K.",
""
]
] |
2504.11335 | Gopichand Bandarupalli | Gopichand Bandarupalli | Code Reborn AI-Driven Legacy Systems Modernization from COBOL to Java | null | null | null | null | cs.SE cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | This study investigates AI-driven modernization of legacy COBOL code into
Java, addressing a critical challenge in aging software systems. Leveraging the
Legacy COBOL 2024 Corpus -- 50,000 COBOL files from public and enterprise
sources -- Java parses the code, AI suggests upgrades, and React visualizes
gains. Achieving 93% accuracy, complexity drops 35% (from 18 to 11.7) and
coupling 33% (from 8 to 5.4), surpassing manual efforts (75%) and rule-based
tools (82%). The approach offers a scalable path to rejuvenate COBOL systems,
vital for industries like banking and insurance.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:07:54 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Bandarupalli",
"Gopichand",
""
]
] |
2504.11337 | Zhihao Xu | Zhihao Xu, Yongqi Tong, Xin Zhang, Jun Zhou, Xiting Wang | REWARD CONSISTENCY: Improving Multi-Objective Alignment from a
Data-Centric Perspective | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Multi-objective preference alignment in language models often encounters a
challenging trade-off: optimizing for one human preference (e.g., helpfulness)
frequently compromises others (e.g., harmlessness) due to the inherent
conflicts between competing objectives. While prior work mainly focuses on
algorithmic solutions, we explore a novel data-driven approach to uncover the
types of data that can effectively mitigate these conflicts. Specifically, we
propose the concept of Reward Consistency (RC), which identifies samples that
align with multiple preference objectives, thereby reducing conflicts during
training. Through gradient-based analysis, we demonstrate that RC-compliant
samples inherently constrain performance degradation during multi-objective
optimization. Building on these insights, we further develop Reward Consistency
Sampling, a framework that automatically constructs preference datasets that
effectively mitigate conflicts during multi-objective alignment. Our generated
data achieves an average improvement of 13.37% in both the harmless rate and
helpfulness win rate when optimizing harmlessness and helpfulness, and can
consistently resolve conflicts in varying multi-objective scenarios.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:09:19 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Xu",
"Zhihao",
""
],
[
"Tong",
"Yongqi",
""
],
[
"Zhang",
"Xin",
""
],
[
"Zhou",
"Jun",
""
],
[
"Wang",
"Xiting",
""
]
] |
2504.11338 | Jerry Lacmou Zeutouo | Alexandre Savi Fayam Mbala Mouen, Jerry Lacmou Zeutouo, Vianney Kengne
Tchendji | Transformer-Based Model for Cold Start Mitigation in FaaS Architecture | null | null | null | null | cs.DC cs.AI | http://creativecommons.org/licenses/by/4.0/ | Serverless architectures, particularly the Function as a Service (FaaS)
model, have become a cornerstone of modern cloud computing due to their ability
to simplify resource management and enhance application deployment agility.
However, a significant challenge remains: the cold start problem. This
phenomenon occurs when an idle FaaS function is invoked, requiring a full
initialization process, which increases latency and degrades user experience.
Existing solutions for cold start mitigation are limited in terms of invocation
pattern generalization and implementation complexity. In this study, we propose
an innovative approach leveraging Transformer models to mitigate the impact of
cold starts in FaaS architectures. Our solution excels in accurately modeling
function initialization delays and optimizing serverless system performance.
Experimental evaluation using a public dataset provided by Azure demonstrates a
significant reduction in cold start times, reaching up to 79\% compared to
conventional methods.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:12:07 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Mouen",
"Alexandre Savi Fayam Mbala",
""
],
[
"Zeutouo",
"Jerry Lacmou",
""
],
[
"Tchendji",
"Vianney Kengne",
""
]
] |
2504.11339 | Marco Feder | Michele Benzi, Marco Feder, Luca Heltai and Federica Mugnaioni | Optimal and Scalable Augmented Lagrangian preconditioners for Fictitious
Domain problems | null | null | null | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present optimal and scalable preconditioning techniques to solve linear
systems of equations with a block two-by-two and three-by-three structure
arising from fictitious domain problems and from finite element discretizations
of immersed boundary methods. In particular, we propose two augmented
Lagrangian-based preconditioners to accelerate the convergence of iterative
solvers for these two classes of linear. We consider two relevant examples to
illustrate the performance of these preconditioners when used in conjunction
with flexible GMRES: the Poisson and the Stokes fictitious domain problems. A
spectral analysis is established for both exact and inexact versions of these
preconditioners. We show the effectiveness of the proposed approach and the
robustness of our preconditioning strategy through extensive numerical tests in
both two and three dimensions.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:12:51 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Benzi",
"Michele",
""
],
[
"Feder",
"Marco",
""
],
[
"Heltai",
"Luca",
""
],
[
"Mugnaioni",
"Federica",
""
]
] |
2504.11340 | Davide Michieletto | Saminathan Ramakrishnan, Auro Varat Patnaik, Guglielmo Grillo, Luca
Tubiana, Davide Michieletto | Organisation and dynamics of individual DNA segments in topologically
complex genomes | null | null | null | null | cond-mat.soft physics.bio-ph | http://creativecommons.org/licenses/by/4.0/ | Capturing the physical organisation and dynamics of genomic regions is one of
the major open challenges in biology. The kinetoplast DNA (kDNA) is a
topologically complex genome, made by thousands of DNA (mini and maxi) circles
interlinked into a two-dimensional Olympic network. The organisation and
dynamics of these DNA circles are poorly understood. In this paper, we show
that dCas9 linked to Quantum Dots can efficiently label different classes of
DNA minicircles in kDNA. We use this method to study the distribution and
dynamics of different classes of DNA minicircles within the network. We
discover that maxicircles display a preference to localise at the periphery of
the network and that they undergo subdiffusive dynamics. From the latter, we
can also quantify the effective network stiffness, confirming previous indirect
estimations via AFM. Our method could be used more generally, to quantify the
location, dynamics and material properties of genomic regions in other complex
genomes, such as that of bacteria, and to study their behaviour in the presence
of DNA-binding proteins.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:12:54 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Ramakrishnan",
"Saminathan",
""
],
[
"Patnaik",
"Auro Varat",
""
],
[
"Grillo",
"Guglielmo",
""
],
[
"Tubiana",
"Luca",
""
],
[
"Michieletto",
"Davide",
""
]
] |
2504.11342 | Roozbeh Hazrat | Tran Quang Do, Roozbeh Hazrat, Tran Giang Nam | Williams' conjecture holds for graphs of Gelfand-Kirillov dimension
three | null | null | null | null | math.RA math.DS math.OA | http://creativecommons.org/licenses/by/4.0/ | A graph of Gelfand-Kirillov dimension three is a connected finite essential
graph such that its Leavitt path algebra has Gelfand-Kirillov dimension three.
We provide number-theoretic criteria for graphs of Gelfand-Kirillov dimension
three to be strong shift equivalent. We then prove that two graphs of
Gelfand-Kirillov dimension three are shift equivalent if and only if they are
strongly shift equivalent, if and only if their corresponding Leavitt path
algebras are graded Morita equivalent, if and only if their graded
$K$-theories, $K^{\text{gr}}_0$, are order-preserving $\mathbb{Z}[x,
x^{-1}]$-module isomorphic. As a consequence, we obtain that the Leavitt path
algebras of graphs of Gelfand-Kirillov dimension three are graded Morita
equivalent if and only if their graph $C^*$-algebras are equivariant Morita
equivalent, and two graphs $E$ and $F$ of Gelfand-Kirillov dimension three are
shift equivalent if and only if the singularity categories
$\text{D}_{\text{sg}}(KE/J_E^2)$ and $\text{D}_{\text{sg}}(KF/J_F^2)$ are
triangulated equivalent.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:14:00 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Do",
"Tran Quang",
""
],
[
"Hazrat",
"Roozbeh",
""
],
[
"Nam",
"Tran Giang",
""
]
] |
2504.11343 | Hanze Dong | Wei Xiong, Jiarui Yao, Yuhui Xu, Bo Pang, Lei Wang, Doyen Sahoo,
Junnan Li, Nan Jiang, Tong Zhang, Caiming Xiong, Hanze Dong | A Minimalist Approach to LLM Reasoning: from Rejection Sampling to
Reinforce | 12 pages, 4 figures | null | null | null | cs.LG cs.AI cs.CL stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reinforcement learning (RL) has become a prevailing approach for fine-tuning
large language models (LLMs) on complex reasoning tasks. Among recent methods,
GRPO stands out for its empirical success in training models such as
DeepSeek-R1, yet the sources of its effectiveness remain poorly understood. In
this work, we revisit GRPO from a reinforce-like algorithm perspective and
analyze its core components. Surprisingly, we find that a simple rejection
sampling baseline, RAFT, which trains only on positively rewarded samples,
yields competitive performance than GRPO and PPO. Our ablation studies reveal
that GRPO's main advantage arises from discarding prompts with entirely
incorrect responses, rather than from its reward normalization. Motivated by
this insight, we propose Reinforce-Rej, a minimal extension of policy gradient
that filters both entirely incorrect and entirely correct samples.
Reinforce-Rej improves KL efficiency and stability, serving as a lightweight
yet effective alternative to more complex RL algorithms. We advocate RAFT as a
robust and interpretable baseline, and suggest that future advances should
focus on more principled designs for incorporating negative samples, rather
than relying on them indiscriminately. Our findings provide guidance for future
work in reward-based LLM post-training.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:15:02 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Xiong",
"Wei",
""
],
[
"Yao",
"Jiarui",
""
],
[
"Xu",
"Yuhui",
""
],
[
"Pang",
"Bo",
""
],
[
"Wang",
"Lei",
""
],
[
"Sahoo",
"Doyen",
""
],
[
"Li",
"Junnan",
""
],
[
"Jiang",
"Nan",
""
],
[
"Zhang",
"Tong",
""
],
[
"Xiong",
"Caiming",
""
],
[
"Dong",
"Hanze",
""
]
] |
2504.11345 | Daniel Sebasti\'an | Luis Miguel Pardo, Daniel Sebasti\'an | Erzeugunsgrad, VC-Dimension and Neural Networks with rational activation
function | 50 pages | null | null | null | cs.LG math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The notion of Erzeugungsgrad was introduced by Joos Heintz in 1983 to bound
the number of non-empty cells occurring after a process of quantifier
elimination. We extend this notion and the combinatorial bounds of Theorem 2 in
Heintz (1983) using the degree for constructible sets defined in
Pardo-Sebasti\'an (2022). We show that the Erzeugungsgrad is the key ingredient
to connect affine Intersection Theory over algebraically closed fields and the
VC-Theory of Computational Learning Theory for families of classifiers given by
parameterized families of constructible sets. In particular, we prove that the
VC-dimension and the Krull dimension are linearly related up to logarithmic
factors based on Intersection Theory. Using this relation, we study the density
of correct test sequences in evasive varieties. We apply these ideas to analyze
parameterized families of neural networks with rational activation function.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:16:38 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Pardo",
"Luis Miguel",
""
],
[
"Sebastián",
"Daniel",
""
]
] |
2504.11348 | Ali\'enor Goubault-Larrecq | Ali\'enor Goubault-Larrecq, K\'evin Perrot | Circuit metaconstruction in logspace for Rice-like complexity lower
bounds in ANs and SGRs | null | null | null | null | cs.CC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A new proof technique combining finite model theory and dynamical systems has
recently been introduced to obtain general complexity lower bounds on any
question one may formulate on the dynamics (seen as a graph) of a given
automata network (AN). ANs are abstract finite dynamical systems of interacting
entities whose evolution rules are encoded as circuits, hence the study also
applies to succinct graph representations (SGRs). In this article, we detail
the construction of circuits to obtain general complexity lower bounds
(metareduction) and show that the reduction is feasible in logarithmic space.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:20:22 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Goubault-Larrecq",
"Aliénor",
""
],
[
"Perrot",
"Kévin",
""
]
] |
2504.11349 | Yuezhe Yang | Yuezhe Yang, Boyu Yang, Yaqian Wang, Yang He, Xingbo Dong, Zhe Jin | Explicit and Implicit Representations in AI-based 3D Reconstruction for
Radiology: A systematic literature review | 43 pages, 5 figures, submit to Medical Image Analysis | null | null | null | cs.CV cs.AI cs.GR | http://creativecommons.org/licenses/by/4.0/ | The demand for high-quality medical imaging in clinical practice and assisted
diagnosis has made 3D reconstruction in radiological imaging a key research
focus. Artificial intelligence (AI) has emerged as a promising approach to
enhancing reconstruction accuracy while reducing acquisition and processing
time, thereby minimizing patient radiation exposure and discomfort and
ultimately benefiting clinical diagnosis. This review explores state-of-the-art
AI-based 3D reconstruction algorithms in radiological imaging, categorizing
them into explicit and implicit approaches based on their underlying
principles. Explicit methods include point-based, volume-based, and Gaussian
representations, while implicit methods encompass implicit prior embedding and
neural radiance fields. Additionally, we examine commonly used evaluation
metrics and benchmark datasets. Finally, we discuss the current state of
development, key challenges, and future research directions in this evolving
field. Our project available on: https://github.com/Bean-Young/AI4Med.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:21:47 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Yang",
"Yuezhe",
""
],
[
"Yang",
"Boyu",
""
],
[
"Wang",
"Yaqian",
""
],
[
"He",
"Yang",
""
],
[
"Dong",
"Xingbo",
""
],
[
"Jin",
"Zhe",
""
]
] |
2504.11350 | Navaneet Villodi | Navaneet Villodi and Prabhu Ramachandran | Adaptive Compressible Smoothed Particle Hydrodynamics | 60 pages, 21 figures | null | null | null | physics.flu-dyn physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Modulating the number of particles in a region is key to accurately capturing
the nuances in compressible flows with Smoothed Particle Hydrodynamics (SPH).
This paper details the implementation of a volume-based adaptive refinement and
derefinement procedure, incorporating state-of-the-art features such as
automatic local adaptivity and solution adaptivity. A shock-aware particle
shifting procedure is introduced to regularize the particle distribution while
preserving the integrity of shocks. To our knowledge, this is the first
demonstration of shock-based solution adaptivity and shock-aware particle
shifting in the literature. A wide variety of test problems, which involve flow
in and around boundaries, are employed to highlight the utility of these
adaptivity features in improving the results and in making simulations faster.
For instance, the adaptive resolution procedure is shown to deliver an order of
magnitude speedup. We also demonstrate the effectiveness of the adaptivity
procedure in resolving existing issues like errors due to interaction with
differently spaced ghost particles at boundaries, formation of spot-like
structures due to particle clumping, and poorly resolved low-density regions.
In essence, the adaptivity technique presented in this paper is positioned as a
powerful tool for simulating compressible flows with enhanced accuracy and
efficiency.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:22:01 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Villodi",
"Navaneet",
""
],
[
"Ramachandran",
"Prabhu",
""
]
] |
2504.11351 | Christian M\"uller | Christian M\"uller, Helmut Pottmann | Isometric Surfaces in Isotropic 3-Space | 40 pages | null | null | null | math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While the notion of isometric deformations of surfaces is straightforward for
surfaces with Euclidean metric, a corresponding notion in isotropic space has
been missing. By making Gauss' Theorema Egregium a necessary condition we
develop a sensible notion of isometric surfaces in isotropic space. The
well-known examples in Euclidean space, like isometries within the associated
family of minimal surfaces, Bour's theorem, and Minding isometries, find their
natural analogues in isotropic space. We also include an extensive treatment of
infinitesimal flexibility, or infinitesimal deformation, of surfaces. We prove
results for the isotropic displacement diagrams in analogy to its well-known
counterparts in Euclidean space culminating in the existence of an isotropic
Darboux wreath consisting of six surfaces. We show several interesting
relations for special parametrizations involving Koenigs and Voss nets of
smooth and discrete surfaces within the Darboux wreath and we encounter
surfaces of constant Gaussian and mean curvature. At several occasions, we
point to connections to statics as the isotropic space is a natural language to
describe the Airy stress function.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:22:37 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Müller",
"Christian",
""
],
[
"Pottmann",
"Helmut",
""
]
] |
2504.11352 | Callum Arnold | Callum R. K. Arnold and Matthew J. Ferrari | Diagnostic Uncertainty Limits the Potential of Early Warning Signals to
Identify Epidemic Emergence | 18 pages, 2 figures, 4 tables | null | null | null | q-bio.OT | http://creativecommons.org/licenses/by/4.0/ | Methods to detect the emergence of infectious diseases, and approach to the
"critical transition" RE = 1, have to potential to avert substantial disease
burden by facilitating preemptive actions like vaccination campaigns. Early
warning signals (EWS), summary statistics of infection case time series, show
promise in providing such advanced warnings. As EWS are computed on test
positive case data, the accuracy of this underlying data is integral to their
predictive ability, but will vary with changes in the diagnostic test accuracy
and the incidence of the target disease relative to clinically-compatible
background noise. We simulated emergent and null time series as the sum of an
SEIR-generated measles time series, and background noise generated by either
independent draws from a Poisson distribution, or an SEIR simulation with
rubella-like parameters. We demonstrate that proactive outbreak detection with
EWS metrics is resilient to decreasing diagnostic accuracy, so long as
background infections remain proportionally low. Under situations with large,
episodic, noise, imperfect diagnostic tests cannot appropriately discriminate
between emergent and null periods. Not all EWS metrics performed equally: we
find that the mean was the least affected by changes to the noise structure and
magnitude, given a moderately accurate diagnostic test (>= to 95% sensitive and
specific), and the autocovariance and variance were the most predictive when
the noise incidence did not exhibit large temporal variations. In these
situations, diagnostic test accuracy should not be a precursor to the
implementation of an EWS metric-based alert system.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:22:58 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Arnold",
"Callum R. K.",
""
],
[
"Ferrari",
"Matthew J.",
""
]
] |
2504.11353 | Jundi Huang | Jundi Huang, Dawei Zhan | An Adaptive Dropout Approach for High-Dimensional Bayesian Optimization | null | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Bayesian optimization (BO) is a widely used algorithm for solving expensive
black-box optimization problems. However, its performance decreases
significantly on high-dimensional problems due to the inherent
high-dimensionality of the acquisition function. In the proposed algorithm, we
adaptively dropout the variables of the acquisition function along the
iterations. By gradually reducing the dimension of the acquisition function,
the proposed approach has less and less difficulty to optimize the acquisition
function. Numerical experiments demonstrate that AdaDropout effectively tackle
high-dimensional challenges and improve solution quality where standard
Bayesian optimization methods often struggle. Moreover, it achieves superior
results when compared with state-of-the-art high-dimensional Bayesian
optimization approaches. This work provides a simple yet efficient solution for
high-dimensional expensive optimization.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:23:25 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Huang",
"Jundi",
""
],
[
"Zhan",
"Dawei",
""
]
] |
2504.11354 | Haiming Wang | Haiming Wang, Mert Unsal, Xiaohan Lin, Mantas Baksys, Junqi Liu, Marco
Dos Santos, Flood Sung, Marina Vinyes, Zhenzhe Ying, Zekai Zhu, Jianqiao Lu,
Hugues de Saxc\'e, Bolton Bailey, Chendong Song, Chenjun Xiao, Dehao Zhang,
Ebony Zhang, Frederick Pu, Han Zhu, Jiawei Liu, Jonas Bayer, Julien Michel,
Longhui Yu, L\'eo Dreyfus-Schmidt, Lewis Tunstall, Luigi Pagani, Moreira
Machado, Pauline Bourigault, Ran Wang, Stanislas Polu, Thibaut Barroyer,
Wen-Ding Li, Yazhe Niu, Yann Fleureau, Yangyang Hu, Zhouliang Yu, Zihan Wang,
Zhilin Yang, Zhengying Liu, Jia Li | Kimina-Prover Preview: Towards Large Formal Reasoning Models with
Reinforcement Learning | 22 pages | null | null | null | cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce Kimina-Prover Preview, a large language model that pioneers a
novel reasoning-driven exploration paradigm for formal theorem proving, as
showcased in this preview release. Trained with a large-scale reinforcement
learning pipeline from Qwen2.5-72B, Kimina-Prover demonstrates strong
performance in Lean 4 proof generation by employing a structured reasoning
pattern we term \textit{formal reasoning pattern}. This approach allows the
model to emulate human problem-solving strategies in Lean, iteratively
generating and refining proof steps. Kimina-Prover sets a new state-of-the-art
on the miniF2F benchmark, reaching 80.7% with pass@8192. Beyond improved
benchmark performance, our work yields several key insights: (1) Kimina-Prover
exhibits high sample efficiency, delivering strong results even with minimal
sampling (pass@1) and scaling effectively with computational budget, stemming
from its unique reasoning pattern and RL training; (2) we demonstrate clear
performance scaling with model size, a trend previously unobserved for neural
theorem provers in formal mathematics; (3) the learned reasoning style,
distinct from traditional search algorithms, shows potential to bridge the gap
between formal verification and informal mathematical intuition. We open source
distilled versions with 1.5B and 7B parameters of Kimina-Prover
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:23:44 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Wang",
"Haiming",
""
],
[
"Unsal",
"Mert",
""
],
[
"Lin",
"Xiaohan",
""
],
[
"Baksys",
"Mantas",
""
],
[
"Liu",
"Junqi",
""
],
[
"Santos",
"Marco Dos",
""
],
[
"Sung",
"Flood",
""
],
[
"Vinyes",
"Marina",
""
],
[
"Ying",
"Zhenzhe",
""
],
[
"Zhu",
"Zekai",
""
],
[
"Lu",
"Jianqiao",
""
],
[
"de Saxcé",
"Hugues",
""
],
[
"Bailey",
"Bolton",
""
],
[
"Song",
"Chendong",
""
],
[
"Xiao",
"Chenjun",
""
],
[
"Zhang",
"Dehao",
""
],
[
"Zhang",
"Ebony",
""
],
[
"Pu",
"Frederick",
""
],
[
"Zhu",
"Han",
""
],
[
"Liu",
"Jiawei",
""
],
[
"Bayer",
"Jonas",
""
],
[
"Michel",
"Julien",
""
],
[
"Yu",
"Longhui",
""
],
[
"Dreyfus-Schmidt",
"Léo",
""
],
[
"Tunstall",
"Lewis",
""
],
[
"Pagani",
"Luigi",
""
],
[
"Machado",
"Moreira",
""
],
[
"Bourigault",
"Pauline",
""
],
[
"Wang",
"Ran",
""
],
[
"Polu",
"Stanislas",
""
],
[
"Barroyer",
"Thibaut",
""
],
[
"Li",
"Wen-Ding",
""
],
[
"Niu",
"Yazhe",
""
],
[
"Fleureau",
"Yann",
""
],
[
"Hu",
"Yangyang",
""
],
[
"Yu",
"Zhouliang",
""
],
[
"Wang",
"Zihan",
""
],
[
"Yang",
"Zhilin",
""
],
[
"Liu",
"Zhengying",
""
],
[
"Li",
"Jia",
""
]
] |
2504.11355 | Alberto Castillo Dr | Alberto Castillo, Elliot Pryor, Anas El Fathi, Boris Kovatchev, Marc
Breton | Neural Networks for on-chip Model Predictive Control: a Method to Build
Optimized Training Datasets and its application to Type-1 Diabetes | null | null | null | null | eess.SY cs.AI cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Training Neural Networks (NNs) to behave as Model Predictive Control (MPC)
algorithms is an effective way to implement them in constrained embedded
devices. By collecting large amounts of input-output data, where inputs
represent system states and outputs are MPC-generated control actions, NNs can
be trained to replicate MPC behavior at a fraction of the computational cost.
However, although the composition of the training data critically influences
the final NN accuracy, methods for systematically optimizing it remain
underexplored. In this paper, we introduce the concept of Optimally-Sampled
Datasets (OSDs) as ideal training sets and present an efficient algorithm for
generating them. An OSD is a parametrized subset of all the available data that
(i) preserves existing MPC information up to a certain numerical resolution,
(ii) avoids duplicate or near-duplicate states, and (iii) becomes saturated or
complete. We demonstrate the effectiveness of OSDs by training NNs to replicate
the University of Virginia's MPC algorithm for automated insulin delivery in
Type-1 Diabetes, achieving a four-fold improvement in final accuracy. Notably,
two OSD-trained NNs received regulatory clearance for clinical testing as the
first NN-based control algorithm for direct human insulin dosing. This
methodology opens new pathways for implementing advanced optimizations on
resource-constrained embedded platforms, potentially revolutionizing how
complex algorithms are deployed.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:25:06 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Castillo",
"Alberto",
""
],
[
"Pryor",
"Elliot",
""
],
[
"Fathi",
"Anas El",
""
],
[
"Kovatchev",
"Boris",
""
],
[
"Breton",
"Marc",
""
]
] |
2504.11356 | Ekta Agrawal | Ekta Agrawal and Saurabh Verma | Dimension preserving set-valued approximation and decomposition via
metric sum | 32 pages, 2 figures | null | null | null | math.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the literature, the Minkowski-sum and the metric-sum of compact sets are
highlighted. While the first is associative, the latter is not. But the major
drawback of the Minkowski combination is that, by increasing the number of
summands, this leads to convexification. The present article is uncovered in
two folds: The initial segment presents a novel approach to approximate a
continuous set-valued function with compact images via a fractal approach using
the metric linear combination of sets. The other segment contains the dimension
analysis of the distance set of graph of set-valued function and solving the
celebrated distance set conjecture. In the end, a decomposition of any
continuous convex compact set-valued function is exhibited that preserves the
Hausdorff dimension, so this will serve as a method for dealing with
complicated set-valued functions.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:25:29 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Agrawal",
"Ekta",
""
],
[
"Verma",
"Saurabh",
""
]
] |
2504.11357 | Alessandro Broggio | Simone Alioli, Georgios Billis, Alessandro Broggio, Giovanni Stagnitto | NNLO predictions with nonlocal subtractions and fiducial power
corrections in GENEVA | 42 pages, 12 figures | null | null | PSI-PR-25-08, UWThPh 2025-14 | hep-ph hep-ex | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the implementation of next-to-next-to-leading order (NNLO) QCD
fully-differential corrections within the GENEVA framework, for both
colour-singlet and colour-singlet+jet processes at hadron colliders, by
employing a nonlocal subtraction approach. In particular, we discuss the
implementation details and the challenges that arise when utilizing a dynamical
infrared cutoff parameter. Additionally, we combine the subtraction with the
projection-to-Born method in order to include fiducial power corrections. As a
test case, we provide predictions for Drell-Yan and $Z$+jet production at the
LHC, using $N$-jettiness as resolution variable. We validate the NNLO
corrections of GENEVA against NNLOJET finding excellent agreement. Finally, we
discuss how to extend our method to calculate the N$^3$LO QCD
fully-differential corrections to colour-singlet production at hadron
colliders.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:26:01 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Alioli",
"Simone",
""
],
[
"Billis",
"Georgios",
""
],
[
"Broggio",
"Alessandro",
""
],
[
"Stagnitto",
"Giovanni",
""
]
] |
2504.11358 | Yupei Liu | Yupei Liu, Yuqi Jia, Jinyuan Jia, Dawn Song, Neil Zhenqiang Gong | DataSentinel: A Game-Theoretic Detection of Prompt Injection Attacks | To appear in IEEE Symposium on Security and Privacy, 2025 | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by/4.0/ | LLM-integrated applications and agents are vulnerable to prompt injection
attacks, where an attacker injects prompts into their inputs to induce
attacker-desired outputs. A detection method aims to determine whether a given
input is contaminated by an injected prompt. However, existing detection
methods have limited effectiveness against state-of-the-art attacks, let alone
adaptive ones. In this work, we propose DataSentinel, a game-theoretic method
to detect prompt injection attacks. Specifically, DataSentinel fine-tunes an
LLM to detect inputs contaminated with injected prompts that are strategically
adapted to evade detection. We formulate this as a minimax optimization
problem, with the objective of fine-tuning the LLM to detect strong adaptive
attacks. Furthermore, we propose a gradient-based method to solve the minimax
optimization problem by alternating between the inner max and outer min
problems. Our evaluation results on multiple benchmark datasets and LLMs show
that DataSentinel effectively detects both existing and adaptive prompt
injection attacks.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:26:21 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Liu",
"Yupei",
""
],
[
"Jia",
"Yuqi",
""
],
[
"Jia",
"Jinyuan",
""
],
[
"Song",
"Dawn",
""
],
[
"Gong",
"Neil Zhenqiang",
""
]
] |
2504.11359 | Thomas Chaney | Thomas P. Chaney, Christine LaPorte Mahajan, Masoud Ghasemi, Andrew J.
Levin, Keith P. White, Scott T. Milner, Enrique D. Gomez and Michael F. Toney | Deciphering the Structure of Push-Pull Conjugated Polymer Aggregates in
Solution and Film | Main text: 29 pages, 6 figures. Supporting Information: 16 pages, 19
figures | null | null | null | cond-mat.mtrl-sci cond-mat.soft | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The morphology of conjugated polymer films is highly tunable, influencing
their performance in organic electronics. Specifically, molecular packing or
crystal structure strongly influence electronic processes such as light
absorption and charge transfer. However, the unit cells of high-performance
electron donor polymers remain unknown, limiting the understanding of how
processing affects structure and device performance. This study characterizes
the aggregate structure of PM6-type push-pull polymers using X-ray scattering,
cryogenic electron microscopy, and molecular dynamics (MD) simulations. A novel
forward simulation approach linking grazing-incidence wide-angle X-ray
scattering (GIWAXS) with MD resolves a monoclinic unit cell that accurately
describes PM6-type polymer aggregates in both thin films and casting solutions.
Intimate pi-pi stacking between donor and acceptor units emerges from this unit
cell. Analysis of experimental GIWAXS using this unit cell quantifies sliding
disorder in these aggregates, which may impact device performance. The shape
and internal structure of solution aggregates are also identified in
chlorobenzene. These findings enhance our understanding of PM6-type polymer
packing, outline a strategy for elucidating the crystal structure of weakly
ordered materials, and provide an opportunity to control optoelectronic
performance through aggregate formation in PM6 and other push-pull conjugated
polymers.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:26:28 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Chaney",
"Thomas P.",
""
],
[
"Mahajan",
"Christine LaPorte",
""
],
[
"Ghasemi",
"Masoud",
""
],
[
"Levin",
"Andrew J.",
""
],
[
"White",
"Keith P.",
""
],
[
"Milner",
"Scott T.",
""
],
[
"Gomez",
"Enrique D.",
""
],
[
"Toney",
"Michael F.",
""
]
] |
2504.11360 | Nicola Bariletto | Nicola Bariletto, Bernardo Flores, Stephen G. Walker | Posterior Consistency in Parametric Models via a Tighter Notion of
Identifiability | 34 pages, 2 figures | null | null | null | math.ST stat.TH | http://creativecommons.org/licenses/by/4.0/ | We investigate Bayesian posterior consistency in the context of parametric
models with proper priors. While typical state-of-the-art approaches rely on
regularity conditions that are difficult to verify and often misaligned with
the actual mechanisms driving posterior consistency, we propose an alternative
framework centered on a simple yet general condition we call ''sequential
identifiability''. This concept strengthens the usual identifiability
assumption by requiring that any sequence of parameters whose induced
distributions converge to the true data-generating distribution must itself
converge to the true parameter value. We demonstrate that sequential
identifiability, combined with a standard Kullback--Leibler prior support
condition, is sufficient to ensure posterior consistency. Moreover, we show
that failure of this condition necessarily entails a specific and pathological
form of oscillations of the model around the true density, which cannot exist
without intentional design. This leads to the important insight that posterior
inconsistency may be safely ruled out, except in the unrealistic scenario where
the modeler possesses precise knowledge of the data-generating distribution and
deliberately incorporates oscillatory pathologies into the model targeting the
corresponding density. Taken together, these results provide a unified
perspective on both consistency and inconsistency in parametric settings,
significantly expanding the class of models for which posterior consistency can
be rigorously established. To illustrate the strength and versatility of our
framework, we construct a one-dimensional model that violates standard
regularity conditions and fails to admit a consistent maximum likelihood
estimator, yet supports a complete posterior consistency analysis via our
approach.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:26:34 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Bariletto",
"Nicola",
""
],
[
"Flores",
"Bernardo",
""
],
[
"Walker",
"Stephen G.",
""
]
] |
2504.11361 | Fernando C. Lombardo | Fernando C. Lombardo and Paula I. Villar | Dynamical Casimir effect in superconducting cavities: from photon
generation to universal quantum gates | 35 pages, 8 figures. Chapter of the Proceedings for the III
International Workshop on Quantum Nonstationary Systems (eds. Alexandre
Dodonov and Lucas Chibebe C\'eleri) | null | null | null | quant-ph cond-mat.other | http://creativecommons.org/licenses/by/4.0/ | This chapter explores various aspects of the Dynamical Casimir Effect (DCE)
and its implications in the context of circuit quantum electrodynamics (cQED).
We begin by reviewing the origin and fundamental properties of the DCE,
including three equivalent mathematical frameworks that offer complementary
perspectives on the phenomenon. These formulations will serve as a foundation
for the subsequent analyses. We then turn our attention to the practical
realization of the DCE in cQED-based architectures, discussing how modern
superconducting circuits can be engineered to exhibit this inherently quantum
effect. Building on this, we examine how the presence of the DCE influences the
performance of a quantum thermal machine operating with a quantum field,
shedding light on the interplay between quantum fluctuations and thermodynamic
processes. Further, we demonstrate how the DCE can be harnessed to implement a
controlled-squeeze gate within a cQED platform, opening a path toward advanced
quantum control and quantum information processing. The chapter concludes with
a synthesis of the main results and a discussion of potential future
directions.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:28:00 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Lombardo",
"Fernando C.",
""
],
[
"Villar",
"Paula I.",
""
]
] |
2504.11362 | Florian Cougoulic | Florian Cougoulic and St\'ephane Peign\'e | $27 \otimes 27$ | 61pages | null | null | null | hep-ph hep-th | http://creativecommons.org/licenses/by/4.0/ | We study the decomposition into $SU(N)$ irreducible representations (irreps)
of the tensor product $27 \otimes 27$, where $27$ is the highest-dimensional
$SU(N)$ irrep present in a two-gluon system, and explicitly construct all
Hermitian projectors on these irreps, as well as transition operators between
them. This yields an explicit basis of the complete $27\otimes 27$ color space
(defined as the space of $27\otimes 27\to 27 \otimes 27$ color maps) in terms
of orthogonal multiplets. This study shows that even complex color structures
can be addressed, with the help of the birdtrack pictorial technique, using
only elementary tools. In particular, we highlight the usefulness of the
quadratic Casimir operator, whose eigenspaces allow efficient filtering of all
projectors and transition operators, and of the permutation operators that
further improve this filtering. The product $27\otimes 27$ also has an
interesting feature: three equivalent irreps $27$ appear in the decomposition,
two of which are symmetric and can therefore be distinguished neither by the
quadratic Casimir operator nor by their symmetry under permutation. In this
case, it is convenient to use Clebsch-Gordan coefficients to derive the two
associated, symmetric projectors. The latter are not uniquely determined (only
their sum is), and we give the set of all solutions. Finally, we explicitly
derive the soft anomalous dimension matrix associated with $27\otimes 27\to
27\otimes 27$, whose block-diagonal main structure is easy to understand, but
whose detailed spectrum properties remain intriguing. The approach presented
for $27 \otimes 27 \to 27 \otimes 27$ could in principle be applied to any
product of $SU(N)$ irreps, and eventually automated.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:28:10 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Cougoulic",
"Florian",
""
],
[
"Peigné",
"Stéphane",
""
]
] |
2504.11363 | Yiping Zhao | Yiping Zhao | Smartphone-Based Undergraduate Physics Labs: A Comprehensive Review of
Innovation, Accessibility, and Pedagogical Impact | 64 pages, 2 figures, 18 tables | null | null | null | physics.ed-ph | http://creativecommons.org/publicdomain/zero/1.0/ | Smartphone-integrated physics laboratories (SmartIPLs) have emerged as
scalable and cost-effective alternatives to traditional lab instruction,
providing accessible, hands-on experiences for diverse learning environments.
This review synthesizes over a decade of research, covering nearly 200
SmartIPLs across key physics domains such as mechanics, optics, acoustics,
electromagnetism, thermodynamics, and modern physics. SmartIPLs are categorized
into two primary types: sensor-based experiments using built-in smartphone
tools and camera-based video/image analysis for motion and optical studies.
Empirical studies show that SmartIPLs support equal or greater gains in
conceptual understanding, science process skills, and student engagement,
especially in remote and under-resourced settings. The review explores their
theoretical foundations, compares them to traditional and virtual labs, and
addresses challenges such as device variability and classroom integration.
Future directions include broader curricular integration, AI-driven student
feedback, expansion into underrepresented physics topics, interdisciplinary
applications, and equity-focused instructional design. Open-access resources,
such as the UGA SmartPhone Intro Physics Lab and Modern Optics YouTube channels
and the SPIE book Use of Smartphones in Optical Experimentation, highlight
community-driven efforts to expand and democratize physics education. As
smartphone technology advances, SmartIPLs will offer a promising path toward
adaptive, intelligent, and inclusive laboratory instruction for the 21st
century.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:28:13 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Zhao",
"Yiping",
""
]
] |
2504.11365 | Juergen Reuter | Krzysztof M\k{e}ka{\l}a, Daniel Jeans, J\"urgen Reuter, Junping Tian,
Aleksander Filip \.Zarnecki | Determination of the first-generation quark couplings at the Z-pole | 22 pages, 8 figures | null | null | DESY-25-063 | hep-ph hep-ex | http://creativecommons.org/licenses/by-sa/4.0/ | Electroweak Precision Measurements are stringent tests of the Standard Model
and sensitive probes to New Physics. Accurate studies of the $Z$-boson
couplings to the first-generation quarks, which are currently constrained from
LEP data to a few percent, could reveal potential discrepancies from the theory
predictions. Future $e^+e^-$ colliders running at the $Z$-pole would be an
excellent tool for an analysis based on a comparison of radiative and
non-radiative $Z$ boson decays. In this paper, we present a method to extract
the values of the $Z$ couplings to light quarks and discuss the uncertainty of
the measurement, including contributions from various systematic effects. We
show that systematic uncertainty in the heavy-flavour tagging performance is
the key factor in the analysis and reducing it to a sub-permille level might be
crucial to fully profit from the high luminosity of future $e^+e^-$ machines.
The measurement could improve the LEP results by at least an order of
magnitude.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:31:21 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Mękała",
"Krzysztof",
""
],
[
"Jeans",
"Daniel",
""
],
[
"Reuter",
"Jürgen",
""
],
[
"Tian",
"Junping",
""
],
[
"Żarnecki",
"Aleksander Filip",
""
]
] |
2504.11366 | Ali J. Ghandour | Hasan Wehbi, Hasan Nasrallah, Mohamad Hasan Zahweh, Zeinab Takach,
Veera Ganesh Yalla, Ali J. Ghandour | A Decade of Wheat Mapping for Lebanon | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Wheat accounts for approximately 20% of the world's caloric intake, making it
a vital component of global food security. Given this importance, mapping wheat
fields plays a crucial role in enabling various stakeholders, including policy
makers, researchers, and agricultural organizations, to make informed decisions
regarding food security, supply chain management, and resource allocation. In
this paper, we tackle the problem of accurately mapping wheat fields out of
satellite images by introducing an improved pipeline for winter wheat
segmentation, as well as presenting a case study on a decade-long analysis of
wheat mapping in Lebanon. We integrate a Temporal Spatial Vision Transformer
(TSViT) with Parameter-Efficient Fine Tuning (PEFT) and a novel post-processing
pipeline based on the Fields of The World (FTW) framework. Our proposed
pipeline addresses key challenges encountered in existing approaches, such as
the clustering of small agricultural parcels in a single large field. By
merging wheat segmentation with precise field boundary extraction, our method
produces geometrically coherent and semantically rich maps that enable us to
perform in-depth analysis such as tracking crop rotation pattern over years.
Extensive evaluations demonstrate improved boundary delineation and field-level
precision, establishing the potential of the proposed framework in operational
agricultural monitoring and historical trend analysis. By allowing for accurate
mapping of wheat fields, this work lays the foundation for a range of critical
studies and future advances, including crop monitoring and yield estimation.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:31:54 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Wehbi",
"Hasan",
""
],
[
"Nasrallah",
"Hasan",
""
],
[
"Zahweh",
"Mohamad Hasan",
""
],
[
"Takach",
"Zeinab",
""
],
[
"Yalla",
"Veera Ganesh",
""
],
[
"Ghandour",
"Ali J.",
""
]
] |
2504.11367 | Jiang Shuyu | Rui Tang, Ziyun Yong, Shuyu Jiang, Xingshu Chen, Yaofang Liu, Yi-Cheng
Zhang, Gui-Quan Sun, Wei Wang | Network Alignment | null | Physics Reports 1107 (2025): 1-45 | 10.1016/j.physrep.2024.11.006 | null | physics.soc-ph cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Complex networks are frequently employed to model physical or virtual complex
systems. When certain entities exist across multiple systems simultaneously,
unveiling their corresponding relationships across the networks becomes
crucial. This problem, known as network alignment, holds significant
importance. It enhances our understanding of complex system structures and
behaviours, facilitates the validation and extension of theoretical physics
research about studying complex systems, and fosters diverse practical
applications across various fields. However, due to variations in the
structure, characteristics, and properties of complex networks across different
fields, the study of network alignment is often isolated within each domain,
with even the terminologies and concepts lacking uniformity. This review
comprehensively summarizes the latest advancements in network alignment
research, focusing on analyzing network alignment characteristics and progress
in various domains such as social network analysis, bioinformatics,
computational linguistics and privacy protection. It provides a detailed
analysis of various methods' implementation principles, processes, and
performance differences, including structure consistency-based methods, network
embedding-based methods, and graph neural network-based (GNN-based) methods.
Additionally, the methods for network alignment under different conditions,
such as in attributed networks, heterogeneous networks, directed networks, and
dynamic networks, are presented. Furthermore, the challenges and the open
issues for future studies are also discussed.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:32:09 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Tang",
"Rui",
""
],
[
"Yong",
"Ziyun",
""
],
[
"Jiang",
"Shuyu",
""
],
[
"Chen",
"Xingshu",
""
],
[
"Liu",
"Yaofang",
""
],
[
"Zhang",
"Yi-Cheng",
""
],
[
"Sun",
"Gui-Quan",
""
],
[
"Wang",
"Wei",
""
]
] |
2504.11368 | Jingkun Chen | Jingkun Chen, Haoran Duan, Xiao Zhang, Boyan Gao, Tao Tan, Vicente
Grau, Jungong Han | From Gaze to Insight: Bridging Human Visual Attention and Vision
Language Model Explanation for Weakly-Supervised Medical Image Segmentation | 10 pages, 5 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Medical image segmentation remains challenging due to the high cost of
pixel-level annotations for training. In the context of weak supervision,
clinician gaze data captures regions of diagnostic interest; however, its
sparsity limits its use for segmentation. In contrast, vision-language models
(VLMs) provide semantic context through textual descriptions but lack the
explanation precision required. Recognizing that neither source alone suffices,
we propose a teacher-student framework that integrates both gaze and language
supervision, leveraging their complementary strengths. Our key insight is that
gaze data indicates where clinicians focus during diagnosis, while VLMs explain
why those regions are significant. To implement this, the teacher model first
learns from gaze points enhanced by VLM-generated descriptions of lesion
morphology, establishing a foundation for guiding the student model. The
teacher then directs the student through three strategies: (1) Multi-scale
feature alignment to fuse visual cues with textual semantics; (2)
Confidence-weighted consistency constraints to focus on reliable predictions;
(3) Adaptive masking to limit error propagation in uncertain areas. Experiments
on the Kvasir-SEG, NCI-ISBI, and ISIC datasets show that our method achieves
Dice scores of 80.78%, 80.53%, and 84.22%, respectively-improving 3-5% over
gaze baselines without increasing the annotation burden. By preserving
correlations among predictions, gaze data, and lesion descriptions, our
framework also maintains clinical interpretability. This work illustrates how
integrating human visual attention with AI-generated semantic context can
effectively overcome the limitations of individual weak supervision signals,
thereby advancing the development of deployable, annotation-efficient medical
AI systems. Code is available at: https://github.com/jingkunchen/FGI.git.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:32:15 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Chen",
"Jingkun",
""
],
[
"Duan",
"Haoran",
""
],
[
"Zhang",
"Xiao",
""
],
[
"Gao",
"Boyan",
""
],
[
"Tan",
"Tao",
""
],
[
"Grau",
"Vicente",
""
],
[
"Han",
"Jungong",
""
]
] |
2504.11369 | Andrea Tagarelli | Lucio La Cava, Andrea Tagarelli | OpenTuringBench: An Open-Model-based Benchmark and Framework for
Machine-Generated Text Detection and Attribution | Under review with ARR | null | null | null | cs.CL cs.AI cs.CY cs.HC physics.soc-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Open Large Language Models (OLLMs) are increasingly leveraged in generative
AI applications, posing new challenges for detecting their outputs. We propose
OpenTuringBench, a new benchmark based on OLLMs, designed to train and evaluate
machine-generated text detectors on the Turing Test and Authorship Attribution
problems. OpenTuringBench focuses on a representative set of OLLMs, and
features a number of challenging evaluation tasks, including
human/machine-manipulated texts, out-of-domain texts, and texts from previously
unseen models. We also provide OTBDetector, a contrastive learning framework to
detect and attribute OLLM-based machine-generated texts. Results highlight the
relevance and varying degrees of difficulty of the OpenTuringBench tasks, with
our detector achieving remarkable capabilities across the various tasks and
outperforming most existing detectors. Resources are available on the
OpenTuringBench Hugging Face repository at
https://huggingface.co/datasets/MLNTeam-Unical/OpenTuringBench
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:36:14 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"La Cava",
"Lucio",
""
],
[
"Tagarelli",
"Andrea",
""
]
] |
2504.11370 | Disson dos Prazeres | Julio C. Correa and Disson dos Prazeres | A two-phase quenching-type problem for the p-Laplacian | null | null | null | null | math.AP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We study minimizers of non-differentiable functionals of the Alp-Phillips
type with two-phases for the $p$-Laplacian , focusing on the geometric and
analytical properties of free boundaries. The main result establishes finite
$(n-1)$-dimensional Hausdorff measure estimates, achieved through optimal
gradient decay estimates, a $BV$-inequality and the known classifications of
blow-up profiles of the linear case.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:36:19 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Correa",
"Julio C.",
""
],
[
"Prazeres",
"Disson dos",
""
]
] |
2504.11371 | James P. Crutchfield | Alexandra Jurgens and James P. Crutchfield | Taxonomy of Prediction | 18 pages, 6 figures;
https://csc.ucdavis.edu/~cmg/compmech/pubs/top.htm | null | null | null | cond-mat.stat-mech cs.IT math.IT nlin.AO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A prediction makes a claim about a system's future given knowledge of its
past. A retrodiction makes a claim about its past given knowledge of its
future. We introduce the ambidextrous hidden Markov chain that does both
optimally -- the bidirectional machine whose state structure makes explicit all
statistical correlations in a stochastic process. We introduce an informational
taxonomy to profile these correlations via a suite of multivariate information
measures. While prior results laid out the different kinds of information
contained in isolated measurements, in addition to being limited to single
measurements the associated informations were challenging to calculate
explicitly. Overcoming these via bidirectional machine states, we expand that
analysis to information embedded across sequential measurements. The result
highlights fourteen new interpretable and calculable information measures that
fully characterize a process' informational structure. Additionally, we
introduce a labeling and indexing scheme that systematizes
information-theoretic analyses of highly complex multivariate systems.
Operationalizing this, we provide algorithms to directly calculate all of these
quantities in closed form for finitely-modeled processes.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:36:20 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Jurgens",
"Alexandra",
""
],
[
"Crutchfield",
"James P.",
""
]
] |
2504.11372 | Zhengbing He | Zhengbing He, Jorge Laval, Yu Han, Ryosuke Nishi, Cathy Wu | A Review of Traffic Wave Suppression Strategies: Variable Speed Limit
vs. Jam-Absorption Driving | null | null | null | null | physics.soc-ph cs.SY eess.SY stat.AP | http://creativecommons.org/licenses/by/4.0/ | The main form of freeway traffic congestion is the familiar stop-and-go wave,
characterized by wide moving jams that propagate indefinitely upstream provided
enough traffic demand. They cause severe, long-lasting adverse effects, such as
reduced traffic efficiency, increased driving risks, and higher vehicle
emissions. This underscores the crucial importance of artificial intervention
in the propagation of stop-and-go waves. Over the past two decades, two
prominent strategies for stop-and-go wave suppression have emerged: variable
speed limit (VSL) and jam-absorption driving (JAD). Although they share similar
research motivations, objectives, and theoretical foundations, the development
of these strategies has remained relatively disconnected. To synthesize
fragmented advances and drive the field forward, this paper first provides a
comprehensive review of the achievements in the stop-and-go wave
suppression-oriented VSL and JAD, respectively. It then focuses on bridging the
two areas and identifying research opportunities from the following
perspectives: fundamental diagrams, traffic dynamics modeling, traffic state
estimation and prediction, stochasticity, scenarios for strategy validation,
and field tests and practical deployment. We expect that through this review,
one area can effectively address its limitations by identifying and leveraging
the strengths of the other, thus promoting the overall research goal of freeway
stop-and-go wave suppression.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:37:23 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"He",
"Zhengbing",
""
],
[
"Laval",
"Jorge",
""
],
[
"Han",
"Yu",
""
],
[
"Nishi",
"Ryosuke",
""
],
[
"Wu",
"Cathy",
""
]
] |
2504.11373 | Wang Bill Zhu | Wang Bill Zhu, Tianqi Chen, Ching Ying Lin, Jade Law, Mazen Jizzini,
Jorge J. Nieva, Ruishan Liu, Robin Jia | Cancer-Myth: Evaluating AI Chatbot on Patient Questions with False
Presuppositions | null | null | null | null | cs.CL cs.CY | http://creativecommons.org/licenses/by/4.0/ | Cancer patients are increasingly turning to large language models (LLMs) as a
new form of internet search for medical information, making it critical to
assess how well these models handle complex, personalized questions. However,
current medical benchmarks focus on medical exams or consumer-searched
questions and do not evaluate LLMs on real patient questions with detailed
clinical contexts. In this paper, we first evaluate LLMs on cancer-related
questions drawn from real patients, reviewed by three hematology oncology
physicians. While responses are generally accurate, with GPT-4-Turbo scoring
4.13 out of 5, the models frequently fail to recognize or address false
presuppositions in the questions-posing risks to safe medical decision-making.
To study this limitation systematically, we introduce Cancer-Myth, an
expert-verified adversarial dataset of 585 cancer-related questions with false
presuppositions. On this benchmark, no frontier LLM -- including GPT-4o,
Gemini-1.Pro, and Claude-3.5-Sonnet -- corrects these false presuppositions
more than 30% of the time. Even advanced medical agentic methods do not prevent
LLMs from ignoring false presuppositions. These findings expose a critical gap
in the clinical reliability of LLMs and underscore the need for more robust
safeguards in medical AI systems.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:37:32 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Zhu",
"Wang Bill",
""
],
[
"Chen",
"Tianqi",
""
],
[
"Lin",
"Ching Ying",
""
],
[
"Law",
"Jade",
""
],
[
"Jizzini",
"Mazen",
""
],
[
"Nieva",
"Jorge J.",
""
],
[
"Liu",
"Ruishan",
""
],
[
"Jia",
"Robin",
""
]
] |
2504.11374 | Yongkang Huo | Yongkang Huo, Fuvio Forni, Rodolphe Sepulchre | A Winner-Takes-All Mechanism for Event Generation | null | null | null | null | eess.SY cs.AI cs.SY | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We present a novel framework for central pattern generator design that
leverages the intrinsic rebound excitability of neurons in combination with
winner-takes-all computation. Our approach unifies decision-making and rhythmic
pattern generation within a simple yet powerful network architecture that
employs all-to-all inhibitory connections enhanced by designable excitatory
interactions. This design offers significant advantages regarding ease of
implementation, adaptability, and robustness. We demonstrate its efficacy
through a ring oscillator model, which exhibits adaptive phase and frequency
modulation, making the framework particularly promising for applications in
neuromorphic systems and robotics.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:40:37 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Huo",
"Yongkang",
""
],
[
"Forni",
"Fuvio",
""
],
[
"Sepulchre",
"Rodolphe",
""
]
] |
2504.11375 | Yunze Liu | Yunze Liu, Congyi Su, Xing Zhao | Ring Artifacts Correction Based on Global-Local Features Interaction
Guidance in the Projection Domain | 13 pages, 14 figures | null | null | null | eess.IV | http://creativecommons.org/licenses/by/4.0/ | Ring artifacts are common artifacts in CT imaging, typically caused by
inconsistent responses of detector units to X-rays, resulting in stripe
artifacts in the projection data. Under circular scanning mode, such artifacts
manifest as concentric rings radiating from the center of rotation, severely
degrading image quality. In the Radon transform domain, even if the object's
density function is piecewise discontinuous in certain regions, the projection
images remain nearly continuous in the angular direction, making the ideal
projections exhibit a smooth global low-frequency characteristic. In practical
scanning, the local disturbances of the same detector unit at different
scanning angles lead to a prominent high-frequency locality of stripe
artifacts. Existing studies generally model ring artifacts disturbances as
fixed additive errors, which overlooks the dynamic variation of detector
responses during practical scanning. However, the degree of detector response
inconsistency is a function of the projection values, as revealed in our
experiments, thereby requiring consideration of the interaction between global
and local features in the process of stripe artifacts extraction and
correction. Therefore, we propose a CT ring artifacts correction method based
on global and local features in the projection domain. We employ the VSS block
and Dense block to respectively correct the low-frequency sub-band, which
capture the global correlations of the projection, and the high-frequency
sub-band, which contain local stripe artifacts after wavelet decomposition.
Specifically, the accuracy of artifacts correction is enhanced by the
interaction guidance between global and local features. Extensive experiments
demonstrate that our method achieves superior performance in both quantitative
metrics and visual quality, verifying its robustness and practical
applicability.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:42:23 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Liu",
"Yunze",
""
],
[
"Su",
"Congyi",
""
],
[
"Zhao",
"Xing",
""
]
] |
2504.11376 | Yilmaz Ege Gonul | Yilmaz Ege Gonul, Baris Taskin | A Multi-Stage Potts Machine based on Coupled CMOS Ring Oscillators | 8 pages, 5 figures, conference: Design, Automation, and Test in
Europe (DATE) Conference 2025, Lyon, France | null | null | null | cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work presents a multi-stage coupled ring oscillator
based Potts machine, designed with phase-shifted Sub
Harmonic-Injection-Locking (SHIL) to represent multi valued Potts spins at
different solution stages with os cillator phases. The proposed Potts machine
is able to
solve a certain class of combinatorial optimization prob lems that natively
require multivalued spins with a divide and-conquer approach, facilitated
through the alternating
phase-shifted SHILs acting on the oscillators. The pro posed architecture
eliminates the need for any external in termediary mappings or usage of
external memory, as the
influence of SHIL allows oscillators to act as both mem ory and computation
units. Planar 4-coloring problems
of sizes up to 2116 nodes are mapped to the proposed
architecture. Simulations demonstrate that the proposed
Potts machine provides exact solutions for smaller prob lems (e.g. 49 nodes)
and generates solutions reaching up
to 97% accuracy for larger problems (e.g. 2116 nodes).
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:47:34 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Gonul",
"Yilmaz Ege",
""
],
[
"Taskin",
"Baris",
""
]
] |
2504.11377 | Isabel Lehenbauer | Kevin Soto, Isabel Hess, Brandon Schrader, Shan He and Patrick
Musgrave | Improving Swimming Performance in Soft Robotic Fish with Distributed
Muscles and Embedded Kinematic Sensing | 7 pages, 5 figures. To be published in the Proceedings of the 8th
IEEE - RAS International Conference on Soft Robotics | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Bio-inspired underwater vehicles could yield improved efficiency,
maneuverability, and environmental compatibility over conventional
propeller-driven underwater vehicles. However, to realize the swimming
performance of biology, there is a need for soft robotic swimmers with both
distributed muscles and kinematic feedback. This study presents the design and
swimming performance of a soft robotic fish with independently controllable
muscles and embedded kinematic sensing distributed along the body. The soft
swimming robot consists of an interior flexible spine, three axially
distributed sets of HASEL artificial muscles, embedded strain gauges, a
streamlined silicone body, and off-board electronics. In a fixed configuration,
the soft robot generates a maximum thrust of 7.9 mN when excited near its first
resonant frequency (2 Hz) with synchronized antagonistic actuation of all
muscles. When excited near its second resonant frequency (8 Hz), synchronized
muscle actuation generates 5.0 mN of thrust. By introducing a sequential phase
offset into the muscle actuation, the thrust at the second resonant frequency
increases to 7.2 mN, a 44% increase from simple antagonistic activation. The
sequential muscle activation improves the thrust by increasing 1) the tail-beat
velocity and 2) traveling wave content in the swimming kinematics by four
times. Further, the second resonant frequency (8 Hz) generates nearly as much
thrust as the first resonance (2 Hz) while requiring only $\approx25$% of the
tail displacement, indicating that higher resonant frequencies have benefits
for swimming in confined environments where a smaller kinematic envelope is
necessary. These results demonstrate the performance benefits of independently
controllable muscles and distributed kinematic sensing, and this type of soft
robotic swimmer provides a platform to address the open challenge of
sensorimotor control.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:48:08 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Soto",
"Kevin",
""
],
[
"Hess",
"Isabel",
""
],
[
"Schrader",
"Brandon",
""
],
[
"He",
"Shan",
""
],
[
"Musgrave",
"Patrick",
""
]
] |
2504.11378 | Ozgur Delice | Hatice \"Ozer, \"Ozg\"ur Delice | Gravitational waves in massive Horndeski theory with a cosmological
constant | 19 pages, revtex 4-2 | null | null | null | gr-qc hep-th | http://creativecommons.org/licenses/by/4.0/ | We study gravitational waves in a de-Sitter background in linearized
Horndeski theory. This background can be obtained in different ways in this
theory. We first consider an arbitrary potential to implement an effective
cosmological constant. The scalar field has long range whereas its potential
can induce a mass term, rendering its range to short distances. A second effect
of the potential is the possibility to induce a background curvature or an
effective cosmological constant via the minimum of the potential. The latter is
usually ignored in the literature to have an asymptotically flat background for
the gravitational waves. In order to fill this gap, we consider the effects of
the scalar potential in the framework of gravitational waves in the theories
having such scalar couplings, and compare its effects with other approaches
such as the presence of a vacuum energy density filling the spacetime or a
linear potential which leads to an effective cosmological constant but now with
a massless scalar field. We discuss properties of the gravitational waves,
their propagation and interaction with the medium in which they propagate. The
minimum of the scalar potential term mimics a cosmological constant and leads
to separate redshifts in the effective frequency and wave number, which also
discriminates the tensorial and massive scalar waves. We see that, depending on
the implementation to have a cosmological constant, the propagation speeds and
polarization properties of scalar modes of these waves is different, as
summarized in the paper.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:51:33 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Özer",
"Hatice",
""
],
[
"Delice",
"Özgür",
""
]
] |
2504.11379 | Liu Yang | Liu Yang, Huiyu Duan, Yucheng Zhu, Xiaohong Liu, Lu Liu, Zitong Xu,
Guangji Ma, Xiongkuo Min, Guangtao Zhai, Patrick Le Callet | Omni$^2$: Unifying Omnidirectional Image Generation and Editing in an
Omni Model | 10 pages | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | $360^{\circ}$ omnidirectional images (ODIs) have gained considerable
attention recently, and are widely used in various virtual reality (VR) and
augmented reality (AR) applications. However, capturing such images is
expensive and requires specialized equipment, making ODI synthesis increasingly
important. While common 2D image generation and editing methods are rapidly
advancing, these models struggle to deliver satisfactory results when
generating or editing ODIs due to the unique format and broad 360$^{\circ}$
Field-of-View (FoV) of ODIs. To bridge this gap, we construct
\textbf{\textit{Any2Omni}}, the first comprehensive ODI generation-editing
dataset comprises 60,000+ training data covering diverse input conditions and
up to 9 ODI generation and editing tasks. Built upon Any2Omni, we propose an
\textbf{\underline{Omni}} model for \textbf{\underline{Omni}}-directional image
generation and editing (\textbf{\textit{Omni$^2$}}), with the capability of
handling various ODI generation and editing tasks under diverse input
conditions using one model. Extensive experiments demonstrate the superiority
and effectiveness of the proposed Omni$^2$ model for both the ODI generation
and editing tasks.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:53:11 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Yang",
"Liu",
""
],
[
"Duan",
"Huiyu",
""
],
[
"Zhu",
"Yucheng",
""
],
[
"Liu",
"Xiaohong",
""
],
[
"Liu",
"Lu",
""
],
[
"Xu",
"Zitong",
""
],
[
"Ma",
"Guangji",
""
],
[
"Min",
"Xiongkuo",
""
],
[
"Zhai",
"Guangtao",
""
],
[
"Callet",
"Patrick Le",
""
]
] |
2504.11380 | Jordan Aiko Deja Mr | Mark Edison Jim, Jan Benjamin Yap, Gian Chill Laolao, Andrei Zachary
Lim and Jordan Aiko Deja | Speak with Confidence: Designing an Augmented Reality Training Tool for
Public Speaking | 7 pages, 3 figures, 2 tables | null | null | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | Public speaking anxiety affects many individuals, yet opportunities for
real-world practice remain limited. This study explores how augmented reality
(AR) can provide an accessible training environment for public speaking.
Drawing from literature on public speaking, VR-based training, self-efficacy,
and behavioral feedback mechanisms, we designed SpeakAR, an AR-based tool that
simulates audience interaction through virtual models. SpeakAR was evaluated
with five participants of varying anxiety levels, each completing six speaking
tasks. Results indicate that AR exposure can enhance confidence, with
participants finding the system useful for practice. Feedback highlighted the
importance of dynamic facial expressions and idle animations in virtual models
to improve realism and engagement. Our findings contribute to the design of
AR-based training tools for public speaking, offering insights into how
immersive environments can support skill development and anxiety reduction.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:53:30 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Jim",
"Mark Edison",
""
],
[
"Yap",
"Jan Benjamin",
""
],
[
"Laolao",
"Gian Chill",
""
],
[
"Lim",
"Andrei Zachary",
""
],
[
"Deja",
"Jordan Aiko",
""
]
] |
2504.11381 | Juan Diego Rodriguez | Juan Diego Rodriguez, Wenxuan Ding, Katrin Erk, Greg Durrett | RankAlign: A Ranking View of the Generator-Validator Gap in Large
Language Models | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Although large language models (LLMs) have become generally more capable and
accurate across many tasks, some fundamental sources of unreliability remain in
their behavior. One key limitation is their inconsistency at reporting the the
same information when prompts are changed. In this paper, we consider the
discrepancy between a model's generated answer and their own verification of
that answer, the generator-validator gap. We define this gap in a more
stringent way than prior work: we expect correlation of scores from a generator
and a validator over the entire set of candidate answers. We show that
according to this measure, a large gap exists in various settings, including
question answering, lexical semantics tasks, and next-word prediction. We then
propose RankAlign, a ranking-based training method, and show that it
significantly closes the gap by 31.8% on average, surpassing all baseline
methods. Moreover, this approach generalizes well to out-of-domain tasks and
lexical items.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:53:31 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Rodriguez",
"Juan Diego",
""
],
[
"Ding",
"Wenxuan",
""
],
[
"Erk",
"Katrin",
""
],
[
"Durrett",
"Greg",
""
]
] |
2504.11382 | Guillaume Olikier | Guillaume Olikier, Petar Mlinari\'c, P.-A. Absil, and Andr\'e
Uschmajew | The tangent cone to the real determinantal variety: various expressions
and four proofs | null | null | null | null | math.OC math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The set of real matrices of upper-bounded rank is a real algebraic variety
called the real generic determinantal variety. An explicit description of the
tangent cone to that variety is given in Theorem 3.2 of Schneider and Uschmajew
[SIAM J. Optim., 25 (2015), pp. 622-646]. The present paper shows that the
proof therein is incomplete and provides four proofs. It also reviews
equivalent descriptions of the tangent cone to that variety. Moreover, it shows
that the tangent cone and the algebraic tangent cone to that variety coincide,
which is not true for all real algebraic varieties.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:53:48 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Olikier",
"Guillaume",
""
],
[
"Mlinarić",
"Petar",
""
],
[
"Absil",
"P. -A.",
""
],
[
"Uschmajew",
"André",
""
]
] |
2504.11384 | Noam Soker | Noam Soker (Technion, Israel) | The morphology of supernova remnant G0.9+0.1 implies explosion by
jittering-jets | Will be submitted in two days to allow for comments (including
missing references) | null | null | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | I examine the morphology of the core-collapse supernova (CCSN) remnant (SNR)
G0.9+0.1 and reveal a point-symmetrical morphology that implies shaping by
three or more pairs of jets, as expected in the jittering jets explosion
mechanism (JJEM). The large northwest protrusion, the ear (or lobe), has two
bright rims. I compare this ear with its rims to an ear with three rims of a
jet-shaped planetary nebula and jets from an active galactic nucleus that
shaped several rims on one side. Based on this similarity, I argue that two
jets or more shaped the northwest ear of SNR G0.9+0.1 and its two rims. I
identified the bright region south of the main shell of SNR G0.9+0.1 as a
jet-shaped blowout formed by a jet that broke out from the main SNR shell. I
base this on the similarity of the blowout of SNR G0.9+0.1 with that of SNR
G309.2-00.6, argued in the past to be shaped by jets. I identify four symmetry
axes along different directions that compose the point-symmetric morphology of
SNR G0.9+0.1. I show that the morphological features of holes, granular
texture, and random filaments exist in CCSNe and planetary nebulae and are
unlikely to result from some unique processes in CCSNe. These structures result
from similar instabilities in the JJEM and the neutrino-driven explosion
mechanism and, unlike a point-symmetric morphology, cannot determine the
explosion mechanism. Identifying SNR G0.9+0.1 as a new point-symmetric CCSN
strengthens the JJEM as the primary explosion mechanism of CCSNe.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:54:04 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Soker",
"Noam",
"",
"Technion, Israel"
]
] |
2504.11385 | Yitian Qian | Yitian Qian, Ting Tao, Shaohua Pan, Houduo Qi | GLL-type Nonmonotone Descent Methods Revisited under
Kurdyka-{\L}ojasiewicz Property | null | null | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The purpose of this paper is to extend the full convergence results of the
classic GLL-type (Grippo-Lampariello-Lucidi) nonmonotone methods to nonconvex
and nonsmooth optimization. We propose a novel iterative framework for the
minimization of a proper and lower semicontinuous function $\Phi$. The
framework consists of the GLL-type nonmonotone decrease condition for a
sequence, a relative error condition for its augmented sequence with respect to
a Kurdyka-{\L}ojasiewicz (KL) function $\Theta$, and a relative gap condition
for the partial maximum objective value sequence. The last condition is shown
to be a product of the prox-regularity of $\Phi$ on the set of cluster points,
and to hold automatically under a mild condition on the objective value
sequence. We prove that for any sequence and its bounded augmented sequence
together falling within the framework, the sequence itself is convergent.
Furthermore, when $\Theta$ is a KL function of exponent $\theta\in(0, 1)$, the
convergence admits a linear rate if $\theta\in(0, 1/2]$ and a sublinear rate if
$\theta\in(1/2, 1)$. As applications, we prove, for the first time, that the
two existing algorithms, namely the nonmonotone proximal gradient (NPG) method
with majorization and NPG with extrapolation both enjoy the full convergence of
the iterate sequences for nonconvex and nonsmooth KL composite optimization
problems.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:54:25 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Qian",
"Yitian",
""
],
[
"Tao",
"Ting",
""
],
[
"Pan",
"Shaohua",
""
],
[
"Qi",
"Houduo",
""
]
] |
2504.11386 | Jiafeng Xiong | Jiafeng Xiong, Rizos Sakellariou | Trajectory Encoding Temporal Graph Networks | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Temporal Graph Networks (TGNs) have demonstrated significant success in
dynamic graph tasks such as link prediction and node classification. Both tasks
comprise transductive settings, where the model predicts links among known
nodes, and in inductive settings, where it generalises learned patterns to
previously unseen nodes. Existing TGN designs face a dilemma under these dual
scenarios. Anonymous TGNs, which rely solely on temporal and structural
information, offer strong inductive generalisation but struggle to distinguish
known nodes. In contrast, non-anonymous TGNs leverage node features to excel in
transductive tasks yet fail to adapt to new nodes. To address this challenge,
we propose Trajectory Encoding TGN (TETGN). Our approach introduces
automatically expandable node identifiers (IDs) as learnable temporal
positional features and performs message passing over these IDs to capture each
node's historical context. By integrating this trajectory-aware module with a
standard TGN using multi-head attention, TETGN effectively balances
transductive accuracy with inductive generalisation. Experimental results on
three real-world datasets show that TETGN significantly outperforms strong
baselines on both link prediction and node classification tasks, demonstrating
its ability to unify the advantages of anonymous and non-anonymous models for
dynamic graph learning.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:57:09 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Xiong",
"Jiafeng",
""
],
[
"Sakellariou",
"Rizos",
""
]
] |
2504.11387 | Andrea Pedicone | Andrea Pedicone and Enzo Orsingher | On the distribution of the telegraph meander and its properties | 28 pages | null | null | null | math.PR | http://creativecommons.org/licenses/by/4.0/ | In this paper we present the distribution of the telegraph meander, a random
function obtained by conditioning the telegraph process to stay above the zero
level. The reflection principle for finite-velocity random motions allows the
law of the telegraph meander to be expressed in terms of the spatial derivative
of the law of the telegraph process with initial negative velocity. As a
result, we are able to obtain the characteristic function, the moments and the
hyperbolic equation that governs the law of the telegraph meander. Furthermore,
we prove that Brownian meander is the weak limit of the telegraph meander.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:57:41 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Pedicone",
"Andrea",
""
],
[
"Orsingher",
"Enzo",
""
]
] |
2504.11390 | Rozie Zangeneh | Rozie Zangeneh | Unraveling Momentum and Heat Intercoupling in Reattaching Turbulent
Boundary Layers Using Dynamic Mode Decomposition | null | null | null | null | physics.flu-dyn | http://creativecommons.org/publicdomain/zero/1.0/ | Dynamic mode decomposition method is deployed to investigate the heat
transfer mechanism in a compressible turbulent shear layer and shockwave. To
this end, highly resolved Large Eddy Simulations are performed to explore the
effect of wall thermal conditions on the behavior of a reattaching free shear
layer interacting with an oblique shock in compressible turbulent flows.
Various different wall temperature conditions, such as cold adiabatic and hot
wall, are considered. Dynamic mode decomposition is used to isolate and study
the structures generated by the shear layer exposed in the boundary layer.
Results reveal that the shear layer flapping is the most energetic mode. The
hot wall gains the highest amplitude for the flapping frequency, and the
vortical motions are most intense in the vicinity of the reattachment point of
the heated wall. The vortex shedding due to the large-scale motion of the shear
layer is associated with the second energetic mode. The cold wall not only has
a higher amplitude of the shedding mode, but it also has a lower frequency
compared to the adiabatic and hot walls. This work sheds light on the
underlying physics of the nonlinear intercoupling of momentum and heat, hence
providing guidelines for designing control systems for high speed flight
vehicles and mitigating aircraft fatigue loading caused by intense wall
pressure fluctuations and heat flux.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 16:59:10 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Zangeneh",
"Rozie",
""
]
] |
2504.11391 | Stefan Dittmaier | Stefan Dittmaier, Tim Engel, Jose Luis Hernando Ariza, Mathieu Pellen | Electroweak corrections to $\tau^+\tau^-$ production in ultraperipheral
heavy-ion collisions at the LHC | 42 pages, latex, 11 figures | null | null | FR-PHENO-2025-004 | hep-ph | http://creativecommons.org/licenses/by/4.0/ | While the anomalous magnetic moments of the electron and the muon have been
measured with remarkable precision, the magnetic moment of the $\tau$-lepton is
only known to rather limited precision. A promising approach to measure it
exploits $\tau^+\tau^-$ production in ultraperipheral collisions of lead ions
at the LHC. In this article, a state-of-the-art theory prediction for
$\tau^+\tau^-$ production including leptonic $\tau$-decays is provided. The
impact of spin correlations between the $\tau$-leptons, of the masses of
final-state leptons, of next-to-leading-order electroweak corrections, and of
the parametrization of the photon flux are discussed.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:00:47 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Dittmaier",
"Stefan",
""
],
[
"Engel",
"Tim",
""
],
[
"Ariza",
"Jose Luis Hernando",
""
],
[
"Pellen",
"Mathieu",
""
]
] |
2504.11393 | Ian Magnusson | Ian Magnusson, Nguyen Tai, Ben Bogin, David Heineman, Jena D. Hwang,
Luca Soldaini, Akshita Bhagia, Jiacheng Liu, Dirk Groeneveld, Oyvind Tafjord,
Noah A. Smith, Pang Wei Koh, Jesse Dodge | DataDecide: How to Predict Best Pretraining Data with Small Experiments | null | null | null | null | cs.LG cs.CL | http://creativecommons.org/licenses/by/4.0/ | Because large language models are expensive to pretrain on different
datasets, using smaller-scale experiments to decide on data is crucial for
reducing costs. Which benchmarks and methods of making decisions from observed
performance at small scale most accurately predict the datasets that yield the
best large models? To empower open exploration of this question, we release
models, data, and evaluations in DataDecide -- the most extensive open suite of
models over differences in data and scale. We conduct controlled pretraining
experiments across 25 corpora with differing sources, deduplication, and
filtering up to 100B tokens, model sizes up to 1B parameters, and 3 random
seeds. We find that the ranking of models at a single, small size (e.g., 150M
parameters) is a strong baseline for predicting best models at our larger
target scale (1B) (~80% of com parisons correct). No scaling law methods among
8 baselines exceed the compute-decision frontier of single-scale predictions,
but DataDecide can measure improvement in future scaling laws. We also identify
that using continuous likelihood metrics as proxies in small experiments makes
benchmarks including MMLU, ARC, HellaSwag, MBPP, and HumanEval >80% predictable
at the target 1B scale with just 0.01% of the compute.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:02:15 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Magnusson",
"Ian",
""
],
[
"Tai",
"Nguyen",
""
],
[
"Bogin",
"Ben",
""
],
[
"Heineman",
"David",
""
],
[
"Hwang",
"Jena D.",
""
],
[
"Soldaini",
"Luca",
""
],
[
"Bhagia",
"Akshita",
""
],
[
"Liu",
"Jiacheng",
""
],
[
"Groeneveld",
"Dirk",
""
],
[
"Tafjord",
"Oyvind",
""
],
[
"Smith",
"Noah A.",
""
],
[
"Koh",
"Pang Wei",
""
],
[
"Dodge",
"Jesse",
""
]
] |
2504.11394 | Jim Coykendall | Jason Boynton, Jim Coykendall, Grant Moles, Chelsey Morrow | Overrings of half-factorial orders | 11 pages | null | null | null | math.AC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The behavior of factorization properties in various ring extensions is a
central theme in commutative algebra. Classically, the UFDs are (completely)
integrally closed and tend to behave well in standard ring extensions, with the
notable exception of power series extension. The half-factorial property is not
as robust; HFDs need not be integrally closed and the half-factorial property
is not necessarily preserved in integral extensions or even localizations. Here
we exhibit classes of HFDs that behave well in (almost) integral extensions,
resolve an open question on the behavior of the boundary map, and give a
squeeze theorem for elasticity in certain domains.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:05:28 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Boynton",
"Jason",
""
],
[
"Coykendall",
"Jim",
""
],
[
"Moles",
"Grant",
""
],
[
"Morrow",
"Chelsey",
""
]
] |
2504.11395 | Xiongxun Huang | Xiongxun Huang and Yonglu Shu | The frequent hypercyclicity of unbounded operator | null | null | null | null | math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We establish two Frequent Hypercyclicity Criteria for unbounded operators,
inspired by the frameworks of Bayart Grivaux and deLaubenfels Emamirad Grosse
Erdmann. These criteria simplify the verification and construction of
frequently hypercyclic operators.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:06:03 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Huang",
"Xiongxun",
""
],
[
"Shu",
"Yonglu",
""
]
] |
2504.11397 | Xingjian Li | Raghav Pant, Sikan Li, Xingjian Li, Hassan Iqbal, Krishna Kumar | MLPs and KANs for data-driven learning in physical problems: A
performance comparison | 30 pages, 18 figures, 8 tables | null | null | null | cs.LG physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | There is increasing interest in solving partial differential equations (PDEs)
by casting them as machine learning problems. Recently, there has been a spike
in exploring Kolmogorov-Arnold Networks (KANs) as an alternative to traditional
neural networks represented by Multi-Layer Perceptrons (MLPs). While showing
promise, their performance advantages in physics-based problems remain largely
unexplored. Several critical questions persist: Can KANs capture complex
physical dynamics and under what conditions might they outperform traditional
architectures? In this work, we present a comparative study of KANs and MLPs
for learning physical systems governed by PDEs. We assess their performance
when applied in deep operator networks (DeepONet) and graph network-based
simulators (GNS), and test them on physical problems that vary significantly in
scale and complexity. Drawing inspiration from the Kolmogorov Representation
Theorem, we examine the behavior of KANs and MLPs across shallow and deep
network architectures. Our results reveal that although KANs do not
consistently outperform MLPs when configured as deep neural networks, they
demonstrate superior expressiveness in shallow network settings, significantly
outpacing MLPs in accuracy over our test cases. This suggests that KANs are a
promising choice, offering a balance of efficiency and accuracy in applications
involving physical systems.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:13:42 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Pant",
"Raghav",
""
],
[
"Li",
"Sikan",
""
],
[
"Li",
"Xingjian",
""
],
[
"Iqbal",
"Hassan",
""
],
[
"Kumar",
"Krishna",
""
]
] |
2504.11398 | Iman Gholami | Ali Ahmadi, Iman Gholami, MohammadTaghi Hajiaghayi, Peyman Jabbarzade,
Mohammad Mahdavi | Breaking a Long-Standing Barrier: 2-$\varepsilon$ Approximation for
Steiner Forest | null | null | null | null | cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Steiner Forest problem, also known as the Generalized Steiner Tree
problem, is a fundamental optimization problem on edge-weighted graphs where,
given a set of vertex pairs, the goal is to select a minimum-cost subgraph such
that each pair is connected. This problem generalizes the Steiner Tree problem,
first introduced in 1811, for which the best known approximation factor is 1.39
[Byrka, Grandoni, Rothvo{\ss}, and Sanit\`a, 2010] (Best Paper award, STOC
2010).
The celebrated work of [Agrawal, Klein, and Ravi, 1989] (30-Year Test-of-Time
award, STOC 2023), along with refinements by [Goemans and Williamson, 1992]
(SICOMP'95), established a 2-approximation for Steiner Forest over 35 years
ago. Jain's (FOCS'98) pioneering iterative rounding techniques later extended
these results to higher connectivity settings. Despite the long-standing
importance of this problem, breaking the approximation factor of 2 has remained
a major challenge, raising suspicions that achieving a better factor -- similar
to Vertex Cover -- might indeed be hard. Notably, fundamental works, including
those by Gupta and Kumar (STOC'15) and Gro{\ss} et al. (ITCS'18), introduced
96- and 69-approximation algorithms, possibly with the hope of paving the way
for a breakthrough in achieving a constant-factor approximation below 2 for the
Steiner Forest problem.
In this paper, we break the approximation barrier of 2 by designing a novel
deterministic algorithm that achieves a $2 - 10^{-11}$ approximation for this
fundamental problem. As a key component of our approach, we also introduce a
novel dual-based local search algorithm for the Steiner Tree problem with an
approximation guarantee of $1.943$, which is of independent interest.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:13:48 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Ahmadi",
"Ali",
""
],
[
"Gholami",
"Iman",
""
],
[
"Hajiaghayi",
"MohammadTaghi",
""
],
[
"Jabbarzade",
"Peyman",
""
],
[
"Mahdavi",
"Mohammad",
""
]
] |
2504.11399 | Phalgun Lolur | Marek Kowalik, Ellen Michael, Peter Pog\'any, Phalgun Lolur | Assessing Tensor Network Quantum Emulators for Hamiltonian Simulation of
Pharmaceutical Molecules: Challenges and Limitations in Drug Discovery
Applications | 31 pages, 15 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum computing holds promise for revolutionizing computational chemistry
simulations, particularly in drug discovery. However, current quantum hardware
is limited by noise and scale, necessitating bridging technologies. This study
provides an initial evaluation of tensor network quantum emulators, narrowed to
matrix product state-based emulators, for Hamiltonian simulation of
pharmaceutical molecules, with a focus on predicting the reactivity of targeted
covalent drugs. We assess runtime scaling, accuracy, and resource requirements
across various active space sizes, comparing performance to traditional state
vector simulation methods. Our results reveal that, for accurate estimation of
the expectation value trajectory of a key measurement operator - used as a
quantum-derived feature for reactivity prediction - the required bond dimension
in matrix product state tensor networks grows rapidly with system size,
effectively negating runtime advantages for larger, chemically relevant
molecules. This study highlights the fundamental challenges in classically
simulating complex quantum chemistry systems and contributes to the support of
the irreplaceability premise of quantum computers to efficiently handle
strongly entangled systems. Such robustness of fault-tolerant quantum computers
leads to practical advantages in drug discovery applications.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:14:01 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Kowalik",
"Marek",
""
],
[
"Michael",
"Ellen",
""
],
[
"Pogány",
"Peter",
""
],
[
"Lolur",
"Phalgun",
""
]
] |
2504.11400 | Alessandro Margara | Fabio Chini, Luca De Martini, Alessandro Margara, Gianpaolo Cugola | FlowUnits: Extending Dataflow for the Edge-to-Cloud Computing Continuum | Preprint. Accepted at the 2nd Workshop on Engineering Techniques for
Distributed Computing Continuum Systems (EDCCS), co-located with IEEE ICDCS
2025 | null | null | null | cs.DC cs.SE | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This paper introduces FlowUnits, a novel programming and deployment model
that extends the traditional dataflow paradigm to address the unique challenges
of edge-to-cloud computing environments. While conventional dataflow systems
offer significant advantages for large-scale data processing in homogeneous
cloud settings, they fall short when deployed across distributed, heterogeneous
infrastructures. FlowUnits addresses three critical limitations of current
approaches: lack of locality awareness, insufficient resource adaptation, and
absence of dynamic update mechanisms. FlowUnits organize processing operators
into cohesive, independently manageable components that can be transparently
replicated across different regions, efficiently allocated on nodes with
appropriate hardware capabilities, and dynamically updated without disrupting
ongoing computations. We implement and evaluate the FlowUnits model within
Renoir, an existing dataflow system, demonstrating significant improvements in
deployment flexibility and resource utilization across the computing continuum.
Our approach maintains the simplicity of dataflow while enabling seamless
integration of edge and cloud resources into unified data processing pipelines.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:14:08 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Chini",
"Fabio",
""
],
[
"De Martini",
"Luca",
""
],
[
"Margara",
"Alessandro",
""
],
[
"Cugola",
"Gianpaolo",
""
]
] |
2504.11401 | Saksham Sharma | Giuseppe De Giorgio, Pasquale D'Angelo, Giuseppe Tarabella, Saksham
Sharma | Advances in Prebiotic Chemistry: the potential of Analog Computing and
Navier-Stokes Nernst-Planck (NPNS) Modeling in Organic Electronics
Technologies (OECTs) | 10 pages, 3 figures | null | null | null | physics.flu-dyn physics.chem-ph | http://creativecommons.org/licenses/by/4.0/ | In this article, we attempt to make a conceptual bridge between the research
in biology, pre-biotic chemistry, biomimetics, and the tools used in organic
bioelectronics in terms of materials and devices. The goal is discussing how
materials and devices of organic bioelectronics can be exploited and used at
the interface with biology, but also how, and at what extent, they can be
adapted to mimicking nature-inspired properties, herein including
unconventional computing strategies. The idea is to provide new hints and solid
hypotheses for designing niche experiments that could benefit from a proper
interaction, even at a basic communicative level, between materials science and
biotechnology. The finale long-term vision goal being the vision of collecting
experimental data that may help to made a step forward toward the
implementation of the transition from inanimate objects to animated beings. The
mathematical model canonically considered in this work is the
Navier-Stokes-Nernst-Planck (NPNS) Model which is often used to model a charged
continuum system such as the organic electrochemical transistors.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:17:45 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"De Giorgio",
"Giuseppe",
""
],
[
"D'Angelo",
"Pasquale",
""
],
[
"Tarabella",
"Giuseppe",
""
],
[
"Sharma",
"Saksham",
""
]
] |
2504.11404 | Semhar Michael | Andrew Simpson and Semhar Michael | Statistical few-shot learning for large-scale classification via
parameter pooling | null | null | null | null | stat.ME | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In large-scale few-shot learning for classification problems, often there are
a large number of classes and few high-dimensional observations per class.
Previous model-based methods, such as Fisher's linear discriminant analysis
(LDA), require the strong assumptions of a shared covariance matrix between all
classes. Quadratic discriminant analysis will often lead to singular or
unstable covariance matrix estimates. Both of these methods can lead to
lower-than-desired classification performance. We introduce a novel,
model-based clustering method that can relax the shared covariance assumptions
of LDA by clustering sample covariance matrices, either singular or
non-singular. In addition, we study the statistical properties of parameter
estimates. This will lead to covariance matrix estimates which are pooled
within each cluster of classes. We show, using simulated and real data, that
our classification method tends to yield better discrimination compared to
other methods.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:19:12 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Simpson",
"Andrew",
""
],
[
"Michael",
"Semhar",
""
]
] |
2504.11405 | Rogerio Deienno | Rogerio Deienno, Andre Izidoro, David Nesvorny, William F. Bottke,
Fernando Roig, Simone Marchi | Size-Frequency Distribution of Terrestrial Leftover Planetesimals and
S-complex Implanted Asteroids | Accepted for publication in The Astrophysical Journal, 13 pages
(single column), 5 figures | null | null | null | astro-ph.EP | http://creativecommons.org/licenses/by/4.0/ | The isotopic composition of meteorites linked to S-complex asteroids has been
used to suggest that these asteroids originated in the terrestrial planet's
region, i.e., within 1.5 au, and later got implanted into the main asteroid
belt (MAB). Dynamical models of planet formation support this view. Yet, it
remains to be demonstrated whether the currently observed size-frequency
distribution (SFD) of S-complex bodies in the MAB can be reproduced via this
implantation process. Here we studied the evolution of the SFD of planetesimals
during the accretion of terrestrial planets with the code LIPAD
self-consistently accounting for growth and fragmentation of planetesimals. In
our simulations we vary the initial surface density of planetesimals, the
gaseous disk lifetime, and the power slope of the initial planetesimals' SFD.
We compared the final SFDs of leftover planetesimals in the terrestrial planet
region with the SFD of observed S-complex MAB objects (D $>$ 100km). We found
that the SFDs of our planetesimal populations and that of S-complex MAB objects
show very similar cumulative power index (i.e., q $\approx$ 3.15 in
N($>$D)$~\propto$ D$^{-q}$) for slopes in the diameter range 100 km $<$ D $<$
400 km by the end of our simulations. Our results support the hypothesis of
S-complex MAB implantation from the terrestrial planet forming region, assuming
implantation is size-independent, and implies that implantation efficiency is
smaller than $\mathcal{O}$(10$^{\rm -2}$--10$^{\rm -4}$) to avoid
over-implantation of (4) Vesta-sized objects or larger.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:19:57 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Deienno",
"Rogerio",
""
],
[
"Izidoro",
"Andre",
""
],
[
"Nesvorny",
"David",
""
],
[
"Bottke",
"William F.",
""
],
[
"Roig",
"Fernando",
""
],
[
"Marchi",
"Simone",
""
]
] |
2504.11406 | Felipe Salvagnini | Felipe Crispim Salvagnini, Jancarlo F. Gomes, Cid A. N. Santos, Silvio
Jamil F. Guimar\~aes, Alexandre X. Falc\~ao | Multi-level Cellular Automata for FLIM networks | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | The necessity of abundant annotated data and complex network architectures
presents a significant challenge in deep-learning Salient Object Detection
(deep SOD) and across the broader deep-learning landscape. This challenge is
particularly acute in medical applications in developing countries with limited
computational resources. Combining modern and classical techniques offers a
path to maintaining competitive performance while enabling practical
applications. Feature Learning from Image Markers (FLIM) methodology empowers
experts to design convolutional encoders through user-drawn markers, with
filters learned directly from these annotations. Recent findings demonstrate
that coupling a FLIM encoder with an adaptive decoder creates a flyweight
network suitable for SOD, requiring significantly fewer parameters than
lightweight models and eliminating the need for backpropagation. Cellular
Automata (CA) methods have proven successful in data-scarce scenarios but
require proper initialization -- typically through user input, priors, or
randomness. We propose a practical intersection of these approaches: using FLIM
networks to initialize CA states with expert knowledge without requiring user
interaction for each image. By decoding features from each level of a FLIM
network, we can initialize multiple CAs simultaneously, creating a multi-level
framework. Our method leverages the hierarchical knowledge encoded across
different network layers, merging multiple saliency maps into a high-quality
final output that functions as a CA ensemble. Benchmarks across two challenging
medical datasets demonstrate the competitiveness of our multi-level CA approach
compared to established models in the deep SOD literature.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:22:24 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Salvagnini",
"Felipe Crispim",
""
],
[
"Gomes",
"Jancarlo F.",
""
],
[
"Santos",
"Cid A. N.",
""
],
[
"Guimarães",
"Silvio Jamil F.",
""
],
[
"Falcão",
"Alexandre X.",
""
]
] |
2504.11407 | Alessandro Montinaro | Alessandro Montinaro | The Higman-M\lowercase{c}Laughlin Theorem for the flag-transitive
$2$-designs with $\lambda$ prime | null | null | null | null | math.CO math.GR | http://creativecommons.org/licenses/by/4.0/ | A famous result of Higman and McLaughlin \cite{HM} in 1961 asserts that any
flag-transitive automorphism group $G$ of a $2$-design $\mathcal{D}$ with
$\lambda=1$ acts point-primitively on $\mathcal{D}$. In this paper, we show
that the Higman and McLaughlin theorem is still true when $\lambda$ is a prime
and $\mathcal{D}$ is not isomorphic to one of the two $2$-$(16,6,2)$ designs as
in [42, Section 1.2], or the $2$-$(45,12,3)$ design as in [44, Construction
4.2], or, when $2^{2^{j}}+1$ is a Fermat prime, a possible
$2$-$(2^{2^{j+1}}(2^{2^{j}}+2),2^{2^{j}}(2^{2^{j}}+1),2^{2^{j}}+1)$ design
having very specific features.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:23:58 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Montinaro",
"Alessandro",
""
]
] |
2504.11408 | Milko Estrada | Milko Estrada and L. C. N. Santos and M. S. Cunha | Five dimensional rotating and Quintessence black hole and their shadows | null | null | null | null | gr-qc | http://creativecommons.org/licenses/by-sa/4.0/ | We present a new five-dimensional rotating quintessence black hole solution.
To obtain this, we employ the $5D$ version of the Janis Newman algorithm, which
incorporates the Hopf bifurcation. The variation of the quintessence parameter
$w_q$ causes the geometry to transition from a regular rotating universe
surrounded by a cosmological horizon to a singular rotating geometry, which can
represent a naked singularity, a singular extremal black hole, or a singular
black hole with both an inner and an outer (event) horizon. We have also
determined the properties of the ergosphere. For the study of the shadow, we
followed a novel approach in which the $2D$ shadow observed by humans
corresponds to cross-sections of the $3D$ shadow. We analyzed how quintessence
affects both the size and shape of the black hole shadow, showing that
increasing the quintessence strength reduces the shadow radius, contrary to the
known results in $4D$. We also propose a speculative methodology to test the
shadow behavior in five-dimensional scenarios, in light of the constraints
provided by the Event Horizon Telescope (EHT) concerning the shadow of the
four-dimensional supermassive black hole M87. We identify scenarios in which
the theoretical $5D$ results could be consistent with these observational
constraints. Finally, we determine the energy conditions required to support
the solution.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:26:18 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Estrada",
"Milko",
""
],
[
"Santos",
"L. C. N.",
""
],
[
"Cunha",
"M. S.",
""
]
] |
2504.11409 | Ali Taghibakhsh | Ali Taghibakhshi, Sharath Turuvekere Sreenivas, Saurav Muralidharan,
Marcin Chochowski, Yashaswi Karnati, Raviraj Joshi, Ameya Sunil
Mahabaleshwarkar, Zijia Chen, Yoshi Suhara, Oluwatobi Olabiyi, Daniel
Korzekwa, Mostofa Patwary, Mohammad Shoeybi, Jan Kautz, Bryan Catanzaro,
Ashwath Aithal, Nima Tajbakhsh, Pavlo Molchanov | Efficient Hybrid Language Model Compression through Group-Aware SSM
Pruning | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Hybrid LLM architectures that combine Attention and State Space Models (SSMs)
achieve state-of-the-art accuracy and runtime performance. Recent work has
demonstrated that applying compression and distillation to Attention-only
models yields smaller, more accurate models at a fraction of the training cost.
In this work, we explore the effectiveness of compressing Hybrid architectures.
We introduce a novel group-aware pruning strategy that preserves the structural
integrity of SSM blocks and their sequence modeling capabilities. Furthermore,
we demonstrate the necessity of such SSM pruning to achieve improved accuracy
and inference speed compared to traditional approaches. Our compression recipe
combines SSM, FFN, embedding dimension, and layer pruning, followed by
knowledge distillation-based retraining, similar to the MINITRON technique.
Using this approach, we compress the Nemotron-H 8B Hybrid model down to 4B
parameters with up to 40x fewer training tokens. The resulting model surpasses
the accuracy of similarly-sized models while achieving 2x faster inference,
significantly advancing the Pareto frontier.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:26:29 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Taghibakhshi",
"Ali",
""
],
[
"Sreenivas",
"Sharath Turuvekere",
""
],
[
"Muralidharan",
"Saurav",
""
],
[
"Chochowski",
"Marcin",
""
],
[
"Karnati",
"Yashaswi",
""
],
[
"Joshi",
"Raviraj",
""
],
[
"Mahabaleshwarkar",
"Ameya Sunil",
""
],
[
"Chen",
"Zijia",
""
],
[
"Suhara",
"Yoshi",
""
],
[
"Olabiyi",
"Oluwatobi",
""
],
[
"Korzekwa",
"Daniel",
""
],
[
"Patwary",
"Mostofa",
""
],
[
"Shoeybi",
"Mohammad",
""
],
[
"Kautz",
"Jan",
""
],
[
"Catanzaro",
"Bryan",
""
],
[
"Aithal",
"Ashwath",
""
],
[
"Tajbakhsh",
"Nima",
""
],
[
"Molchanov",
"Pavlo",
""
]
] |
2504.11410 | David Torregrosa-Bel\'en | Pedro P\'erez-Aros, David Torregrosa-Bel\'en | Randomized block proximal method with locally Lipschitz continuous
gradient | null | null | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Block-coordinate algorithms are recognized to furnish efficient iterative
schemes for addressing large-scale problems, especially when the computation of
full derivatives entails substantial memory requirements and computational
efforts. In this paper, we investigate a randomized block proximal gradient
algorithm for minimizing the sum of a differentiable function and a separable
proper lower-semicontinuous function, both possibly nonconvex. In contrast to
previous works, we only assume that the partial gradients of the differentiable
function are locally Lipschitz continuous. At each iteration, the method
adaptively selects a proximal stepsize to satisfy a sufficient decrease
condition without prior knowledge of the local Lipschitz moduli of the partial
gradients of the differentiable function. In addition, we incorporate the
possibility of conducting an additional linesearch to enhance the performance
of the algorithm. Our main result establishes subsequential convergence to a
stationary point of the problem almost surely. Finally, we provide numerical
validation of the method in an experiment in image compression using a
nonnegative matrix factorization model.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:26:40 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Pérez-Aros",
"Pedro",
""
],
[
"Torregrosa-Belén",
"David",
""
]
] |
2504.11411 | Khac-Hoang Ngo | Khac-Hoang Ngo and Erik G. Larsson | Breaking the TDD Flow for Over-the-Air Phase Synchronization in
Distributed Antenna Systems | submitted to IEEE GLOBECOM 2025 | null | null | null | cs.IT eess.SP math.IT | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Phase synchronization between distributed antenna arrays requires
measurements that break the standard time-division duplex (TDD) operation. We
present a feasibility study on implementing such synchronization and analyze
its impact on the quality of service. Considering two antenna arrays with
independent local oscillators (LOs), we propose a modified TDD flow to
accommodate the transmission of phase synchronization signals, formulate the
phase estimation and compensation problem, and derive the achievable downlink
spectral efficiency (SE). Numerical results show that frequent re-estimation of
the interarray phase disparity is essential for maximizing SE in systems with
low-quality LOs. Furthermore, applying a Kalman filter for phase tracking
substantially improves the SE, especially if phase estimation errors are large
compared to LOs phase drifts.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:26:42 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Ngo",
"Khac-Hoang",
""
],
[
"Larsson",
"Erik G.",
""
]
] |
2504.11412 | Yudong Luo | Yudong Luo, Yangchen Pan, Jiaqi Tan, Pascal Poupart | Measures of Variability for Risk-averse Policy Gradient | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Risk-averse reinforcement learning (RARL) is critical for decision-making
under uncertainty, which is especially valuable in high-stake applications.
However, most existing works focus on risk measures, e.g., conditional
value-at-risk (CVaR), while measures of variability remain underexplored. In
this paper, we comprehensively study nine common measures of variability,
namely Variance, Gini Deviation, Mean Deviation, Mean-Median Deviation,
Standard Deviation, Inter-Quantile Range, CVaR Deviation, Semi_Variance, and
Semi_Standard Deviation. Among them, four metrics have not been previously
studied in RARL. We derive policy gradient formulas for these unstudied
metrics, improve gradient estimation for Gini Deviation, analyze their gradient
properties, and incorporate them with the REINFORCE and PPO frameworks to
penalize the dispersion of returns.
Our empirical study reveals that variance-based metrics lead to unstable
policy updates. In contrast, CVaR Deviation and Gini Deviation show consistent
performance across different randomness and evaluation domains, achieving high
returns while effectively learning risk-averse policies. Mean Deviation and
Semi_Standard Deviation are also competitive across different scenarios. This
work provides a comprehensive overview of variability measures in RARL,
offering practical insights for risk-aware decision-making and guiding future
research on risk metrics and RARL algorithms.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:28:15 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Luo",
"Yudong",
""
],
[
"Pan",
"Yangchen",
""
],
[
"Tan",
"Jiaqi",
""
],
[
"Poupart",
"Pascal",
""
]
] |
2504.11413 | Christian Kumpf | Niclas Tilgner, Susanne Wolff, Serguei Soubatch, Tien-Lin Lee, Fabian
G\"ohler, F. Stefan Tautz, Thomas Seyller, Philip Sch\"adlich, Christian
Kumpf | Solving the Phase Problem of Diffraction: X-ray Standing Waves Imaging
on Bismuthene/SiC(0001) | Five Figures, supplement with three figures | null | null | null | cond-mat.mes-hall cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | The phase retrieval problem is a fundamental shortcoming of all
diffraction-based methods, arising from the inability to measure the phase of
scattered waves. The (normal incidence) X-ray standing wave (NIXSW) technique
circumvents this issue by introducing a (Bragg-generated) X-ray standing wave
field throughout the sample, relative to which any atomic species can be
localized by probing its fluorescence or photoelectron yield. In essence, in a
single measurement the complex scattering factor (i.e., its amplitude
\textit{and} phase) corresponding to the used Bragg reflection is determined.
Performing this for multiple Bragg reflections enables one to reconstruct the
scattering density of the sample in three dimensions, straightforwardly as the
Fourier sum of all measured (complex) scattering factors. Here, we utilize this
technique to reveal the structural key features involved in the formation of
the quantum spin Hall insulator bismuthene on silicon carbide. In this
prominent example, the two-dimensional Bi layer is confined between a 4H-SiC
substrate crystal and an epitaxial graphene layer. The key finding is a change
in the adsorption site of the Bi atoms underneath the graphene upon
hydrogenation, caused by the H-saturation of one (out of three) Si dangling
bonds per unit cell. This structural change, clearly revealed by our NIXSW
imaging experiment, is the key feature leading to the formation of the
characteristic band structure of the 2D bismuthene honeycomb.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:28:24 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Tilgner",
"Niclas",
""
],
[
"Wolff",
"Susanne",
""
],
[
"Soubatch",
"Serguei",
""
],
[
"Lee",
"Tien-Lin",
""
],
[
"Göhler",
"Fabian",
""
],
[
"Tautz",
"F. Stefan",
""
],
[
"Seyller",
"Thomas",
""
],
[
"Schädlich",
"Philip",
""
],
[
"Kumpf",
"Christian",
""
]
] |
2504.11415 | Nikolette Pedersen | Nikolette Pedersen, Regitze Sydendal, Andreas Wulff, Ralf Raumanns,
Eike Petersen and Veronika Cheplygina | Robustness and sex differences in skin cancer detection: logistic
regression vs CNNs | 16 pages (excluding appendix), 2 figures (excluding appendix),
submitted to MIUA 2025 conference (response pending) | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning has been reported to achieve high performances in the detection
of skin cancer, yet many challenges regarding the reproducibility of results
and biases remain. This study is a replication (different data, same analysis)
of a study on Alzheimer's disease [28] which studied robustness of logistic
regression (LR) and convolutional neural networks (CNN) across patient sexes.
We explore sex bias in skin cancer detection, using the PAD-UFES-20 dataset
with LR trained on handcrafted features reflecting dermatological guidelines
(ABCDE and the 7-point checklist), and a pre-trained ResNet-50 model. We
evaluate these models in alignment with [28]: across multiple training datasets
with varied sex composition to determine their robustness. Our results show
that both the LR and the CNN were robust to the sex distributions, but the
results also revealed that the CNN had a significantly higher accuracy (ACC)
and area under the receiver operating characteristics (AUROC) for male patients
than for female patients. We hope these findings to contribute to the growing
field of investigating potential bias in popular medical machine learning
methods. The data and relevant scripts to reproduce our results can be found in
our Github.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:31:46 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Pedersen",
"Nikolette",
""
],
[
"Sydendal",
"Regitze",
""
],
[
"Wulff",
"Andreas",
""
],
[
"Raumanns",
"Ralf",
""
],
[
"Petersen",
"Eike",
""
],
[
"Cheplygina",
"Veronika",
""
]
] |
2504.11416 | Panagiotis Agrafiotis | Panagiotis Agrafiotis and Beg\"um Demir | Deep Learning-based Bathymetry Retrieval without In-situ Depths using
Remote Sensing Imagery and SfM-MVS DSMs with Data Gaps | Accepted for publication in ISPRS Journal of Photogrammetry and
Remote Sensing | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Accurate, detailed, and high-frequent bathymetry is crucial for shallow
seabed areas facing intense climatological and anthropogenic pressures. Current
methods utilizing airborne or satellite optical imagery to derive bathymetry
primarily rely on either SfM-MVS with refraction correction or Spectrally
Derived Bathymetry (SDB). However, SDB methods often require extensive manual
fieldwork or costly reference data, while SfM-MVS approaches face challenges
even after refraction correction. These include depth data gaps and noise in
environments with homogeneous visual textures, which hinder the creation of
accurate and complete Digital Surface Models (DSMs) of the seabed. To address
these challenges, this work introduces a methodology that combines the
high-fidelity 3D reconstruction capabilities of the SfM-MVS methods with
state-of-the-art refraction correction techniques, along with the spectral
analysis capabilities of a new deep learning-based method for bathymetry
prediction. This integration enables a synergistic approach where SfM-MVS
derived DSMs with data gaps are used as training data to generate complete
bathymetric maps. In this context, we propose Swin-BathyUNet that combines
U-Net with Swin Transformer self-attention layers and a cross-attention
mechanism, specifically tailored for SDB. Swin-BathyUNet is designed to improve
bathymetric accuracy by capturing long-range spatial relationships and can also
function as a standalone solution for standard SDB with various training depth
data, independent of the SfM-MVS output. Experimental results in two completely
different test sites in the Mediterranean and Baltic Seas demonstrate the
effectiveness of the proposed approach through extensive experiments that
demonstrate improvements in bathymetric accuracy, detail, coverage, and noise
reduction in the predicted DSM. The code is available at
https://github.com/pagraf/Swin-BathyUNet.
| [
{
"version": "v1",
"created": "Tue, 15 Apr 2025 17:31:48 GMT"
}
] | 2025-04-16T00:00:00 | [
[
"Agrafiotis",
"Panagiotis",
""
],
[
"Demir",
"Begüm",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.