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
2025-05-16 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.20929 | Evgenii Strugovshchikov | Evgenii Strugovshchikov, Viktor Mandrolko, Dominika Lesnicki,
Mariachiara Pastore, Laurent Chaput and Mykola Isaiev | Interfacial Behavior from the Atomic Blueprint: Machine Learning-Guided
Design of Spatially Functionalized a-SiO2 Surfaces | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Precise control over surface chemistry is essential for tuning interfacial
behavior in technologies ranging from catalysis and protective coatings to
energy conversion systems. Although chemical functionalization of alpha-quartz
(alpha-SiO2) with hydroxyl (OH) and methyl (CH3) groups has been extensively
studied, the impact of their spatial distribution at the atomic scale remains
largely uncharted. In this work, we integrate density functional theory (DFT),
ab initio molecular dynamics (AIMD), and on-the-fly machine-learned force
fields (MLFFs) to systematically investigate how different arrangements of
OH/CH3 groups modulate surface properties. Our results reveal that spatial
patterning governs the formation of hydrogen-bonding networks, alters
vibrational signatures, and has a significant influence on the thermodynamic
stability of the functionalized surfaces. The MLFF framework enables
high-fidelity simulations at unprecedented scales, bridging the gap between
quantum accuracy and statistical sampling. By uncovering structure-property
relationships inaccessible to conventional approaches, this study establishes
spatial arrangement of functionalized groups as a critical and tunable design
axis, paving the way for the predictive engineering of silica-based materials
with optimized interfacial performance.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:47:37 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Strugovshchikov",
"Evgenii",
""
],
[
"Mandrolko",
"Viktor",
""
],
[
"Lesnicki",
"Dominika",
""
],
[
"Pastore",
"Mariachiara",
""
],
[
"Chaput",
"Laurent",
""
],
[
"Isaiev",
"Mykola",
""
]
] |
2504.20930 | Ziqing Fan | Ziqing Fan, Cheng Liang, Chaoyi Wu, Ya Zhang, Yanfeng Wang, and Weidi
Xie | ChestX-Reasoner: Advancing Radiology Foundation Models with Reasoning
through Step-by-Step Verification | null | null | null | null | cs.AI cs.CL cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in reasoning-enhanced large language models (LLMs) and
multimodal LLMs (MLLMs) have significantly improved performance in complex
tasks, yet medical AI models often overlook the structured reasoning processes
inherent in clinical practice. In this work, we present ChestX-Reasoner, a
radiology diagnosis MLLM designed to leverage process supervision mined
directly from clinical reports, reflecting the step-by-step reasoning followed
by radiologists. We construct a large dataset by extracting and refining
reasoning chains from routine radiology reports. Our two-stage training
framework combines supervised fine-tuning and reinforcement learning guided by
process rewards to better align model reasoning with clinical standards. We
introduce RadRBench-CXR, a comprehensive benchmark featuring 59K visual
question answering samples with 301K clinically validated reasoning steps, and
propose RadRScore, a metric evaluating reasoning factuality, completeness, and
effectiveness. ChestX-Reasoner outperforms existing medical and general-domain
MLLMs in both diagnostic accuracy and reasoning ability, achieving 16%, 5.9%,
and 18% improvements in reasoning ability compared to the best medical MLLM,
the best general MLLM, and its base model, respectively, as well as 3.3%, 24%,
and 27% improvements in outcome accuracy. All resources are open-sourced to
facilitate further research in medical reasoning MLLMs.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:48:23 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Fan",
"Ziqing",
""
],
[
"Liang",
"Cheng",
""
],
[
"Wu",
"Chaoyi",
""
],
[
"Zhang",
"Ya",
""
],
[
"Wang",
"Yanfeng",
""
],
[
"Xie",
"Weidi",
""
]
] |
2504.20932 | Taisuke Kobayashi | Taisuke Kobayashi | Improvements of Dark Experience Replay and Reservoir Sampling towards
Better Balance between Consolidation and Plasticity | 29 pages, 8 figures | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Continual learning is the one of the most essential abilities for autonomous
agents, which can incrementally learn daily-life skills. For this ultimate
goal, a simple but powerful method, dark experience replay (DER), has been
proposed recently. DER mitigates catastrophic forgetting, in which the skills
acquired in the past are unintentionally forgotten, by stochastically storing
the streaming data in a reservoir sampling (RS) buffer and by relearning them
or retaining the past outputs for them. However, since DER considers multiple
objectives, it will not function properly without appropriate weighting of
them. In addition, the ability to retain past outputs inhibits learning if the
past outputs are incorrect due to distribution shift or other effects. This is
due to a tradeoff between memory consolidation and plasticity. The tradeoff is
hidden even in the RS buffer, which gradually stops storing new data for new
skills in it as data is continuously passed to it. To alleviate the tradeoff
and achieve better balance, this paper proposes improvement strategies to each
of DER and RS. Specifically, DER is improved with automatic adaptation of
weights, block of replaying erroneous data, and correction of past outputs. RS
is also improved with generalization of acceptance probability, stratification
of plural buffers, and intentional omission of unnecessary data. These
improvements are verified through multiple benchmarks including regression,
classification, and reinforcement learning problems. As a result, the proposed
methods achieve steady improvements in learning performance by balancing the
memory consolidation and plasticity.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:50:05 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Kobayashi",
"Taisuke",
""
]
] |
2504.20933 | Guanying Peng | Xavier Lamy, Andrew Lorent, Guanying Peng | Another regularizing property of the 2D eikonal equation | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A weak solution of the two-dimensional eikonal equation amounts to a vector
field $m\colon\Omega\subset\mathbb R^2\to\mathbb R^2$ such that $|m|=1$ a.e.
and $\mathrm{div}\,m=0$ in $\mathcal D'(\Omega)$. It is known that, if $m$ has
some low regularity, e.g., continuous or $W^{1/3,3}$, then $m$ is automatically
more regular: locally Lipschitz outside a locally finite set. A long-standing
conjecture by Aviles and Giga, if true, would imply the same regularizing
effect under the Besov regularity assumption $m\in B^{1/3}_{p,\infty}$ for
$p>3$. In this note we establish that regularizing effect in the borderline
case $p=6$, above which the Besov regularity assumption implies continuity. If
the domain is a disk and $m$ satisfies tangent boundary conditions, we also
prove this for $p$ slightly below $6$.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:50:37 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Lamy",
"Xavier",
""
],
[
"Lorent",
"Andrew",
""
],
[
"Peng",
"Guanying",
""
]
] |
2504.20934 | Xingzhi Zhang | Jinyan Xu, Yangye Zhou, Xingzhi Zhang, Yinshuai Li, Qinhan Tan,
Yinqian Zhang, Yajin Zhou, Rui Chang, Wenbo Shen | DejaVuzz: Disclosing Transient Execution Bugs with Dynamic Swappable
Memory and Differential Information Flow Tracking assisted Processor Fuzzing | This paper has been accepted to ASPLOS 2025 | null | null | null | cs.AR cs.CR | http://creativecommons.org/licenses/by-sa/4.0/ | Transient execution vulnerabilities have emerged as a critical threat to
modern processors. Hardware fuzzing testing techniques have recently shown
promising results in discovering transient execution bugs in large-scale
out-of-order processor designs. However, their poor microarchitectural
controllability and observability prevent them from effectively and efficiently
detecting transient execution vulnerabilities.
This paper proposes DejaVuzz, a novel pre-silicon stage processor transient
execution bug fuzzer. DejaVuzz utilizes two innovative operating primitives:
dynamic swappable memory and differential information flow tracking, enabling
more effective and efficient transient execution vulnerability detection. The
dynamic swappable memory enables the isolation of different instruction streams
within the same address space. Leveraging this capability, DejaVuzz generates
targeted training for arbitrary transient windows and eliminates ineffective
training, enabling efficient triggering of diverse transient windows. The
differential information flow tracking aids in observing the propagation of
sensitive data across the microarchitecture. Based on taints, DejaVuzz designs
the taint coverage matrix to guide mutation and uses taint liveness annotations
to identify exploitable leakages. Our evaluation shows that DejaVuzz
outperforms the state-of-the-art fuzzer SpecDoctor, triggering more
comprehensive transient windows with lower training overhead and achieving a
4.7x coverage improvement. And DejaVuzz also mitigates control flow
over-tainting with acceptable overhead and identifies 5 previously undiscovered
transient execution vulnerabilities (with 6 CVEs assigned) on BOOM and
XiangShan.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:51:55 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Xu",
"Jinyan",
""
],
[
"Zhou",
"Yangye",
""
],
[
"Zhang",
"Xingzhi",
""
],
[
"Li",
"Yinshuai",
""
],
[
"Tan",
"Qinhan",
""
],
[
"Zhang",
"Yinqian",
""
],
[
"Zhou",
"Yajin",
""
],
[
"Chang",
"Rui",
""
],
[
"Shen",
"Wenbo",
""
]
] |
2504.20935 | Matthieu Petiteau | Sylvain Gravier, Matthieu Petiteau, Isabelle Sivignon | Note about the complexity of the acyclic orientation with parity
constraint problem | null | null | null | null | cs.DM math.CO | http://creativecommons.org/licenses/by/4.0/ | Let $G = (V, E)$ be a connected graph, and let $T$ in $V$ be a subset of
vertices. An orientation of $G$ is called $T$-odd if any vertex $v \in V$ has
odd in-degree if and only if it is in $T$. Finding a T -odd orientation of G
can be solved in polynomial time as shown by Chevalier, Jaeger, Payan and Xuong
(1983). Since then, $T$-odd orientations have continued to attract interest,
particularly in the context of global constraints on the orientation. For
instance, Frank and Kir\'aly (2002) investigated $k$-connected $T$-odd
orientations and raised questions about acyclic $T$-odd orientations. This
problem is now recognized as an Egres problem and is known as the "Acyclic
orientation with parity constraints" problem. Szegedy ( 005) proposed a
randomized polynomial algorithm to address this problem. An easy consequence of
his work provides a polynomial time algorithm for planar graphs whenever $|T |
= |V | - 1$. Nevertheless, it remains unknown whether it exists in general. In
this paper we contribute to the understanding of the complexity of this problem
by studying a more general one. We prove that finding a $T$-odd acyclic
orientation on graphs having some directed edges is NP-complete.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:56:48 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Gravier",
"Sylvain",
""
],
[
"Petiteau",
"Matthieu",
""
],
[
"Sivignon",
"Isabelle",
""
]
] |
2504.20936 | You Wang | You Wang and Yunhe Sheng | Phase space of a Poisson algebra and the induced Pre-Poisson bialgebra | null | null | null | null | math-ph math.MP math.RA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we first introduce the notion of a phase space of a Poisson
algebra, and show that a Poisson algebra has a phase space if and only if it is
sub-adjacent to a pre-Poisson algebra. Moreover, we introduce the notion of
Manin triples of pre-Poisson algebras and show that there is a one-to-one
correspondence between Manin triples of pre-Poisson algebras and phase spaces
of Poisson algebras. Then we introduce the notion of pre-Poisson bialgebras,
which is equivalent to Manin triples of pre-Poisson algebras. We study
coboundary pre-Poisson bialgebras, which leads to an analogue of the classical
Yang-Baxter equation. Furthermore, we introduce the notions of quasi-triangular
and factorizable pre-Poisson bialgebras as special cases. A quasi-triangular
pre-Poisson bialgebra gives rise to a relative Rota-Baxter operator of weight
$1$. The double of a pre-Poisson bialgebra enjoys a natural factorizable
pre-Poisson bialgebra structure. Finally, we introduce the notion of quadratic
Rota-Baxter pre-Poisson algebras and show that there is a one-to-one
correspondence between quadratic Rota-Baxter pre-Poisson algebras and
factorizable pre-Poisson bialgebras. Based on this construction, we give a
phase space for a Rota-Baxter symplectic Poisson algebra.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:58:27 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Wang",
"You",
""
],
[
"Sheng",
"Yunhe",
""
]
] |
2504.20937 | Francisco Carter | Francisco Carter, Nancy Hitschfeld, Crist\'obal A. Navarro | M\`imir: A real-time interactive visualization library for CUDA programs | null | null | null | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Real-time visualization of computational simulations running over graphics
processing units (GPU) is a valuable feature in modern science and
technological research, as it allows researchers to visually assess the quality
and correctness of their computational models during the simulation. Due to the
high throughput involved in GPU-based simulations, classical visualization
approaches such as ones based on copying to RAM or storage are not feasible
anymore, as they imply large memory transfers between GPU and CPU at each
moment, reducing both computational performance and interactivity. Implementing
real-time visualizers for GPU simulation codes is a challenging task as it
involves dealing with i) low-level integration of graphics APIs (e.g, OpenGL
and Vulkan) into the general-purpose GPU code, ii) a careful and efficient
handling of memory spaces and iii) finding a balance between rendering and
computing as both need the GPU resources. In this work we present M\`imir, a
CUDA/Vulkan interoperability C++ library that allows users to add real-time
2D/3D visualization to CUDA codes with low programming effort. With M\`imir,
researchers can leverage state-of-the-art CUDA/Vulkan interoperability features
without needing to invest time in learning the complex low-level technical
aspects involved. Internally, M\`imir streamlines the interoperability mapping
between CUDA device memory containing simulation data and Vulkan graphics
resources, so that changes on the data are instantly reflected in the
visualization. This abstraction scheme allows generating visualizations with
minimal alteration over the original source code, needing only to replace the
GPU memory allocation lines of the data to be visualized by the API calls
provided by M\`imir among other optional changes.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:02:57 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Carter",
"Francisco",
""
],
[
"Hitschfeld",
"Nancy",
""
],
[
"Navarro",
"Cristóbal A.",
""
]
] |
2504.20938 | Zhengfu He | Zhengfu He and Junxuan Wang and Rui Lin and Xuyang Ge and Wentao Shu
and Qiong Tang and Junping Zhang and Xipeng Qiu | Towards Understanding the Nature of Attention with Low-Rank Sparse
Decomposition | null | null | null | null | cs.LG cs.CL | http://creativecommons.org/licenses/by/4.0/ | We propose Low-Rank Sparse Attention (Lorsa), a sparse replacement model of
Transformer attention layers to disentangle original Multi Head Self Attention
(MHSA) into individually comprehensible components. Lorsa is designed to
address the challenge of attention superposition to understand
attention-mediated interaction between features in different token positions.
We show that Lorsa heads find cleaner and finer-grained versions of previously
discovered MHSA behaviors like induction heads, successor heads and attention
sink behavior (i.e., heavily attending to the first token). Lorsa and Sparse
Autoencoder (SAE) are both sparse dictionary learning methods applied to
different Transformer components, and lead to consistent findings in many ways.
For instance, we discover a comprehensive family of arithmetic-specific Lorsa
heads, each corresponding to an atomic operation in Llama-3.1-8B. Automated
interpretability analysis indicates that Lorsa achieves parity with SAE in
interpretability while Lorsa exhibits superior circuit discovery properties,
especially for features computed collectively by multiple MHSA heads. We also
conduct extensive experiments on architectural design ablation, Lorsa scaling
law and error analysis.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:03:03 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"He",
"Zhengfu",
""
],
[
"Wang",
"Junxuan",
""
],
[
"Lin",
"Rui",
""
],
[
"Ge",
"Xuyang",
""
],
[
"Shu",
"Wentao",
""
],
[
"Tang",
"Qiong",
""
],
[
"Zhang",
"Junping",
""
],
[
"Qiu",
"Xipeng",
""
]
] |
2504.20939 | Neda Moghim | Nasrin Gholami, Neda Moghim, Behrouz Shahgholi Ghahfarokhi, Pouyan
Salavati, Christo Kurisummoottil Thomas, Sachin Shetty, Tahereh Rahmati | Flexible Semantic-Aware Resource Allocation: Serving More Users Through
Similarity Range Constraints | null | null | null | null | cs.NI eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Semantic communication (SemCom) aims to enhance the resource efficiency of
next-generation networks by transmitting the underlying meaning of messages,
focusing on information relevant to the end user. Existing literature on SemCom
primarily emphasizes learning the encoder and decoder through end-to-end deep
learning frameworks, with the objective of minimizing a task-specific semantic
loss function. Beyond its influence on the physical and application layer
design, semantic variability across users in multi-user systems enables the
design of resource allocation schemes that incorporate user-specific semantic
requirements. To this end, \emph{a semantic-aware resource allocation} scheme
is proposed with the objective of maximizing transmission and semantic
reliability, ultimately increasing the number of users whose semantic
requirements are met. The resulting resource allocation problem is a non-convex
mixed-integer nonlinear program (MINLP), which is known to be NP-hard. To make
the problem tractable, it is decomposed into a set of sub-problems, each of
which is efficiently solved via geometric programming techniques. Finally,
simulations demonstrate that the proposed method improves user satisfaction by
up to $17.1\%$ compared to state of the art methods based on quality of
experience-aware SemCom methods.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:04:48 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Gholami",
"Nasrin",
""
],
[
"Moghim",
"Neda",
""
],
[
"Ghahfarokhi",
"Behrouz Shahgholi",
""
],
[
"Salavati",
"Pouyan",
""
],
[
"Thomas",
"Christo Kurisummoottil",
""
],
[
"Shetty",
"Sachin",
""
],
[
"Rahmati",
"Tahereh",
""
]
] |
2504.20940 | Maximilian Stupp | Maximilian Stupp, P. S. Koutsourelakis | Energy-Based Coarse-Graining in Molecular Dynamics: A Flow-Based
Framework Without Data | null | null | null | null | physics.chem-ph cs.LG physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Coarse-grained (CG) models offer an effective route to reducing the
complexity of molecular simulations, yet conventional approaches depend heavily
on long all-atom molecular dynamics (MD) trajectories to adequately sample
configurational space. This data-driven dependence limits their accuracy and
generalizability, as unvisited configurations remain excluded from the
resulting CG model. We introduce a data-free generative framework for
coarse-graining that directly targets the all-atom Boltzmann distribution. Our
model defines a structured latent space comprising slow collective variables,
which are statistically associated with multimodal marginal densities capturing
metastable states, and fast variables, which represent the remaining degrees of
freedom with simple, unimodal conditional distributions. A potentially
learnable, bijective map from the full latent space to the all-atom
configuration space enables automatic and accurate reconstruction of molecular
structures. The model is trained using an energy-based objective that minimizes
the reverse Kullback-Leibler divergence, relying solely on the interatomic
potential rather than sampled trajectories. A tempering scheme is used to
stabilize training and promote exploration of diverse configurations. Once
trained, the model can generate unbiased, one-shot equilibrium all-atom
samples. We validate the method on two synthetic systems-a double-well
potential and a Gaussian mixture-as well as on the benchmark alanine dipeptide.
The model captures all relevant modes of the Boltzmann distribution, accurately
reconstructs atomic configurations, and learns physically meaningful
coarse-grained representations, all without any simulation data.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:05:27 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Stupp",
"Maximilian",
""
],
[
"Koutsourelakis",
"P. S.",
""
]
] |
2504.20941 | Peilin He | Peilin He, Liou Tang, M. Amin Rahimian, James Joshi | Conformal-DP: Differential Privacy on Riemannian Manifolds via Conformal
Transformation | Submitted and do not distribute | null | null | null | cs.CR math.DG stat.OT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Differential Privacy (DP) has been established as a safeguard for private
data sharing by adding perturbations to information release. Prior research on
DP has extended beyond data in the flat Euclidean space and addressed data on
curved manifolds, e.g., diffusion tensor MRI, social networks, or organ shape
analysis, by adding perturbations along geodesic distances. However, existing
manifold-aware DP methods rely on the assumption that samples are uniformly
distributed across the manifold. In reality, data densities vary, leading to a
biased noise imbalance across manifold regions, weakening the privacy-utility
trade-offs. To address this gap, we propose a novel mechanism: Conformal-DP,
utilizing conformal transformations on the Riemannian manifold to equalize
local sample density and to redefine geodesic distances accordingly while
preserving the intrinsic geometry of the manifold. Our theoretical analysis
yields two main results. First, we prove that the conformal factor computed
from local kernel-density estimates is explicitly data-density-aware; Second,
under the conformal metric, the mechanism satisfies $ \varepsilon
$-differential privacy on any complete Riemannian manifold and admits a
closed-form upper bound on the expected geodesic error that depends only on the
maximal density ratio, not on global curvatureof the manifold. Our experimental
results validate that the mechanism achieves high utility while providing the $
\varepsilon $-DP guarantee for both homogeneous and especially heterogeneous
manifold data.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:05:55 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"He",
"Peilin",
""
],
[
"Tang",
"Liou",
""
],
[
"Rahimian",
"M. Amin",
""
],
[
"Joshi",
"James",
""
]
] |
2504.20942 | Christopher Watson | Christopher Watson, Rajeev Alur, Divya Gopinath, Ravi Mangal, Corina
S. Pasareanu | Scenario-based Compositional Verification of Autonomous Systems with
Neural Perception | null | null | null | null | cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in deep learning have enabled the development of autonomous
systems that use deep neural networks for perception. Formal verification of
these systems is challenging due to the size and complexity of the perception
DNNs as well as hard-to-quantify, changing environment conditions. To address
these challenges, we propose a probabilistic verification framework for
autonomous systems based on the following key concepts: (1) Scenario-based
Modeling: We decompose the task (e.g., car navigation) into a composition of
scenarios, each representing a different environment condition. (2)
Probabilistic Abstractions: For each scenario, we build a compact abstraction
of perception based on the DNN's performance on an offline dataset that
represents the scenario's environment condition. (3) Symbolic Reasoning and
Acceleration: The abstractions enable efficient compositional verification of
the autonomous system via symbolic reasoning and a novel acceleration proof
rule that bounds the error probability of the system under arbitrary variations
of environment conditions. We illustrate our approach on two case studies: an
experimental autonomous system that guides airplanes on taxiways using
high-dimensional perception DNNs and a simulation model of an F1Tenth
autonomous car using LiDAR observations.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:06:22 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Watson",
"Christopher",
""
],
[
"Alur",
"Rajeev",
""
],
[
"Gopinath",
"Divya",
""
],
[
"Mangal",
"Ravi",
""
],
[
"Pasareanu",
"Corina S.",
""
]
] |
2504.20943 | Rudnei Ramos | Rudnei O. Ramos and Gabriel S. Rodrigues | Viability of warm inflation with standard model interactions | 8 pages, 5 figures | null | null | null | hep-ph astro-ph.CO gr-qc hep-th | http://creativecommons.org/licenses/by/4.0/ | The minimal warm inflation scenario proposed in Ref. [1] -- featuring an
axion-like inflaton coupled to Standard Model (SM) gluons via the standard
interaction $\phi G \tilde G$ -- offers a compelling bridge between
inflationary dynamics and SM particle content. While the model retains only the
inflaton as a beyond-SM field, its original analysis relied on some approximate
treatments of warm inflation's (WI) dynamics. Here, we revisit this scenario
using WI2easy, a precision computational tool for WI dynamics [2], to
rigorously evaluate the model's viability and full range of model's parameters
compatible with the observational parameters. Overall, we find that the results
of Ref. [1] hold, but with significant differences in the weak and strong
dissipative regimes of WI.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:10:02 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Ramos",
"Rudnei O.",
""
],
[
"Rodrigues",
"Gabriel S.",
""
]
] |
2504.20944 | Kleanthis Avramidis | Kleanthis Avramidis, Woojae Jeong, Aditya Kommineni, Sudarsana R.
Kadiri, Marcus Ma, Colin McDaniel, Myzelle Hughes, Thomas McGee, Elsi Kaiser,
Dani Byrd, Assal Habibi, B. Rael Cahn, Idan A. Blank, Kristina Lerman,
Takfarinas Medani, Richard M. Leahy, Shrikanth Narayanan | Deep Learning Characterizes Depression and Suicidal Ideation from Eye
Movements | Preprint. 12 pages, 5 figures | null | null | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Identifying physiological and behavioral markers for mental health conditions
is a longstanding challenge in psychiatry. Depression and suicidal ideation, in
particular, lack objective biomarkers, with screening and diagnosis primarily
relying on self-reports and clinical interviews. Here, we investigate eye
tracking as a potential marker modality for screening purposes. Eye movements
are directly modulated by neuronal networks and have been associated with
attentional and mood-related patterns; however, their predictive value for
depression and suicidality remains unclear. We recorded eye-tracking sequences
from 126 young adults as they read and responded to affective sentences, and
subsequently developed a deep learning framework to predict their clinical
status. The proposed model included separate branches for trials of positive
and negative sentiment, and used 2D time-series representations to account for
both intra-trial and inter-trial variations. We were able to identify
depression and suicidal ideation with an area under the receiver operating
curve (AUC) of 0.793 (95% CI: 0.765-0.819) against healthy controls, and
suicidality specifically with 0.826 AUC (95% CI: 0.797-0.852). The model also
exhibited moderate, yet significant, accuracy in differentiating depressed from
suicidal participants, with 0.609 AUC (95% CI 0.571-0.646). Discriminative
patterns emerge more strongly when assessing the data relative to response
generation than relative to the onset time of the final word of the sentences.
The most pronounced effects were observed for negative-sentiment sentences,
that are congruent to depressed and suicidal participants. Our findings
highlight eye tracking as an objective tool for mental health assessment and
underscore the modulatory impact of emotional stimuli on cognitive processes
affecting oculomotor control.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:11:13 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Avramidis",
"Kleanthis",
""
],
[
"Jeong",
"Woojae",
""
],
[
"Kommineni",
"Aditya",
""
],
[
"Kadiri",
"Sudarsana R.",
""
],
[
"Ma",
"Marcus",
""
],
[
"McDaniel",
"Colin",
""
],
[
"Hughes",
"Myzelle",
""
],
[
"McGee",
"Thomas",
""
],
[
"Kaiser",
"Elsi",
""
],
[
"Byrd",
"Dani",
""
],
[
"Habibi",
"Assal",
""
],
[
"Cahn",
"B. Rael",
""
],
[
"Blank",
"Idan A.",
""
],
[
"Lerman",
"Kristina",
""
],
[
"Medani",
"Takfarinas",
""
],
[
"Leahy",
"Richard M.",
""
],
[
"Narayanan",
"Shrikanth",
""
]
] |
2504.20945 | Lucrezia Ravera | J. Fran\c{c}ois, L. Ravera | There is no boundary problem | 8 pages | null | null | null | gr-qc hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In General Relativity and gauge field theory, one often encounters a claim,
which may be called the boundary problem, according to which "boundaries break
diffeomorphism and gauge symmetries". We argue that this statement has the same
conceptual structure as the hole argument, and is thus likewise defused by the
point-coincidence argument: We show that the boundary problem dissolves once it
is understood that a physical region, thus its boundary, is relationally and
invariantly defined. This insight can be technically implemented via the
Dressing Field Method, a systematic tool to exhibit the gauge-invariant content
of general-relativistic gauge field theories, whereby physical
field-theoretical degrees of freedoms co-define each other and define,
coordinatize, the physical spacetime. We illustrate our claim with a simple
application to the case of General Relativity.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:12:19 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"François",
"J.",
""
],
[
"Ravera",
"L.",
""
]
] |
2504.20947 | Shinkyu Park | Norah K. Alghamdi and Shinkyu Park | Opinion-Driven Decision-Making for Multi-Robot Navigation through Narrow
Corridors | null | null | null | null | cs.RO cs.MA | http://creativecommons.org/licenses/by/4.0/ | We propose an opinion-driven navigation framework for multi-robot traversal
through a narrow corridor. Our approach leverages a multi-agent decision-making
model known as the Nonlinear Opinion Dynamics (NOD) to address the narrow
corridor passage problem, formulated as a multi-robot navigation game. By
integrating the NOD model with a multi-robot path planning algorithm, we
demonstrate that the framework effectively reduces the likelihood of deadlocks
during corridor traversal. To ensure scalability with an increasing number of
robots, we introduce a game reduction technique that enables efficient
coordination in larger groups. Extensive simulation studies are conducted to
validate the effectiveness of the proposed approach.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:14:55 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Alghamdi",
"Norah K.",
""
],
[
"Park",
"Shinkyu",
""
]
] |
2504.20949 | Jaehyeok Lee | Jaehyeok Lee | Prekosmic Grothendieck/Galois Categories | This is the author's Ph.D. thesis | null | null | null | math.AG | http://creativecommons.org/licenses/by/4.0/ | We establish a generalized version of the duality between groups and the
categories of their representations on sets. Given an abstract symmetric
monoidal category $K$ called Galois prekosmos, we define pre-Galois objects in
$K$ and study the categories of their representations internal to $K$. The
motivating example of $K$ is the cartesian monoidal category $\textit{Set}$ of
sets, and pre-Galois objects in $\textit{Set}$ are groups. We present an
axiomatic definition of pre-Galois $K$-categories, which is a complete abstract
characterization of the categories of representations of pre-Galois objects in
$K$. The category of covering spaces over a well-connected topological space is
a prototype of a pre-Galois $\textit{Set}$-category. We establish a perfect
correspondence between pre-Galois objects in $K$ and pre-Galois $K$-categories
pointed with pre-fiber functors.
We also establish a generalized version of the duality between flat affine
group schemes and the categories of their linear representations. Given an
abstract symmetric monoidal category $K$ called Grothendieck prekosmos, we
define what are pre-Grothendieck objects in $K$ and study the categories of
their representations internal to $K$. The motivating example of $K$ is the
symmetric monoidal category $\textit{Vec}_k$ of vector spaces over a field $k$,
and pre-Grothendieck objects in $\textit{Vec}_k$ are affine group $k$-schemes.
We present an axiomatic definition of pre-Grothendieck $K$-categories, which is
a complete abstract characterization of the categories of representations of
pre-Grothendieck objects in $K$. The indization of a neutral Tannakian category
over a field $k$ is a prototype of a pre-Grothendieck
$\textit{Vec}_k$-category. We establish a perfect correspondence between
pre-Grothendieck objects in $K$ and pre-Grothendieck $K$-categories pointed
with pre-fiber functors.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:15:40 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Lee",
"Jaehyeok",
""
]
] |
2504.20950 | Yakov Shalunov | Yakov Shalunov | Improved Bounds on the Space Complexity of Circuit Evaluation | 8 pages | null | null | null | cs.CC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Williams (STOC 2025) recently proved that time-$t$ multitape Turing machines
can be simulated using $O(\sqrt{t \log t})$ space using the Cook-Mertz (STOC
2024) tree evaluation procedure. As Williams notes, applying this result to
fast algorithms for the circuit value problem implies an $O(\sqrt{s} \cdot
\mathrm{polylog}\; s)$ algorithm for evaluating size $s$ circuits.
In this work, we provide a direct reduction from circuit value to tree
evaluation without passing through Turing machines, simultaneously improving
the bound to $O(\sqrt{s \log s})$ space and providing a proof with fewer
abstraction layers.
This result can be thought of as a "sibling" result to Williams' for circuit
complexity instead of time; in particular, using the fact that time-$t$ Turing
machines have size $O(t \log t)$ circuits, we can recover a slightly weakened
version of Williams' result, simulating time-$t$ machines in space $O(\sqrt{t}
\log t)$.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:18:52 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Shalunov",
"Yakov",
""
]
] |
2504.20951 | Maryna Vyshnyvetska | Maryna Vyshnyvetska | Information Gravity: A Field-Theoretic Model for Token Selection in
Large Language Models | 12 pages, 1 figure | null | 10.5281/zenodo.15289890 | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We propose a theoretical model called "information gravity" to describe the
text generation process in large language models (LLMs). The model uses
physical apparatus from field theory and spacetime geometry to formalize the
interaction between user queries and the probability distribution of generated
tokens. A query is viewed as an object with "information mass" that curves the
semantic space of the model, creating gravitational potential wells that
"attract" tokens during generation. This model offers a mechanism to explain
several observed phenomena in LLM behavior, including hallucinations (emerging
from low-density semantic voids), sensitivity to query formulation (due to
semantic field curvature changes), and the influence of sampling temperature on
output diversity.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:21:20 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Vyshnyvetska",
"Maryna",
""
]
] |
2504.20952 | Qinghao Shen | Q. Shen, A. Pikalev, F.J.J. Peeters, V. Guerra, M.C.M.van de Sanden | Pinpointing energy transfer mechanisms in the quenching process of
microwave air plasma | 38 pages, 20 Figures | null | null | null | physics.plasm-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A time-dependent multi-temperature quenching model at atmospheric pressure,
incorporating chemical and vibrational kinetics, is introduced. The model
provides insights into the pathways of NOx formation and destruction in the
downstream region of a microwave air plasma. The relaxation of the temperatures
during the forced cooling trajectory by the wall is modelled. A Continuous
Stirred Tank Reactor model and a Plug Flow Reactor model represent the plasma
and quenching regions, respectively. For the non-thermal conditions, where gas
and vibrational temperatures differ, most reaction rate coefficients, except
those obtained from molecular dynamics methods, are determined based on a
generalized Fridman-Macheret scheme. The energy transfer channels involved in
the quenching process are tracked across different time scales. By varying the
gas temperature in the plasma region and the cooling rate, the reaction
pathways for the NOx synthesis mechanism are analysed. This research provides a
first step for the further advancement and optimisation of plasma reactors for
efficient NOx production.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:22:19 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Shen",
"Q.",
""
],
[
"Pikalev",
"A.",
""
],
[
"Peeters",
"F. J. J.",
""
],
[
"Guerra",
"V.",
""
],
[
"van de Sanden",
"M. C. M.",
""
]
] |
2504.20953 | David Llamas | David Llamas, Dmitry Chistikov, Adrian Kent, Mike Paterson, Olga
Goulko | Optimal Local Simulations of a Quantum Singlet | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Bell's seminal work showed that no local hidden variable (LHV) model can
fully reproduce the quantum correlations of a two-qubit singlet state. His
argument and later developments by Clauser et al. effectively rely on gaps
between the anticorrelations achievable by classical models and quantum theory
for projective measurements along randomly chosen axes separated by a fixed
angle. However, the size of these gaps has to date remained unknown. Here we
numerically determine the LHV models maximizing anticorrelations for random
axes separated by any fixed angle, by mapping the problem onto ground state
configurations of fixed-range spin models. We identify angles where this gap is
largest and thus best suited for Bell tests. These findings enrich the
understanding of Bell non-locality as a physical resource in quantum
information theory and quantum cryptography.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:24:32 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Llamas",
"David",
""
],
[
"Chistikov",
"Dmitry",
""
],
[
"Kent",
"Adrian",
""
],
[
"Paterson",
"Mike",
""
],
[
"Goulko",
"Olga",
""
]
] |
2504.20954 | Ingrid Pelisoli | Ingrid Pelisoli, Boris G\"ansicke, Keith Horne, Danny Steeghs, Gijs
Nelemans, Kevin Burdge, David Buckley, Vik Dhillon, Stuart Littlefair, Daniel
Bayliss | 21 years of Astronomy at Warwick: celebrating the legacy of Prof. Tom
Marsh | This is the authors' version of an article featured in Astronomy &
Geophysics, Published by Oxford University Press | null | 10.1093/astrogeo/ataf009 | null | astro-ph.SR astro-ph.EP astro-ph.IM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Between the 4th and 6th of September 2024, the Astronomy & Astrophysics group
at the University of Warwick held a meeting to celebrate 21 years of astronomy
at Warwick and the scientific legacy of the late Prof. Tom Marsh, the group
founder. More than a hundred people attended the meeting, with about half of
the attendees being external delegates and coming from as far afield as the USA
and South Africa. Tom Marsh moved to the University of Warwick from Southampton
in 2003, after the Department of Physics decided to expand the scope of its
research. From its humble beginnings with only two staff members, Tom himself
and Boris G\"ansicke, one postdoc and a couple of PhD students, the group has
now grown to more than 95 members, including 25 staff. Tom pioneered the
development of Doppler tomography, led key discoveries in the field of
double-degenerate binary systems and made extensive contributions to
instrumentation, primarily to developing the high-speed imaging photometers
ULTRACAM, ULTRASPEC and HiPERCAM. This article provides a summary of Tom's
legacy and Warwick's history as presented in the 21 years of Astronomy at
Warwick meeting.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:25:07 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Pelisoli",
"Ingrid",
""
],
[
"Gänsicke",
"Boris",
""
],
[
"Horne",
"Keith",
""
],
[
"Steeghs",
"Danny",
""
],
[
"Nelemans",
"Gijs",
""
],
[
"Burdge",
"Kevin",
""
],
[
"Buckley",
"David",
""
],
[
"Dhillon",
"Vik",
""
],
[
"Littlefair",
"Stuart",
""
],
[
"Bayliss",
"Daniel",
""
]
] |
2504.20957 | Alexander Balandin | Subhajit Ghosh, Nicholas Sesing, Tina Salguero, Sergey Rumyantsev,
Roger K. Lake, and Alexander A. Balandin | A Quieter State of Charge -- Ultra-Low-Noise Collective Current in
Charge-Density-Wave Nanowires | 29 pages; 12 pages | null | null | null | cond-mat.str-el cond-mat.mes-hall cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In quasi-one-dimensional (quasi-1D) charge-density-wave (CDW) systems,
electric current comprises normal electrons and a collective, electron-lattice
condensate current associated with CDW sliding. While achieving the
dissipation-less Frohlich current of the sliding condensate is impossible in
real materials, one can imagine an important related target, namely reaching
the electron transport regime where electronic noise is inhibited due to the
collective, strongly-correlated nature of the electron-lattice condensate
current. Here we report that in nanowires of the fully-gapped CDW material
(TaSe4)2I, low-frequency electronic noise is suppressed below the limit of
thermalized charge carriers in passive resistors. When the current is dominated
by the sliding Frohlich condensate, the normalized noise spectral density
decreases linearly with current -- a striking departure from the constant value
observed in conventional conductors. This discovery signals intrinsically lower
current fluctuations within a correlated transport regime. The dominant noise
source due to fluctuations in the CDW depinning threshold is extrinsic and
caused by lattice imperfections that locally pin the condensate. Once the bias
voltage is well past threshold and the sliding mode is established, the
normalized noise drops below the noise of normal electrons. No residual minimum
noise level is observed for the current of the condensate. Since flicker noise
limits phase stability in communication systems, reduces the sensitivity and
selectivity of sensors, and degrades coherence in quantum devices, our
discovery introduces a fundamentally new strategy for achieving ultra-low-noise
performance in nanoscale and quantum electronics using strongly correlated
materials.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:28:20 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Ghosh",
"Subhajit",
""
],
[
"Sesing",
"Nicholas",
""
],
[
"Salguero",
"Tina",
""
],
[
"Rumyantsev",
"Sergey",
""
],
[
"Lake",
"Roger K.",
""
],
[
"Balandin",
"Alexander A.",
""
]
] |
2504.20958 | Steffen Wittrock | Steffen Wittrock, Christopher Klose, Salvatore Perna, Korbinian
Baumgaertl, Andrea Mucchietto, Michael Schneider, Josefin Fuchs, Victor
Deinhart, Tamer Karaman, Dirk Grundler, Stefan Eisebitt, Bastian Pfau, Daniel
Schick | Soft-X-ray momentum microscopy of nonlinear magnon interactions below
100-nm wavelength | Ancillary files include videos presenting MMM images during (a) a
frequency sweep and (b) a power sweep | null | null | null | cond-mat.mes-hall cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | Magnons represent quantised collective motions of long-range ordered spins.
For wavelength below 100 nm, exchange interactions dominate their physics,
which gives rise to a so far unexplored regime of nonlinearities and couplings
between magnons and other quasiparticles. Besides their selective excitation,
also the detection of such short-wavelength spin waves remains a challenge of
current research and technology. Here, we probe the amplitude and wave vector
of magnons by means of quasi-elastic resonant soft-X-ray scattering. This
Magnon Momentum Microscopy (MMM) can access magnons directly in momentum space
with remarkable sensitivity and high photon efficiency up to THz frequencies
and down to few-nanometre wavelengths. The two-dimensional information obtained
by this light-scattering-based technique is especially valuable for studying
the nonlinear interactions of exchange-dominated magnons within technologically
relevant thin-film samples. In doing so, we uncover a rich variety of deeply
nonlinear magnon interactions, highlighting their potential for applications in
novel computing schemes. With its intrinsic element-selectivity and ability to
probe also buried layers, soft-X-ray MMM has the potential to establish itself
as an advanced tool for ultrabroadband studies of short-wavelength magnonics.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:31:15 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Wittrock",
"Steffen",
""
],
[
"Klose",
"Christopher",
""
],
[
"Perna",
"Salvatore",
""
],
[
"Baumgaertl",
"Korbinian",
""
],
[
"Mucchietto",
"Andrea",
""
],
[
"Schneider",
"Michael",
""
],
[
"Fuchs",
"Josefin",
""
],
[
"Deinhart",
"Victor",
""
],
[
"Karaman",
"Tamer",
""
],
[
"Grundler",
"Dirk",
""
],
[
"Eisebitt",
"Stefan",
""
],
[
"Pfau",
"Bastian",
""
],
[
"Schick",
"Daniel",
""
]
] |
2504.20959 | Sophie Huczynska | Sophie Huczynska, Christopher Jefferson and Struan McCartney | Digraph-defined external difference families and new circular external
difference families | null | null | null | null | math.CO | http://creativecommons.org/licenses/by/4.0/ | External difference families (EDFs) are combinatorial objects which were
introduced in the early 2000s, motivated by information security applications;
various generalizations have since been defined and investigated, in particular
strong EDFs (SEDFs) and circular EDFs (CEDFs). We present a framework based on
graphs and digraphs which offers a new unified way to view these structures,
and leads to natural new research questions. We present constructions and
structural results about these digraph-defined EDFs, and we obtain new explicit
constructions for infinite families of CEDFs, in particular
$(ml^2,m,l,1)$-CEDFs. We construct the first infinite family of such CEDFs in
non-cyclic abelian groups; these have odd values of $m$ and $l$. We also
present the first CEDF in a non-abelian group.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:32:07 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Huczynska",
"Sophie",
""
],
[
"Jefferson",
"Christopher",
""
],
[
"McCartney",
"Struan",
""
]
] |
2504.20960 | Mahdi Mirzakhani | M. Mirzakhani, W. Baker, M. Chaudhuri, J. B. Dent, R. Dey, B. Dutta,
V. Iyer, A. Jastram, V. K. S. Kashyap, A. Kubik, K. Lang, R. Mahapatra, S.
Maludze, N. Mirabolfathi, B. Mohanty, D. Mondal, H. Neog, J. L. Newstead, M.
Platt, S. Sahoo, J. Sander, L. E. Strigari, J. Walker | MINER Reactor Based Search for Axion-Like Particles Using Sapphire
(Al2O3) Detectors | 8 pages, 12 figures | null | null | null | hep-ex | http://creativecommons.org/licenses/by/4.0/ | The absence of definitive results for WIMP dark matter has sparked growing
interest in alternative dark matter candidates, such as axions and Axion-Like
Particles (ALPs), which also provide insight into the strong CP problem. The
Mitchell Institute Neutrino Experiment at Reactor (MINER), conducted at the
Nuclear Science Center of Texas A&M University, investigated ALPs near a 1 MW
TRIGA nuclear reactor core, positioned approximately 4 meters away. This
experiment employed cryogenic sapphire detectors with a low detection threshold
(approximately 100 eV), equipped with a Transition Edge Sensor capable of
detecting athermal phonons. Due to the low-background environment, we were able
to exclude ALPs with axion-photon coupling and axion-electron coupling as small
as \(g_{a\gamma\gamma} = 10^{-5}\) and \(g_{aee} = 10^{-7}\), respectively.
Energy depositions below 3 keV were not considered and remain blinded for our
Coherent Elastic Neutrino Nucleus Scattering (CEvNS) analysis. This is the
first result demonstrating the MINER experiment's potential to probe low-mass
ALPs, enabled by its low-threshold detector and proximity to a reactor.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:32:20 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Mirzakhani",
"M.",
""
],
[
"Baker",
"W.",
""
],
[
"Chaudhuri",
"M.",
""
],
[
"Dent",
"J. B.",
""
],
[
"Dey",
"R.",
""
],
[
"Dutta",
"B.",
""
],
[
"Iyer",
"V.",
""
],
[
"Jastram",
"A.",
""
],
[
"Kashyap",
"V. K. S.",
""
],
[
"Kubik",
"A.",
""
],
[
"Lang",
"K.",
""
],
[
"Mahapatra",
"R.",
""
],
[
"Maludze",
"S.",
""
],
[
"Mirabolfathi",
"N.",
""
],
[
"Mohanty",
"B.",
""
],
[
"Mondal",
"D.",
""
],
[
"Neog",
"H.",
""
],
[
"Newstead",
"J. L.",
""
],
[
"Platt",
"M.",
""
],
[
"Sahoo",
"S.",
""
],
[
"Sander",
"J.",
""
],
[
"Strigari",
"L. E.",
""
],
[
"Walker",
"J.",
""
]
] |
2504.20962 | Andrzej A. Zdziarski | Andrzej A. Zdziarski, Callan M. Wood, Francesco Carotenuto | A Novel Method of Modeling Extended Emission of Compact Jets:
Application to Swift J1727.8-1613 | Submitted to ApJ, comments are welcome | null | null | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | Flat radio spectra of compact jets launched by both supermassive and
stellar-mass black holes are explained by an interplay of self-absorbed
synchrotron emission up to some distance along the jet and optically thin
synchrotron at larger distances (Blandford & Konigl 1979). Their spatial
structure is usually studied using core shifts, in which the position of the
peak (core) of the emission depends on the frequency. Here, we propose a novel
method to fit the spatial dependence of the flux density at a given frequency
of the jet and counterjet (when observed) using the theoretical spatial
dependencies, which we provide as simple analytical formulae. We apply our
method to the spatial structure of the jets in the luminous hard spectral state
of the black-hole X-ray binary Swift J1727.8--1613. It was the most resolved
continuous jet from an X-ray binary ever observed. We find that the observed
approaching jet is significantly intrinsically stronger than the receding one,
which we attribute to an increase in the emission of both jets with time
(observationally confirmed), together with the light travel effect, causing the
receding jet to be observed at an earlier epoch than the approaching one. The
jets are relatively slow, with the velocity $\sim(0.3$--$0.4)c$. Our findings
imply that the magnetic field strength increased with time. Also, the magnetic
flux is much lower than in jets launched by `Magnetically Arrested Disks'. Our
method is general, and we propose that it be applied to jets launched by
stellar-mass and supermassive black holes.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:32:38 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Zdziarski",
"Andrzej A.",
""
],
[
"Wood",
"Callan M.",
""
],
[
"Carotenuto",
"Francesco",
""
]
] |
2504.20963 | Pascal Maillard | Heng Ma and Pascal Maillard | Exponential moments of truncated branching random walk martingales | 14 pages | null | null | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For a branching random walk that drifts to infinity, consider its Malthusian
martingale, i.e.~the additive martingale with parameter $\theta$ being the
smallest root of the characteristic equation. When particles are killed below
the origin, we show that the limit of this martingale admits an exponential
tail, contrary to the case without killing, where the tail is polynomial. In
the critical case, where the characteristic equation has a single root, the
same holds for the (truncated) derivative martingale, as we show. This study is
motivated by recent work on first passage percolation on Erd\H{o}s-R\'enyi
graphs.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:32:43 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Ma",
"Heng",
""
],
[
"Maillard",
"Pascal",
""
]
] |
2504.20964 | Shangyu Li | Shangyu Li, Juyong Jiang, Tiancheng Zhao, Jiasi Shen | OSVBench: Benchmarking LLMs on Specification Generation Tasks for
Operating System Verification | null | null | null | null | cs.CL cs.AI cs.OS cs.PL cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce OSVBench, a new benchmark for evaluating Large Language Models
(LLMs) in generating complete specification code pertaining to operating system
kernel verification tasks. The benchmark first defines the specification
generation problem into a program synthesis problem within a confined scope of
syntax and semantics by providing LLMs with the programming model. The LLMs are
required to understand the provided verification assumption and the potential
syntax and semantics space to search for, then generate the complete
specification for the potentially buggy operating system code implementation
under the guidance of the high-level functional description of the operating
system. This benchmark is built upon a real-world operating system kernel,
Hyperkernel, and consists of 245 complex specification generation tasks in
total, each is a long context task of about 20k-30k tokens. Our comprehensive
evaluation of 12 LLMs exhibits the limited performance of the current LLMs on
the specification generation tasks for operating system verification.
Significant disparities in their performance on the benchmark highlight
differences in their ability to handle long-context code generation tasks. The
evaluation toolkit and benchmark are available at
https://github.com/lishangyu-hkust/OSVBench.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:34:49 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Li",
"Shangyu",
""
],
[
"Jiang",
"Juyong",
""
],
[
"Zhao",
"Tiancheng",
""
],
[
"Shen",
"Jiasi",
""
]
] |
2504.20965 | Zikui Cai | Zikui Cai, Shayan Shabihi, Bang An, Zora Che, Brian R. Bartoldson,
Bhavya Kailkhura, Tom Goldstein, Furong Huang | AegisLLM: Scaling Agentic Systems for Self-Reflective Defense in LLM
Security | ICLR 2025 Workshop BuildingTrust | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | We introduce AegisLLM, a cooperative multi-agent defense against adversarial
attacks and information leakage. In AegisLLM, a structured workflow of
autonomous agents - orchestrator, deflector, responder, and evaluator -
collaborate to ensure safe and compliant LLM outputs, while self-improving over
time through prompt optimization. We show that scaling agentic reasoning system
at test-time - both by incorporating additional agent roles and by leveraging
automated prompt optimization (such as DSPy)- substantially enhances robustness
without compromising model utility. This test-time defense enables real-time
adaptability to evolving attacks, without requiring model retraining.
Comprehensive evaluations across key threat scenarios, including unlearning and
jailbreaking, demonstrate the effectiveness of AegisLLM. On the WMDP unlearning
benchmark, AegisLLM achieves near-perfect unlearning with only 20 training
examples and fewer than 300 LM calls. For jailbreaking benchmarks, we achieve
51% improvement compared to the base model on StrongReject, with false refusal
rates of only 7.9% on PHTest compared to 18-55% for comparable methods. Our
results highlight the advantages of adaptive, agentic reasoning over static
defenses, establishing AegisLLM as a strong runtime alternative to traditional
approaches based on model modifications. Code is available at
https://github.com/zikuicai/aegisllm
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:36:05 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Cai",
"Zikui",
""
],
[
"Shabihi",
"Shayan",
""
],
[
"An",
"Bang",
""
],
[
"Che",
"Zora",
""
],
[
"Bartoldson",
"Brian R.",
""
],
[
"Kailkhura",
"Bhavya",
""
],
[
"Goldstein",
"Tom",
""
],
[
"Huang",
"Furong",
""
]
] |
2504.20966 | Zayd Muhammad Kawakibi Zuhri | Zayd M. K. Zuhri, Erland Hilman Fuadi, Alham Fikri Aji | Softpick: No Attention Sink, No Massive Activations with Rectified
Softmax | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | We introduce softpick, a rectified, not sum-to-one, drop-in replacement for
softmax in transformer attention mechanisms that eliminates attention sink and
massive activations. Our experiments with 340M parameter models demonstrate
that softpick maintains performance parity with softmax on standard benchmarks
while achieving 0% sink rate. The softpick transformer produces hidden states
with significantly lower kurtosis (340 vs 33,510) and creates sparse attention
maps (46.97% sparsity). Models using softpick consistently outperform softmax
when quantized, with particularly pronounced advantages at lower bit
precisions. Our analysis and discussion shows how softpick has the potential to
open new possibilities for quantization, low-precision training, sparsity
optimization, pruning, and interpretability. Our code is available at
https://github.com/zaydzuhri/softpick-attention.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:36:18 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Zuhri",
"Zayd M. K.",
""
],
[
"Fuadi",
"Erland Hilman",
""
],
[
"Aji",
"Alham Fikri",
""
]
] |
2504.20967 | Karola Meszaros | Tam\'as K\'alm\'an, Karola M\'esz\'aros, Alexander Postnikov | Trapezodial property of the generalized Alexander polynomial | null | null | null | null | math.CO math.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fox's conjecture from 1962, that the absolute values of the coefficients of
the Alexander polynomial of an alternating link are trapezoidal, has remained
stubbornly open to this date. Recently Fox's conjecture was settled for all
special alternating links. In this paper we take a broad view of the Alexander
polynomials of special alternating links, showing that they are a generating
function for a statistic on certain vector configurations. We study three types
of vector configurations: (1) vectors arising from cographic matroids, (2)
vectors arising from graphic matroids, (3) vectors arising from totally
positive matrices. We prove that Alexander polynomials of special alternating
links belong to both classes (1) and (2), and prove log-concavity, respectively
trapezoidal, properties for classes (2) and (3). As a special case of our
results, we obtain a new proof of Fox's conjecture for special alternating
links.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:37:31 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Kálmán",
"Tamás",
""
],
[
"Mészáros",
"Karola",
""
],
[
"Postnikov",
"Alexander",
""
]
] |
2504.20968 | Stefan Mitrovic | Stefan Mitrovic | The Redei-Berge function in noncommuting variables | null | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, Stanley and Grinberg introduced a symmetric function associated to
digraphs, called the Redei-Berge symmetric function. This function, however,
does not satisfy the deletion-contraction property, which is a very powerful
tool for proving various identities using induction. In this paper, we
introduce an analogue of this function in noncommuting variables which does
have such property. Furthermore, it specializes to the ordinary Redei-Berge
function when the variables are allowed to commute. This modification allows us
to further generalize properties that are already proved for the original
function and to deduce many new ones.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:37:44 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Mitrovic",
"Stefan",
""
]
] |
2504.20969 | Yiting Zhang | Yiting Zhang, Shichen Li, Elena Shrestha | XPG-RL: Reinforcement Learning with Explainable Priority Guidance for
Efficiency-Boosted Mechanical Search | 13 pages, 5 figures | null | null | null | cs.RO cs.LG | http://creativecommons.org/licenses/by/4.0/ | Mechanical search (MS) in cluttered environments remains a significant
challenge for autonomous manipulators, requiring long-horizon planning and
robust state estimation under occlusions and partial observability. In this
work, we introduce XPG-RL, a reinforcement learning framework that enables
agents to efficiently perform MS tasks through explainable, priority-guided
decision-making based on raw sensory inputs. XPG-RL integrates a task-driven
action prioritization mechanism with a learned context-aware switching strategy
that dynamically selects from a discrete set of action primitives such as
target grasping, occlusion removal, and viewpoint adjustment. Within this
strategy, a policy is optimized to output adaptive threshold values that govern
the discrete selection among action primitives. The perception module fuses
RGB-D inputs with semantic and geometric features to produce a structured scene
representation for downstream decision-making. Extensive experiments in both
simulation and real-world settings demonstrate that XPG-RL consistently
outperforms baseline methods in task success rates and motion efficiency,
achieving up to 4.5$\times$ higher efficiency in long-horizon tasks. These
results underscore the benefits of integrating domain knowledge with learnable
decision-making policies for robust and efficient robotic manipulation.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:37:45 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Zhang",
"Yiting",
""
],
[
"Li",
"Shichen",
""
],
[
"Shrestha",
"Elena",
""
]
] |
2504.20970 | Mete Erdogan | Mete Erdogan, Sebnem Demirtas | SVD Based Least Squares for X-Ray Pneumonia Classification Using Deep
Features | Preprint submitted to IEEE International Workshop on Machine Learning
for Signal Processing (MLSP), 2025 | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate and early diagnosis of pneumonia through X-ray imaging is essential
for effective treatment and improved patient outcomes. Recent advancements in
machine learning have enabled automated diagnostic tools that assist
radiologists in making more reliable and efficient decisions. In this work, we
propose a Singular Value Decomposition-based Least Squares (SVD-LS) framework
for multi-class pneumonia classification, leveraging powerful feature
representations from state-of-the-art self-supervised and transfer learning
models. Rather than relying on computationally expensive gradient based
fine-tuning, we employ a closed-form, non-iterative classification approach
that ensures efficiency without compromising accuracy. Experimental results
demonstrate that SVD-LS achieves competitive performance while offering
significantly reduced computational costs, making it a viable alternative for
real-time medical imaging applications.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:39:16 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Erdogan",
"Mete",
""
],
[
"Demirtas",
"Sebnem",
""
]
] |
2504.20971 | Olaf Post | Olaf Post and Jan Simmer | A distance between operators acting in different Hilbert spaces and
operator convergence | 15 pages, 1 figure; the reference list is as in the original paper
and reflects the state of the art at time of publishing (2019/2020) | Proceedings in "Control theory of infinite-dimensional systems",
pages 53-72, Oper. Theory Adv. Appl., 277, Birkh\"auser/Springer, 2020 | null | null | math.FA math-ph math.MP math.SP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The aim of the present article is to give an introduction to the concept of
quasi-unitary equivalence and to define several (pseudo-)metrics on the space
of self-adjoint operators acting possibly in different Hilbert spaces. As some
of the ``metrics'' do not fulfill all properties of a metric (e.g. some lack
the triangle inequality or the definiteness), we call them ``distances'' here.
To the best of our knowledge, such distances are treated for the first time
here. The present article shall serve as a starting point of further research
(see e.g. arXiv:2412.13165).
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:39:21 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Post",
"Olaf",
""
],
[
"Simmer",
"Jan",
""
]
] |
2504.20972 | Yifan Wei | Yifan Wei and Xiaoyan Yu and Ran Song and Hao Peng and Angsheng Li | SetKE: Knowledge Editing for Knowledge Elements Overlap | The CR version will be updated subsequently | IJCAI 2025 | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) excel in tasks such as retrieval and question
answering but require updates to incorporate new knowledge and reduce
inaccuracies and hallucinations. Traditional updating methods, like fine-tuning
and incremental learning, face challenges such as overfitting and high
computational costs. Knowledge Editing (KE) provides a promising alternative
but often overlooks the Knowledge Element Overlap (KEO) phenomenon, where
multiple triplets share common elements, leading to editing conflicts. We
identify the prevalence of KEO in existing KE datasets and show its significant
impact on current KE methods, causing performance degradation in handling such
triplets. To address this, we propose a new formulation, Knowledge Set Editing
(KSE), and introduce SetKE, a method that edits sets of triplets
simultaneously. Experimental results demonstrate that SetKE outperforms
existing methods in KEO scenarios on mainstream LLMs. Additionally, we
introduce EditSet, a dataset containing KEO triplets, providing a comprehensive
benchmark.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:40:29 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Wei",
"Yifan",
""
],
[
"Yu",
"Xiaoyan",
""
],
[
"Song",
"Ran",
""
],
[
"Peng",
"Hao",
""
],
[
"Li",
"Angsheng",
""
]
] |
2504.20973 | Sara Barja-Martinez | S. Barja-Martinez, M. Aragues-Penalba, A. Sumper, R.
Villafafila-Robles | Local energy communities optimization considering cost and greenhouse
gases minimization | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Local Energy Communities (LECs) facilitate consumer involvement in local
electricity generation and distribution, offering a significant opportunity for
society to participate in the energy transition. This paper presents the
optimization of a renewable energy community in Spain, consisting of four
office buildings, a collectively owned centralized photovoltaic system, and a
Li-ion battery storage system. The case study assesses the performance and
feasibility of the proposed solutions. The results indicate a 6% reduction in
emissions and a 20% reduction in electricity costs, demonstrating the potential
of LECs to enhance energy security by saving costs and emission while
mitigating the vulnerability of local areas to power outages and disruptions
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:42:43 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Barja-Martinez",
"S.",
""
],
[
"Aragues-Penalba",
"M.",
""
],
[
"Sumper",
"A.",
""
],
[
"Villafafila-Robles",
"R.",
""
]
] |
2504.20974 | Elias Nyholm | Elias Nyholm, Oscar Carlsson, Maurice Weiler, Daniel Persson | Equivariant non-linear maps for neural networks on homogeneous spaces | 45 pages,10 figures | null | null | null | cs.LG math.RT stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a novel framework for non-linear equivariant neural
network layers on homogeneous spaces. The seminal work of Cohen et al. on
equivariant $G$-CNNs on homogeneous spaces characterized the representation
theory of such layers in the linear setting, finding that they are given by
convolutions with kernels satisfying so-called steerability constraints.
Motivated by the empirical success of non-linear layers, such as self-attention
or input dependent kernels, we set out to generalize these insights to the
non-linear setting. We derive generalized steerability constraints that any
such layer needs to satisfy and prove the universality of our construction. The
insights gained into the symmetry-constrained functional dependence of
equivariant operators on feature maps and group elements informs the design of
future equivariant neural network layers. We demonstrate how several common
equivariant network architectures - $G$-CNNs, implicit steerable kernel
networks, conventional and relative position embedded attention based
transformers, and LieTransformers - may be derived from our framework.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:42:56 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Nyholm",
"Elias",
""
],
[
"Carlsson",
"Oscar",
""
],
[
"Weiler",
"Maurice",
""
],
[
"Persson",
"Daniel",
""
]
] |
2504.20975 | Stefan Mitrovic | Stefan Mitrovic | Linear function of a poset | null | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Stanley and Grinberg introduced a symmetric function associated with digraphs
and named it the Redei-Berge symmetric function. This function arises from a
suitable combinatorial Hopf algebra on digraphs, which made it possible to
assign the Redei-Berge function to posets. In this paper, we define a new
combinatorial Hopf algebra of posets whose character is a close cousin of the
Redei-Berge character for posets. Further, we investigate the properties of the
symmetric function that arises from this algebra and explore its expansions in
various natural bases of $QSym$ and $Sym$. Finally, we obtain an interesting
method for decomposing a poset.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:43:50 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Mitrovic",
"Stefan",
""
]
] |
2504.20976 | Farhan Sadaf | Dabbrata Das, Argho Deb Das and Farhan Sadaf | Real-Time Wayfinding Assistant for Blind and Low-Vision Users | null | null | null | null | cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Navigating unfamiliar places continues to be one of the most persistent and
essential everyday obstacles for those who are blind or have limited vision
(BLV). Existing assistive technologies, such as GPS-based navigation systems,
AI-powered smart glasses, and sonar-equipped canes, often face limitations in
real-time obstacle avoidance, precise localization, and adaptability to dynamic
surroundings. To investigate potential solutions, we introduced PathFinder, a
novel map-less navigation system that explores different models for
understanding 2D images, including Vision Language Models (VLMs), Large
Language Models (LLMs), and employs monocular depth estimation for free-path
detection. Our approach integrates a Depth-First Search (DFS) algorithm on
depth images to determine the longest obstacle-free path, ensuring optimal
route selection while maintaining computational efficiency. We conducted
comparative evaluations against existing AI-powered navigation methods and
performed a usability study with BLV participants. The results demonstrate that
PathFinder achieves a favorable balance between accuracy, computational
efficiency, and real-time responsiveness. Notably, it reduces mean absolute
error (MAE) and improves decision-making speed in outdoor navigation compared
to AI-based alternatives. Participant feedback emphasizes the system's
usability and effectiveness in outside situations, but also identifies issues
in complicated indoor locations and low-light conditions. Usability testing
revealed that 73% of participants understood how to use the app in about a
minute, and 80% praised its balance of accuracy, quick response, and overall
convenience.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:45:04 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Das",
"Dabbrata",
""
],
[
"Das",
"Argho Deb",
""
],
[
"Sadaf",
"Farhan",
""
]
] |
2504.20977 | Andrey Pikelner | Dennis Horstmann, Kirill Melnikov, Ming-Ming Long, Andrey Pikelner | Integral of the double-emission eikonal function for a massive and a
massless emitter at an arbitrary angle | 44 pages; numerical code available from https://github.com/apik/SSm0 | null | null | SI-HEP-2025-08, TTP25-015, P3H-25-030 | hep-ph | http://creativecommons.org/licenses/by/4.0/ | We present an analytic calculation of the integrated double-emission eikonal
function of a massive and a massless emitter whose momenta are at an arbitrary
angle to each other. This quantity provides one of the required ingredients for
extending the nested soft-collinear subtraction scheme to processes with
massive final-state particles. To calculate it, we use the standard methodology
involving reverse unitarity and its extension to cases with Heaviside
functions, integration-by-parts technology and reduction to master integrals,
and differential equations. In addition, we also describe a semi-numerical
method based on the subtraction of infra-red and collinear singularities from
the eikonal function, allowing us to extract divergences of the integrated
eikonal function analytically, and to derive a simple integral representation
for the finite remainder.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:48:17 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Horstmann",
"Dennis",
""
],
[
"Melnikov",
"Kirill",
""
],
[
"Long",
"Ming-Ming",
""
],
[
"Pikelner",
"Andrey",
""
]
] |
2504.20978 | Shamil Asgarli | Shamil Asgarli, Sara Krehbiel, Howard W. Levinson | Coloring graphs as complete graph invariants | 10 pages; some of the results in this paper were independently
arrived by other authors in arXiv: 2504.19783. We acknowledge that their work
was posted a day earlier than our work. See Remark 1.4 | null | null | null | math.CO | http://creativecommons.org/licenses/by/4.0/ | For a graph $G$, the $k$-coloring graph $\mathcal{C}_{k}(G)$ is obtained by
taking the vertices to be the proper $k$-colorings of $G$ and two $k$-colorings
are incident if they differ exactly on one vertex of $G$. Hogan, Scott,
Tamitegama and Tan recently proved that $\mathcal{C}_{k_1}(G_1)\cong
\mathcal{C}_{k_2}(G_2)$ with $k_1>5n_1^2$ and $k_2>5n_2^2$ implies $G_1\cong
G_2$, where $n_i$ is the number of vertices of $G_i$ for $i=1,2$. We sharpen
this result by showing that $\mathcal{C}_{k_1}(G_1)\cong
\mathcal{C}_{k_2}(G_2)$ with $k_1>\chi(G_1)$ and $k_2>\chi(G_2)$ implies
$G_1\cong G_2$, where $\chi(G)$ is the chromatic number of $G$. In other words,
given any graph $\mathcal{C}$, there is at most one pair of $G$ and $k>\chi(G)$
such that $\mathcal C=\mathcal C_k(G)$, and we additionally provide an
algorithm to reconstruct $(G,k)$ from $\mathcal C$ or determine that no such
pair exists. Moreover, our requirement that $k>\chi$ is tight because coloring
graphs are empty for $k<\chi$ and there exist examples of non-isomorphic
$G_1,G_2$ with $k=\chi(G_1)=\chi(G_2)$ such that $\mathcal C_k(G_1)\cong
\mathcal C_k(G_2)$.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:49:06 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Asgarli",
"Shamil",
""
],
[
"Krehbiel",
"Sara",
""
],
[
"Levinson",
"Howard W.",
""
]
] |
2504.20979 | Tiju Cherian John | Tiju Cherian John, Hemant K. Mishra, Saikat Guha | A Complete Characterization of Passive Unitary Normalizable (PUN)
Gaussian States | A shorter version of this article is accepted to IEEE International
Symposium on Information Theory. This arXiv version contains detailed proofs,
more references, a figure and two tables | null | null | null | quant-ph math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We provide a complete characterization of the class of multimode quantum
Gaussian states that can be reduced to a tensor product of thermal states using
only a passive unitary operator. We call these states \textit{passive unitary
normalizable} (PUN) Gaussian states. The characterization of PUN Gaussian
states is given in three different ways: $(i)$ in terms of their covariance
matrices, $(ii)$ using gauge-invariance (a special class of Glauber--Sudarshan
$p$-functions), and $(iii)$ with respect to the recently obtained $(A,\Lambda)$
parametrization of Gaussian states in [J. Math. Phys. 62, 022102 (2021)]. In
terms of the covariance matrix, our characterization states that an $n$-mode
quantum Gaussian state is PUN if and only if its $2n\times 2n$ quantum
covariance matrix $S$ commutes with the standard symplectic matrix $J$. It is
well-known that the so-called gauge-invariant Gaussian states are PUN, but
whether the converse is true is not known in the literature to the best of our
knowledge. We establish the converse in affirmation. Lastly, in terms of the
$(A,\Lambda)$-parameterization, we show that a Gaussian state with parameters
$(A,\Lambda)$ is PUN if and only if $A=0$.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:49:28 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"John",
"Tiju Cherian",
""
],
[
"Mishra",
"Hemant K.",
""
],
[
"Guha",
"Saikat",
""
]
] |
2504.20980 | Neil F. Johnson | Neil F. Johnson, Frank Yingjie Huo | Jekyll-and-Hyde Tipping Point in an AI's Behavior | null | null | null | null | cs.AI cs.CY nlin.AO physics.comp-ph physics.soc-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Trust in AI is undermined by the fact that there is no science that predicts
-- or that can explain to the public -- when an LLM's output (e.g. ChatGPT) is
likely to tip mid-response to become wrong, misleading, irrelevant or
dangerous. With deaths and trauma already being blamed on LLMs, this
uncertainty is even pushing people to treat their 'pet' LLM more politely to
'dissuade' it (or its future Artificial General Intelligence offspring) from
suddenly turning on them. Here we address this acute need by deriving from
first principles an exact formula for when a Jekyll-and-Hyde tipping point
occurs at LLMs' most basic level. Requiring only secondary school mathematics,
it shows the cause to be the AI's attention spreading so thin it suddenly
snaps. This exact formula provides quantitative predictions for how the
tipping-point can be delayed or prevented by changing the prompt and the AI's
training. Tailored generalizations will provide policymakers and the public
with a firm platform for discussing any of AI's broader uses and risks, e.g. as
a personal counselor, medical advisor, decision-maker for when to use force in
a conflict situation. It also meets the need for clear and transparent answers
to questions like ''should I be polite to my LLM?''
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:50:29 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Johnson",
"Neil F.",
""
],
[
"Huo",
"Frank Yingjie",
""
]
] |
2504.20981 | Micha{\l} Gawe{\l}czyk Dr. | M. Gawe{\l}czyk, K. Gawarecki | Optical Activity of Group III-V Quantum Dots Directly Embedded in
Silicon | 6 pages, 4 figures | null | null | null | cond-mat.mes-hall quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Optically active III-V group semiconductor quantum dots (QDs) are the leading
element of the upcoming safe quantum communication. However, the entire
electronic and IT infrastructure relies on silicon-based devices, with silicon
also providing a natural platform for photonic integration. Combining
semiconductor optics with silicon electronics is thus a major technological
challenge. This obstacle cannot be directly solved because silicon is optically
inactive. Interfacing III-V quantum dots with silicon is thus a sought-after
solution. A radical approach is to embed III-V material grains directly into
silicon. The first realization of such technology was developed, and it gave
InAs and core-shell InAs/GaAs QDs embedded in Si with bright and narrow
single-QD emission lines. No theory has been given, though, and, as we show
here, it is not even obvious if and how such QDs can be optically active. We
first use general arguments, also supported by atomistic calculations, that
InAs/Si QDs cannot confine both carrier types unless the structural strain is
mostly relaxed, meaning many defects at the interface. This explains the lack
of light emission from those dots. Then we show that the InAs/GaAs/Si QDs can
confine both carrier types. Their electron states are, however, highly
influenced by $k$-space valley mixing, which impacts emission spectra and
deteriorates optical properties. We propose to overcome this by adding an
additional wider-bandgap material layer.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:50:46 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Gawełczyk",
"M.",
""
],
[
"Gawarecki",
"K.",
""
]
] |
2504.20983 | Gianmarco Parretti | Giuseppe De Giacomo, Gianmarco Parretti, Shufang Zhu | LTLf Adaptive Synthesis for Multi-Tier Goals in Nondeterministic Domains | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | We study a variant of LTLf synthesis that synthesizes adaptive strategies for
achieving a multi-tier goal, consisting of multiple increasingly challenging
LTLf objectives in nondeterministic planning domains. Adaptive strategies are
strategies that at any point of their execution (i) enforce the satisfaction of
as many objectives as possible in the multi-tier goal, and (ii) exploit
possible cooperation from the environment to satisfy as many as possible of the
remaining ones. This happens dynamically: if the environment cooperates (ii)
and an objective becomes enforceable (i), then our strategies will enforce it.
We provide a game-theoretic technique to compute adaptive strategies that is
sound and complete. Notably, our technique is polynomial, in fact quadratic, in
the number of objectives. In other words, it handles multi-tier goals with only
a minor overhead compared to standard LTLf synthesis.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:53:16 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"De Giacomo",
"Giuseppe",
""
],
[
"Parretti",
"Gianmarco",
""
],
[
"Zhu",
"Shufang",
""
]
] |
2504.20985 | Inhyeok Choi | Inhyeok Choi | Acylindrically hyperbolic groups and counting problems | 18 pages, 2 figures. Comments welcome! | null | null | null | math.GR math.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We show that Morse elements are generic in acylindrically hyperbolic groups.
As an application, we observe that fully irreducible outer automorphisms are
generic in the outer automorphism group of a finite-rank free group.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:55:53 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Choi",
"Inhyeok",
""
]
] |
2504.20986 | Pei-Wei Tu | Pei-Wei Tu, Ji-Wei Xie, Di-Chang Chen, Ji-Lin Zhou | Planets Across Space and Time (PAST). VI. Age Dependence of the
Occurrence and Architecture of Ultra-Short-Period Planet Systems | authors' preprint, 47 pages, 4 figures, 21 supplementary figure, 2
supplementary table, published in Nature Astronomy, the published version is
available at in https://www.nature.com/articles/s41550-025-02539-1 | Nat Astron (2025) | 10.1038/s41550-025-02539-1 | null | astro-ph.EP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Ultra-short-period (USP) planets, with orbital periods shorter than one day,
represent a unique class of exoplanets whose origin remains puzzling.
Determining their age distribution and temporal evolution is vital for
uncovering their formation and evolutionary pathways. Using a sample of over
1,000 short-period planets around Sun-like stars, we find that the host stars
of USP planets are relatively older and have a higher prevalence in the
Galactic thick disk compared to stars hosting other short-period planets.
Furthermore, we find that the occurrence of USP planets increases with stellar
age and uncover evidence indicating that USP planetary system architectures
evolve on Gyr timescales. This includes a distinct dip-pileup in period
distributions around ~1 day and an expansion of orbital spacings with time. In
addition, younger USP planet systems are observed to have fewer multiple
transiting planets, implying fewer nearby companions and/or larger mutual
orbital inclinations. Our findings suggest that USP planets continuously form
through inward migration driven by tidal dissipation over Gyr timescales, and
that younger and older USP planets may have originated via different specific
tidal migration pathways.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:56:03 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Tu",
"Pei-Wei",
""
],
[
"Xie",
"Ji-Wei",
""
],
[
"Chen",
"Di-Chang",
""
],
[
"Zhou",
"Ji-Lin",
""
]
] |
2504.20987 | Cheryne Jonay | Cheryne Jonay, Frank Pollmann | Localized Fock Space Cages in Kinetically Constrained Models | 7 pages, 7 figures | null | null | null | quant-ph cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate a mechanism for non-ergodic behavior in many-body quantum
systems arising from destructive interference, leading to localization in Fock
space. Drawing parallels with single-particle flat-band localization and
Aharonov-Bohm cages, we identify conditions under which similar interference
effects in the many-body domain produce Fock space cages (FSCs)-highly
localized many-body eigenstates. By interpreting Fock space as a graph where
nodes represent bitstring basis states and edges denote non-zero transition
amplitudes of the Hamiltonian, we analyze different kinetically constrained
models. The FSCs cause non-ergodic dynamics when the system is initialized
within their support, highlighting a universal interference-driven localization
mechanism in many-body systems.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:56:25 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Jonay",
"Cheryne",
""
],
[
"Pollmann",
"Frank",
""
]
] |
2504.20988 | Atul Sharma | Atul Sharma, Kavindu Herath, Saurabh Bagchi, Chaoyue Liu, Somali
Chaterji | Hubs and Spokes Learning: Efficient and Scalable Collaborative Machine
Learning | null | null | null | null | cs.LG cs.AI cs.DC | http://creativecommons.org/licenses/by/4.0/ | We introduce the Hubs and Spokes Learning (HSL) framework, a novel paradigm
for collaborative machine learning that combines the strengths of Federated
Learning (FL) and Decentralized Learning (P2PL). HSL employs a two-tier
communication structure that avoids the single point of failure inherent in FL
and outperforms the state-of-the-art P2PL framework, Epidemic Learning Local
(ELL). At equal communication budgets (total edges), HSL achieves higher
performance than ELL, while at significantly lower communication budgets, it
can match ELL's performance. For instance, with only 400 edges, HSL reaches the
same test accuracy that ELL achieves with 1000 edges for 100 peers (spokes) on
CIFAR-10, demonstrating its suitability for resource-constrained systems. HSL
also achieves stronger consensus among nodes after mixing, resulting in
improved performance with fewer training rounds. We substantiate these claims
through rigorous theoretical analyses and extensive experimental results,
showcasing HSL's practicality for large-scale collaborative learning.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:56:55 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Sharma",
"Atul",
""
],
[
"Herath",
"Kavindu",
""
],
[
"Bagchi",
"Saurabh",
""
],
[
"Liu",
"Chaoyue",
""
],
[
"Chaterji",
"Somali",
""
]
] |
2504.20989 | L\'eo Monbroussou | L\'eo Monbroussou, Beatrice Polacchi, Verena Yacoub, Eugenio Caruccio,
Giovanni Rodari, Francesco Hoch, Gonzalo Carvacho, Nicol\`o Spagnolo, Taira
Giordani, Mattia Bossi, Abhiram Rajan, Niki Di Giano, Riccardo Albiero,
Francesco Ceccarelli, Roberto Osellame, Elham Kashefi, and Fabio Sciarrino | Photonic Quantum Convolutional Neural Networks with Adaptive State
Injection | null | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | Linear optical architectures have been extensively investigated for quantum
computing and quantum machine learning applications. Recently, proposals for
photonic quantum machine learning have combined linear optics with resource
adaptivity, such as adaptive circuit reconfiguration, which promises to enhance
expressivity and improve algorithm performances and scalability. Moreover,
linear optical platforms preserve some subspaces due to the fixed number of
particles during the computation, a property recently exploited to design a
novel quantum convolutional neural networks. This last architecture has shown
an advantage in terms of running time complexity and of the number of
parameters needed with respect to other quantum neural network proposals. In
this work, we design and experimentally implement the first photonic quantum
convolutional neural network (PQCNN) architecture based on particle-number
preserving circuits equipped with state injection, an approach recently
proposed to increase the controllability of linear optical circuits.
Subsequently, we experimentally validate the PQCNN for a binary image
classification on a photonic platform utilizing a semiconductor quantum
dot-based single-photon source and programmable integrated photonic
interferometers comprising 8 and 12 modes. In order to investigate the
scalability of the PQCNN design, we have performed numerical simulations on
datasets of different sizes. We highlight the potential utility of a simple
adaptive technique for a nonlinear Boson Sampling task, compatible with
near-term quantum devices.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:57:01 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Monbroussou",
"Léo",
""
],
[
"Polacchi",
"Beatrice",
""
],
[
"Yacoub",
"Verena",
""
],
[
"Caruccio",
"Eugenio",
""
],
[
"Rodari",
"Giovanni",
""
],
[
"Hoch",
"Francesco",
""
],
[
"Carvacho",
"Gonzalo",
""
],
[
"Spagnolo",
"Nicolò",
""
],
[
"Giordani",
"Taira",
""
],
[
"Bossi",
"Mattia",
""
],
[
"Rajan",
"Abhiram",
""
],
[
"Di Giano",
"Niki",
""
],
[
"Albiero",
"Riccardo",
""
],
[
"Ceccarelli",
"Francesco",
""
],
[
"Osellame",
"Roberto",
""
],
[
"Kashefi",
"Elham",
""
],
[
"Sciarrino",
"Fabio",
""
]
] |
2504.20990 | Claudia Ojeda-Aristizabal | Anise Mansour, Deanna Diaz, Movindu Kawshan Dissanayake, Erin
Henkhaus, Jungyoun Cho, Vinh Tran, Francisco Ramirez, Eric Corona-Oceguera,
Joshua Luna, Kenta Kodama, Yueyun Chen, Ho Chan, Jacob Weber, Blake Koford,
Patrick Barfield, Maya Martinez, Kenji Watanabe, Takashi Taniguchi, B. C.
Regan, Matthew Mecklenburg, Thomas Gredig, Claudia Ojeda-Aristizabal | Modification of the scattering mechanisms in bilayer graphene in
proximity to a molecular thin film probed in the mesoscopic regime | 10 pages, 9 figures | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum coherent effects can be probed in multilayer graphene through
electronic transport measurements at low temperatures. In particular, bilayer
graphene is known to be susceptible to quantum interference corrections of the
conductivity, presenting weak localization at all electronic densities, and
dependent on different scattering mechanisms as well as on the trigonal warping
of the electron dispersion near the K and K' valleys. Proximity effects with a
molecular thin film influence these scattering mechanisms, which can be
quantified through the known theory of magnetoconductance for bilayer graphene.
Here, we present weak localization measurements in a copper-phthalocyanine /
bilayer graphene / h-BN heterostructure that suggest an important suppression
of trigonal warping effects in bilayer graphene (BLG), restoring the
manifestation of the chirality of the charge carriers in the localization
properties of BLG. Additionally, we observe a charge transfer of
3.6$\times$10$^{12}$cm$^{-2}$ from the BLG to the molecules, as well as a very
small degradation of the mobility of the BLG/h-BN heterostructure upon the
deposition of copper phthalocyanine (CuPc). The molecular arrangement of the
CuPc thin film is characterized in a control sample through transmission
electron microscopy, that we relate to the electronic transport results.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:57:24 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Mansour",
"Anise",
""
],
[
"Diaz",
"Deanna",
""
],
[
"Dissanayake",
"Movindu Kawshan",
""
],
[
"Henkhaus",
"Erin",
""
],
[
"Cho",
"Jungyoun",
""
],
[
"Tran",
"Vinh",
""
],
[
"Ramirez",
"Francisco",
""
],
[
"Corona-Oceguera",
"Eric",
""
],
[
"Luna",
"Joshua",
""
],
[
"Kodama",
"Kenta",
""
],
[
"Chen",
"Yueyun",
""
],
[
"Chan",
"Ho",
""
],
[
"Weber",
"Jacob",
""
],
[
"Koford",
"Blake",
""
],
[
"Barfield",
"Patrick",
""
],
[
"Martinez",
"Maya",
""
],
[
"Watanabe",
"Kenji",
""
],
[
"Taniguchi",
"Takashi",
""
],
[
"Regan",
"B. C.",
""
],
[
"Mecklenburg",
"Matthew",
""
],
[
"Gredig",
"Thomas",
""
],
[
"Ojeda-Aristizabal",
"Claudia",
""
]
] |
2504.20992 | Giulio Fabbian | Giulio Fabbian, David Alonso, Kate Storey-Fisher, Thomas Cornish | Constraints on primordial non-Gaussianity from Quaia | Prepared for submission to JCAP, comments welcome! Material relevant
for this analysis is available at https://github.com/gfabbian/quaia-fnl | null | null | null | astro-ph.CO gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We analyse the large-scale angular clustering of quasars in the Gaia-unWISE
quasar catalog, Quaia, and their cross-correlation with maps of the lensing
convergence of the Cosmic Microwave Background (CMB), to constrain the level of
primordial non-Gaussianity (PNG). Specifically, we target the scale-dependent
bias that would be induced by PNG on biased tracers of the matter
inhomogeneities on large scales. The Quaia sample is particularly well suited
for this analysis, given the large effective volume covered, and our ability to
map out the main potential sources of systematic contamination and mitigate
their impact. Using the universality relation to characterise the response of
the quasar overdensity to PNG ($p_\phi=1$), we report constraints on the
local-type PNG parameter $f_{\rm NL}$ of $f_{\rm NL}=-20.5^{+19.0}_{-18.1}$
(68\% C.L.) by combining the quasar auto-correlation and its cross-correlation
with CMB lensing in two tomographic redshift bins (or $f_{\rm
NL}=-28.7^{+26.1}_{-24.6}$ if assuming a lower response for quasars,
$p_\phi=1.6$). Using the CMB lensing cross-correlations alone, we find $f_{\rm
NL}=-13.8^{+26.7}_{-25.0}$. These are the tightest constraints on $f_{\rm NL}$
to date from angular clustering statistics and cross-correlations with CMB
lensing.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:58:04 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Fabbian",
"Giulio",
""
],
[
"Alonso",
"David",
""
],
[
"Storey-Fisher",
"Kate",
""
],
[
"Cornish",
"Thomas",
""
]
] |
2504.20993 | Robert K{\l}opotek | Alina Landowska, Robert A. K{\l}opotek, Dariusz Filip, Konrad
Raczkowski | GDP-GFCF Dynamics Across Global Economies: A Comparative Study of Panel
Regressions and Random Forest | 31 pages, 5 figures, 6 tables, data and source code can be access via
OFS
https://osf.io/xpyhw/files/osfstorage?view_only=aa41c1d52ea54058b5f7d89aa3ea166c | null | null | null | econ.GN q-fin.EC | http://creativecommons.org/licenses/by/4.0/ | This study examines the relationship between GDP growth and Gross Fixed
Capital Formation (GFCF) across developed economies (G7, EU-15, OECD) and
emerging markets (BRICS). We integrate Random Forest machine learning
(non-linear regression) with traditional econometric models (linear regression)
to better capture non-linear interactions in investment analysis. Our findings
reveal that while GDP growth positively influences corporate investment, its
impact varies significantly by region. Developed economies show stronger
GDP-GFCF linkages due to stable financial systems, while emerging markets
demonstrate weaker connections due to economic heterogeneity and structural
constraints. Random Forest models indicate that GDP growth's importance is
lower than suggested by traditional econometrics, with lagged GFCF emerging as
the dominant predictor-confirming investment follows path-dependent patterns
rather than short-term GDP fluctuations. Regional variations in investment
drivers are substantial: taxation significantly influences developed economies
but minimally affects BRICS, while unemployment strongly drives investment in
BRICS but less so elsewhere. We introduce a parallelized p-value importance
algorithm for Random Forest that enhances computational efficiency while
maintaining statistical rigor through sequential testing methods (SPRT and
SAPT). The research demonstrates that hybrid methodologies combining machine
learning with econometric techniques provide more nuanced understanding of
investment dynamics, supporting region-specific policy design and improving
forecasting accuracy.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:58:47 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Landowska",
"Alina",
""
],
[
"Kłopotek",
"Robert A.",
""
],
[
"Filip",
"Dariusz",
""
],
[
"Raczkowski",
"Konrad",
""
]
] |
2504.20994 | Gonzalo Villa | Anish Ghoshal, Filippo Revello and Gonzalo Villa | Cosmic superstrings in large volume compactifications: PTAs, LISA and
time-varying tension | 40 pages, 5 figures | null | null | null | astro-ph.CO hep-ph hep-th | http://creativecommons.org/licenses/by/4.0/ | The Stochastic Gravitational Wave Background (SGWB) from cosmic superstrings
offers one of the few known possibilities to test String Theory within current
experimental reach. However, in order to be compatible with the existing
constraints, the tension of a cosmic superstring network is required to lie
several orders of magnitude below the Planck scale. This is naturally realized
in string compactifications where the volume of the extra dimensions is
parametrically large (in string units). We estimate the GW spectrum arising
from cosmic superstrings in such scenarios, providing analytical formulae as
well as numerical results. Crucially, we do so within a fully-fledged string
cosmology, taking into account various modified cosmological epochs (such as
kination or early matter domination) induced by the presence of moduli and a
time-dependent string tension. We show that part of the spectrum generically
lies within reach of LISA and ET, with a large class of models predicting a
characteristic drop in the amplitude which may be robustly probed by LISA. The
corresponding signal would encode information on the dynamics of moduli and
reheating. On the other hand, the ultra-high frequency part of the spectrum can
be significantly enhanced by a long, early phase of kination with time-varying
tension, yielding a spectral index unique to this set-up.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:59:23 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Ghoshal",
"Anish",
""
],
[
"Revello",
"Filippo",
""
],
[
"Villa",
"Gonzalo",
""
]
] |
2504.20995 | Haoyu Zhen | Haoyu Zhen, Qiao Sun, Hongxin Zhang, Junyan Li, Siyuan Zhou, Yilun Du,
Chuang Gan | TesserAct: Learning 4D Embodied World Models | Project Page: https://tesseractworld.github.io/ | null | null | null | cs.CV cs.RO | http://creativecommons.org/licenses/by/4.0/ | This paper presents an effective approach for learning novel 4D embodied
world models, which predict the dynamic evolution of 3D scenes over time in
response to an embodied agent's actions, providing both spatial and temporal
consistency. We propose to learn a 4D world model by training on RGB-DN (RGB,
Depth, and Normal) videos. This not only surpasses traditional 2D models by
incorporating detailed shape, configuration, and temporal changes into their
predictions, but also allows us to effectively learn accurate inverse dynamic
models for an embodied agent. Specifically, we first extend existing robotic
manipulation video datasets with depth and normal information leveraging
off-the-shelf models. Next, we fine-tune a video generation model on this
annotated dataset, which jointly predicts RGB-DN (RGB, Depth, and Normal) for
each frame. We then present an algorithm to directly convert generated RGB,
Depth, and Normal videos into a high-quality 4D scene of the world. Our method
ensures temporal and spatial coherence in 4D scene predictions from embodied
scenarios, enables novel view synthesis for embodied environments, and
facilitates policy learning that significantly outperforms those derived from
prior video-based world models.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:59:30 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Zhen",
"Haoyu",
""
],
[
"Sun",
"Qiao",
""
],
[
"Zhang",
"Hongxin",
""
],
[
"Li",
"Junyan",
""
],
[
"Zhou",
"Siyuan",
""
],
[
"Du",
"Yilun",
""
],
[
"Gan",
"Chuang",
""
]
] |
2504.20996 | Sicheng Mo | Sicheng Mo, Thao Nguyen, Xun Huang, Siddharth Srinivasan Iyer, Yijun
Li, Yuchen Liu, Abhishek Tandon, Eli Shechtman, Krishna Kumar Singh, Yong Jae
Lee, Bolei Zhou, Yuheng Li | X-Fusion: Introducing New Modality to Frozen Large Language Models | Project Page: https://sichengmo.github.io/XFusion/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We propose X-Fusion, a framework that extends pretrained Large Language
Models (LLMs) for multimodal tasks while preserving their language
capabilities. X-Fusion employs a dual-tower design with modality-specific
weights, keeping the LLM's parameters frozen while integrating vision-specific
information for both understanding and generation. Our experiments demonstrate
that X-Fusion consistently outperforms alternative architectures on both
image-to-text and text-to-image tasks. We find that incorporating
understanding-focused data improves generation quality, reducing image data
noise enhances overall performance, and feature alignment accelerates
convergence for smaller models but has minimal impact on larger ones. Our
findings provide valuable insights into building efficient unified multimodal
models.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:59:45 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Mo",
"Sicheng",
""
],
[
"Nguyen",
"Thao",
""
],
[
"Huang",
"Xun",
""
],
[
"Iyer",
"Siddharth Srinivasan",
""
],
[
"Li",
"Yijun",
""
],
[
"Liu",
"Yuchen",
""
],
[
"Tandon",
"Abhishek",
""
],
[
"Shechtman",
"Eli",
""
],
[
"Singh",
"Krishna Kumar",
""
],
[
"Lee",
"Yong Jae",
""
],
[
"Zhou",
"Bolei",
""
],
[
"Li",
"Yuheng",
""
]
] |
2504.20997 | Dilip Arumugam | Dilip Arumugam and Thomas L. Griffiths | Toward Efficient Exploration by Large Language Model Agents | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | A burgeoning area within reinforcement learning (RL) is the design of
sequential decision-making agents centered around large language models (LLMs).
While autonomous decision-making agents powered by modern LLMs could facilitate
numerous real-world applications, such successes demand agents that are capable
of data-efficient RL. One key obstacle to achieving data efficiency in RL is
exploration, a challenge that we demonstrate many recent proposals for LLM
agent designs struggle to contend with. Meanwhile, classic algorithms from the
RL literature known to gracefully address exploration require technical
machinery that can be challenging to operationalize in purely natural language
settings. In this work, rather than relying on finetuning or in-context
learning to coax LLMs into implicitly imitating a RL algorithm, we illustrate
how LLMs can be used to explicitly implement an existing RL algorithm
(Posterior Sampling for Reinforcement Learning) whose capacity for
statistically-efficient exploration is already well-studied. We offer empirical
results demonstrating how our LLM-based implementation of a known,
data-efficient RL algorithm can be considerably more effective in natural
language tasks that demand prudent exploration.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:59:48 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Arumugam",
"Dilip",
""
],
[
"Griffiths",
"Thomas L.",
""
]
] |
2504.20998 | Thao Nguyen | Thao Nguyen, Krishna Kumar Singh, Jing Shi, Trung Bui, Yong Jae Lee,
Yuheng Li | YoChameleon: Personalized Vision and Language Generation | CVPR 2025; Project page: https://thaoshibe.github.io/YoChameleon | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large Multimodal Models (e.g., GPT-4, Gemini, Chameleon) have evolved into
powerful tools with millions of users. However, they remain generic models and
lack personalized knowledge of specific user concepts. Previous work has
explored personalization for text generation, yet it remains unclear how these
methods can be adapted to new modalities, such as image generation. In this
paper, we introduce Yo'Chameleon, the first attempt to study personalization
for large multimodal models. Given 3-5 images of a particular concept,
Yo'Chameleon leverages soft-prompt tuning to embed subject-specific information
to (i) answer questions about the subject and (ii) recreate pixel-level details
to produce images of the subject in new contexts. Yo'Chameleon is trained with
(i) a self-prompting optimization mechanism to balance performance across
multiple modalities, and (ii) a ``soft-positive" image generation approach to
enhance image quality in a few-shot setting.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:59:57 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Nguyen",
"Thao",
""
],
[
"Singh",
"Krishna Kumar",
""
],
[
"Shi",
"Jing",
""
],
[
"Bui",
"Trung",
""
],
[
"Lee",
"Yong Jae",
""
],
[
"Li",
"Yuheng",
""
]
] |
2504.20999 | Amadeo Castro-Gonz\'alez | A. Castro-Gonz\'alez, F. Bouchy, A. C. M. Correia, A. Sozzetti, J.
Lillo-Box, P. Figueira, B. Lavie, C. Lovis, M. J. Hobson, S. G. Sousa, V.
Adibekyan, M. R. Standing, N. C. Hara, D. Barrado, A. M. Silva, V. Bourrier,
J. Korth, N. C. Santos, M. Damasso, M. R. Zapatero Osorio, J. Rodrigues, Y.
Alibert, S. Cristiani, P. Di Marcantonio, J. I. Gonz\'alez Hern\'andez, G. Lo
Curto, C. J. A. P. Martins, N. J. Nunes, E. Palle, F. Pepe, A. Su\'arez
Mascare\~no, and H. M. Tabernero | Two neighbours to the ultra-short-period Earth-sized planet K2-157 b in
the warm Neptunian savanna | Accepted for publication in Astronomy & Astrophysics. Abstract
shortened. 31 pages, 28 figures | null | null | null | astro-ph.EP | http://creativecommons.org/licenses/by/4.0/ | The formation and evolution of ultra-short-period (USP) planets is poorly
understood. However, it is widely thought that these planets migrated inwards
through interactions with outer neighbours. We aimed to confirm and
characterize the USP Earth-sized planet K2-157 b ($P_{\rm orb}$ = 8.8 h). To do
so, We measured 49 radial velocities (RVs) with the ESPRESSO spectrograph and
derived the properties of the system through a RV and transit model. We detect
two additional super-Neptune-mass planets within the warm Neptunian savanna,
K2-157 c ($P_{\rm orb, c}$ = $25.942^{+0.045}_{-0.044}$ d, $M_{\rm p, c} \,
\textrm{sin} \, i$ = $30.8 \pm 1.9$ $\rm M_{\oplus}$), and K2-157 d ($P_{\rm
orb, d}$ = $66.50^{+0.71}_{-0.59}$ d, $M_{\rm p,d} \, \textrm{sin}\,i$ = $23.3
\pm 2.5$ $\rm M_{\oplus}$). The mass of K2-157 b, $M_{\rm p,b}$ =
$1.14^{+0.39}_{-0.41}$ $\rm M_{\oplus}$ ($<$ 2.3 $\rm M_{\oplus}$ at
3$\sigma$), together with its radius, $R_{\rm p}$ = 0.935 $\pm$ 0.090 $\rm
R_{\oplus}$, make the planet compatible with a rocky composition. K2 data
discard non-grazing transit configurations for K2-157 c ($i_{\rm c}$ $<$
88.4$^{\circ}$ at 3$\sigma$), and ESPRESSO data constrain the eccentricities of
K2-157 c and K2-157 d to $e_{\rm c}$ $<$ 0.2 and $e_{\rm d}$ $<$ 0.5 at
3$\sigma$. At a population level, we find that the trend that the closest USP
planets tend to orbit late-type stars does not hold when scaling the orbital
separation to the Roche limit, which suggests that the orbital distribution of
the closest planets across spectral types is determined by tidal disruption.
The orbital architecture of K2-157 is unusual, with only one similar case
reported to date: 55 Cnc. The USP planets of these systems, being accompanied
by massive, long-period, relatively spaced, and possibly misaligned neighbours,
could have migrated inwards through eccentricity-based mechanisms triggered by
secular interactions.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:59:57 GMT"
}
] | 2025-04-30T00:00:00 | [
[
"Castro-González",
"A.",
""
],
[
"Bouchy",
"F.",
""
],
[
"Correia",
"A. C. M.",
""
],
[
"Sozzetti",
"A.",
""
],
[
"Lillo-Box",
"J.",
""
],
[
"Figueira",
"P.",
""
],
[
"Lavie",
"B.",
""
],
[
"Lovis",
"C.",
""
],
[
"Hobson",
"M. J.",
""
],
[
"Sousa",
"S. G.",
""
],
[
"Adibekyan",
"V.",
""
],
[
"Standing",
"M. R.",
""
],
[
"Hara",
"N. C.",
""
],
[
"Barrado",
"D.",
""
],
[
"Silva",
"A. M.",
""
],
[
"Bourrier",
"V.",
""
],
[
"Korth",
"J.",
""
],
[
"Santos",
"N. C.",
""
],
[
"Damasso",
"M.",
""
],
[
"Osorio",
"M. R. Zapatero",
""
],
[
"Rodrigues",
"J.",
""
],
[
"Alibert",
"Y.",
""
],
[
"Cristiani",
"S.",
""
],
[
"Di Marcantonio",
"P.",
""
],
[
"Hernández",
"J. I. González",
""
],
[
"Curto",
"G. Lo",
""
],
[
"Martins",
"C. J. A. P.",
""
],
[
"Nunes",
"N. J.",
""
],
[
"Palle",
"E.",
""
],
[
"Pepe",
"F.",
""
],
[
"Mascareño",
"A. Suárez",
""
],
[
"Tabernero",
"H. M.",
""
]
] |
0705.2576 | Michael Frank | Michael Frank and Kamran Sharifi | Adjointability of densely defined closed operators and the
Magajna-Schweizer Theorem | 13 pages | J. Operator Theory 63(2010), no. 2, 271-282 | null | SFB 478 - Geometrical Structures in Mathematics - SFB preprint
series v. 466(2007), Univ. M | math.OA math-ph math.FA math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this notes unbounded regular operators on Hilbert $C^*$-modules over
arbitrary $C^*$-algebras are discussed. A densely defined operator $t$
possesses an adjoint operator if the graph of $t$ is an orthogonal summand.
Moreover, for a densely defined operator $t$ the graph of $t$ is orthogonally
complemented and the range of $P_FP_{G(t)^\bot}$ is dense in its biorthogonal
complement if and only if $t$ is regular. For a given $C^*$-algebra $\mathcal
A$ any densely defined $\mathcal A$-linear closed operator $t$ between Hilbert
$C^*$-modules is regular, if and only if any densely defined $\mathcal
A$-linear closed operator $t$ between Hilbert $C^*$-modules admits a densely
defined adjoint operator, if and only if $\mathcal A$ is a $C^*$-algebra of
compact operators. Some further characterizations of closed and regular modular
operators are obtained.
Changes 1: Improved results, corrected misprints, added references. Accepted
by J. Operator Theory, August 2007 / Changes 2: Filled gap in the proof of Thm.
3.1, changes in the formulations of Cor. 3.2 and Thm. 3.4, updated references
and address of the second author.
| [
{
"version": "v1",
"created": "Thu, 17 May 2007 18:07:27 GMT"
},
{
"version": "v2",
"created": "Wed, 29 Aug 2007 18:46:42 GMT"
},
{
"version": "v3",
"created": "Fri, 22 Jan 2010 16:55:35 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Frank",
"Michael",
""
],
[
"Sharifi",
"Kamran",
""
]
] |
0710.0586 | Michael Frank | Michael Frank, Alexander A. Pavlov | Strict essential extensions of C*-algebras and Hilbert C*-modules | 12 pages | J. Operator Theory 64(2010), no. 1, 103-116 | null | null | math.OA | null | In the present paper we develop both ideas of D. Baki\'c and B. Gulja{\v{s}}
and the categorical approach to multipliers from E.C. Lance's book and
publications of the second author, for the introduction and study of left
multipliers of Hilbert $C^*$-modules. Some properties and, in particular, the
property of maximality among all strictly essential extensions of a Hilbert
$C^*$-module for left multipliers are proved. Also relations between left
essential and left strictly essential extensions in different contexts are
obtained. Left essential and left strictly essential extensions of matrix
algebras are considered. In the final paragraph the topological approach to the
left multiplier theory of Hilbert $C^*$-modules is worked out.
| [
{
"version": "v1",
"created": "Tue, 2 Oct 2007 17:05:08 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Frank",
"Michael",
""
],
[
"Pavlov",
"Alexander A.",
""
]
] |
0806.0162 | Michael Frank | Michael Frank, Kamran Sharifi | Generalized inverses and polar decomposition of unbounded regular
operators on Hilbert $C^*$-modules | 11 pages / corrected typos and cross-references | J. Operator Theory 64(2010), no. 2, 377-386 | null | null | math.OA math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this note we show that an unbounded regular operator $t$ on Hilbert
$C^*$-modules over an arbitrary $C^*$ algebra $ \mathcal{A}$ has polar
decomposition if and only if the closures of the ranges of $t$ and $|t|$ are
orthogonally complemented, if and only if the operators $t$ and $t^*$ have
unbounded regular generalized inverses. For a given $C^*$-algebra $
\mathcal{A}$ any densely defined $\mathcal A$-linear closed operator $t$
between Hilbert $C^*$-modules has polar decomposition, if and only if any
densely defined $\mathcal A$-linear closed operator $t$ between Hilbert
$C^*$-modules has generalized inverse, if and only if $\mathcal A$ is a
$C^*$-algebra of compact operators.
| [
{
"version": "v1",
"created": "Sun, 1 Jun 2008 18:35:54 GMT"
},
{
"version": "v2",
"created": "Sun, 22 Aug 2010 12:50:05 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Frank",
"Michael",
""
],
[
"Sharifi",
"Kamran",
""
]
] |
0907.2983 | Michael Frank | Michael Frank, Alexander S. Mishchenko, Alexander A. Pavlov | Orthogonality-preserving, C*-conformal and conformal module mappings on
Hilbert C*-modules | 13 pages / Thm. 1.3, second paragraph of proof - corrected, minor
changes of formulations in the text, references updated | J. Funct. Anal. 260(2011), 327-339 | null | null | math.OA math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate orthonormality-preserving, C*-conformal and conformal module
mappings on Hilbert C*-modules to obtain their general structure.
Orthogonality-preserving bounded module maps T act as a multiplication by an
element \lambda of the center of the multiplier algebra of the C*-algebra of
coefficients combined with an isometric module operator as long as some polar
decomposition conditions for the specific element \lambda are fulfilled inside
that multiplier algebra. Generally, T always fulfils the equality $<T(x),T(y) >
= | \lambda |^2 < x,y>$ for any elements x,y of the Hilbert C*-module. At the
contrary, C*-conformal and conformal bounded C*-linear mappings are shown to be
only the positive real multiples of isometric module operators.
| [
{
"version": "v1",
"created": "Fri, 17 Jul 2009 07:40:08 GMT"
},
{
"version": "v2",
"created": "Sun, 30 Aug 2009 14:57:54 GMT"
},
{
"version": "v3",
"created": "Tue, 27 Jul 2010 08:26:26 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Frank",
"Michael",
""
],
[
"Mishchenko",
"Alexander S.",
""
],
[
"Pavlov",
"Alexander A.",
""
]
] |
1101.4922 | Anthony O'Farrell | Stephen M. Buckley and Anthony G. O'Farrell | Wiring Switches to Light Bulbs | 22 pages, 14 figures | Irish Math Soc Bulletin 94 Winter 2024, 69-88 | 10.33232/BIMS.0094.69.88 | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Given n buttons and n bulbs so that the ith button toggles the ith bulb and
at most two other bulbs, we compute the sharp lower bound on the number of
bulbs that can be lit regardless of the action of the buttons.
| [
{
"version": "v1",
"created": "Tue, 25 Jan 2011 20:33:35 GMT"
},
{
"version": "v2",
"created": "Mon, 30 Sep 2024 14:54:34 GMT"
},
{
"version": "v3",
"created": "Sat, 5 Oct 2024 10:27:52 GMT"
},
{
"version": "v4",
"created": "Mon, 16 Dec 2024 12:29:57 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Buckley",
"Stephen M.",
""
],
[
"O'Farrell",
"Anthony G.",
""
]
] |
1504.06293 | Mohammad Sal Moslehian | Mohammad Sal Moslehian, Ali Zamani and Michael Frank | Angle Preserving Mappings | 15 pages, to appear in Z. Anal. Anwend | Z. Anal. Anwend. 34 (2015), no. 4, 485-500 | 10.4171/ZAA/1551 | null | math.FA math.OA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we give some characterizations of orthogonality preserving
mappings between inner product spaces. Furthermore, we study the linear
mappings that preserve some angles. One of our main results states that if
$\mathcal{X}, \mathcal{Y}$ are real inner product spaces and $\theta\in(0,
\pi)$, then an injective nonzero linear mapping $T:\mathcal{X}\longrightarrow
\mathcal{Y}$ is a similarity whenever (i) $x\underset{\theta}{\angle} y\,
\Leftrightarrow \,Tx\underset{\theta}{\angle} Ty$ for all $x, y\in
\mathcal{X}$; (ii) for all $x, y\in \mathcal{X}$, $\|x\|=\|y\|$ and
$x\underset{\theta}{\angle} y$ ensure that $\|Tx\|=\|Ty\|$. We also investigate
orthogonality preserving mappings in the setting of inner product
$C^{*}$-modules. Another result shows that if
$\mathbb{K}(\mathscr{H})\subseteq\mathscr{A}\subseteq\mathbb{B}(\mathscr{H})$
is a $C^{*}$-algebra and $T\,:\mathscr{E}\longrightarrow \mathscr{F}$ is an
$\mathscr{A}$-linear mapping between inner product $\mathscr{A}$-modules, then
$T$ is orthogonality preserving if and only if $|x|\leq|y|\, \Rightarrow
\,|Tx|\leq|Ty|$ for all $x, y\in \mathscr{E}$.
| [
{
"version": "v1",
"created": "Thu, 23 Apr 2015 18:39:18 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Moslehian",
"Mohammad Sal",
""
],
[
"Zamani",
"Ali",
""
],
[
"Frank",
"Michael",
""
]
] |
1511.05134 | Pierre Portal | Pascal Auscher, Sylvie Monniaux, Pierre Portal | On existence and uniqueness for non-autonomous parabolic Cauchy problems
with rough coefficients | Added a remark before Proposition 2.12 to explain that its proof has
a gap, but that the gap has been closed in a recent paper, and that all
results in the present paper remain valid | null | null | null | math.AP math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider existence and uniqueness issues for the initial value problem of
parabolic equations $\partial_{t} u = {\rm div} A \nabla u$ on the upper half
space, with initial data in $L^p$ spaces. The coefficient matrix $A$ is assumed
to be uniformly elliptic, but merely bounded measurable in space and time. For
real coefficients and a single equation, this is an old topic for which a
comprehensive theory is available, culminating in the work of Aronson. Much
less is understood for complex coefficients or systems of equations except for
the work of Lions, mainly because of the failure of maximum principles. In this
paper, we come back to this topic with new methods that do not rely on maximum
principles. This allows us to treat systems in this generality when $p\geq 2$,
or under certain assumptions such as bounded variation in the time variable (a
much weaker assumption that the usual H\"older continuity assumption) when $p<
2$. We reobtain results for real coefficients, and also complement them. For
instance, we obtain uniqueness for arbitrary $L^p$ data, $1\leq p \leq \infty$,
in the class $L^\infty(0,T; L^p({\mathbb{R}}^n))$. Our approach to the
existence problem relies on a careful construction of propagators for an
appropriate energy space, encompassing previous constructions. Our approach to
the uniqueness problem, the most novel aspect here, relies on a parabolic
version of the Kenig-Pipher maximal function, used in the context of elliptic
equations on non-smooth domains. We also prove comparison estimates involving
conical square functions of Lusin type and prove some Fatou type results about
non-tangential convergence of solutions. Recent results on maximal regularity
operators in tent spaces that do not require pointwise heat kernel bounds are
key tools in this study.
| [
{
"version": "v1",
"created": "Sat, 14 Nov 2015 20:11:25 GMT"
},
{
"version": "v2",
"created": "Thu, 19 Nov 2015 06:32:26 GMT"
},
{
"version": "v3",
"created": "Sun, 27 Apr 2025 17:01:13 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Auscher",
"Pascal",
""
],
[
"Monniaux",
"Sylvie",
""
],
[
"Portal",
"Pierre",
""
]
] |
1612.01347 | Chi Xiong | Hwee-Boon Low and Chi Xiong | Fantasia of a Superfluid Universe -- In memory of Kerson Huang | 15 pages, 10 figures | Memorial Volume for Kerson Huang, Edited by K. K. Phua, H. B. Low,
and C. Xiong, World Scientific Publishing, 2017, 55-69, ISBN
978-981-3207-42-4 | 10.1142/9789813207431_0010 | null | astro-ph.CO gr-qc hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This article introduces Kerson Huang's theory on superfluid universe in these
aspects: I. choose the asymptotically free Halpern-Huang scalar field(s) to
drive inflation; II. use quantum turbulence to create matter; III. consider
dark energy as the energy density of the cosmic superfluid and dark matter the
deviation of the superfluid density from its equilibrium value; IV. use quantum
vorticity to explain phenomena such as the non-thermal filaments at the
galactic center, the large voids in the galactic distribution, and the
gravitational collapse of stars to fast-rotating blackholes.
| [
{
"version": "v1",
"created": "Wed, 30 Nov 2016 20:41:09 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Low",
"Hwee-Boon",
""
],
[
"Xiong",
"Chi",
""
]
] |
1612.07638 | Quy Pham Hung | Nguyen Tu Cuong, Pham Hung Quy | On the structure of finitely generated modules over quotients of
Cohen-Macaulay local rings | 27 pages, Journal of Pure and Applied Algebra | null | null | null | math.AC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $(R, \frak m)$ be a homomorphic image of a Cohen-Macaulay local ring and
$M$ a finitely generated $R$-module. We use the splitting of local cohomology
to shed a new light on the structure of non-Cohen-Macaulay modules. Namely, we
show that every finitely generated $R$-module $M$ is associated by a sequence
of invariant modules. This modules sequence expresses the deviation of $M$ with
the Cohen-Macaulay property. This result generalizes the unmixed theorem of
Cohen-Macaulayness for any finitely generated $R$-module. As an application we
construct a new extended degree in sense of Vasconcelos.
| [
{
"version": "v1",
"created": "Thu, 22 Dec 2016 15:13:28 GMT"
},
{
"version": "v2",
"created": "Mon, 17 Apr 2017 02:57:08 GMT"
},
{
"version": "v3",
"created": "Tue, 20 Jun 2017 15:24:51 GMT"
},
{
"version": "v4",
"created": "Mon, 28 Apr 2025 09:03:50 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Cuong",
"Nguyen Tu",
""
],
[
"Quy",
"Pham Hung",
""
]
] |
1711.04724 | Michael Frank | Michael Frank, M. S. Moslehian, and Ali Zamani | Orthogonality preserving property for pairs of operators on Hilbert
$C^*$-modules | 23 pages, In this last revision several new examples are added and
some minor changes appeared in the text. To appear in Aequat. Math | Aequat. Math. 95 (2021), 867-887 | 10.1007/s00010-021-00790-1 | null | math.OA math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the orthogonality preserving property for pairs of mappings on
inner product $C^*$-modules extending existing results for a single
orthogonality-preserving mapping. Guided by the point of view that the
$C^*$-valued inner product structure of a Hilbert $C^*$-module is determined
essentially by the module structure and by the orthogonality structure, pairs
of linear and local orthogonality-preserving mappings are investigated, not a
priori bounded. The intuition is that most often $C^*$-linearity and
boundedness can be derived from the settings under consideration. In
particular, we obtain that if $\mathscr{A}$ is a $C^{*}$-algebra and $T,
S:\mathscr{E}\longrightarrow \mathscr{F}$ are two bounded ${\mathscr A}$-linear
mappings between full Hilbert $\mathscr{A}$-modules, then $\langle x, y\rangle
= 0$ implies $\langle T(x), S(y)\rangle = 0$ for all $x, y\in \mathscr{E}$ if
and only if there exists an element $\gamma$ of the center $Z(M({\mathscr A}))$
of the multiplier algebra $M({\mathscr A})$ of ${\mathscr A}$ such that
$\langle T(x), S(y)\rangle = \gamma \langle x, y\rangle$ for all $x, y\in
\mathscr{E}$. In particular, for adjointable operators $S$ we have
$T=(S^*)^{-1}$, and any bounded invertible module operator $T$ may appear.
Varying the conditions on the mappings $T$ and $S$ we obtain further
affirmative results for local operators and for pairs of a bounded and of an
unbounded module operator with bounded inverse, among others. Also, unbounded
operators with disjoint ranges are considered. The proving techniques give new
insights.
| [
{
"version": "v1",
"created": "Fri, 10 Nov 2017 13:09:19 GMT"
},
{
"version": "v2",
"created": "Tue, 8 Oct 2019 19:05:12 GMT"
},
{
"version": "v3",
"created": "Tue, 27 Apr 2021 12:37:54 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Frank",
"Michael",
""
],
[
"Moslehian",
"M. S.",
""
],
[
"Zamani",
"Ali",
""
]
] |
1802.01007 | Pawe{\l} Pietrzycki | Pawe{\l} Pietrzycki | Reduced commutativity of moduli of operators | null | Linear Algebra and Its Applications vol. 557 (2018), 375-402 | 10.1016/j.laa.2018.08.007 | null | math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we investigate the question of when the equations
$A^{*s}A^{s}=(A^{*}A)^{s}$ for all $s \in S$, where $S$ is a finite set of
positive integers, imply the quasinormality or normality of $A$. In particular,
it is proved that if $S=\{p,m,m+p,n,n+p\}$, where $2\leq m < n$, then $A$ is
quasinormal. Moreover, if $A$ is invertible and $S=\{m,n,n+m\}$, where $m \leq
n$, then $A$ is normal. Furthermore, the case when $S=\{m,m+n\}$ and $A^{*n}A^n
\leq (A^*A)^n$ is discussed.
| [
{
"version": "v1",
"created": "Sat, 3 Feb 2018 17:45:55 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Pietrzycki",
"Paweł",
""
]
] |
1804.02500 | Matt Visser | Matt Visser (Victoria University of Wellington) | Variants on Andrica's conjecture with and without the Riemann hypothesis | V1: 12 pages; V2: 9 pages. Discussion simplified and streamlined.
Various numerical constants improved. Extra section added on what can be
proved unconditionally. Updated discussion on using maximal prime gaps to
verify the standard Andrica conjecture up to 1.8 x 10^{19}. V3: 10 pages; 4
references added; some cosmetic changes; more discussion of numerics; closely
resembles published version | Mathematics Vol 6, No 12, (2018), 289 | 10.3390/math6120289 | null | math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The gap between what we can explicitly prove regarding the distribution of
primes and what we suspect regarding the distribution of primes is enormous. It
is (reasonably) well-known that the Riemann hypothesis is not sufficient to
prove Andrica's conjecture: $\forall n\geq 1$, is $\sqrt{p_{n+1}}-\sqrt{p_n}
\leq 1$? But can one at least get tolerably close? I shall first show that with
a logarithmic modification, provided one assumes the Riemann hypothesis, one
has \[ {\sqrt{p_{n+1}}\over\ln p_{n+1}} -{\sqrt{p_n}\over\ln p_n} <
{11\over25}; \qquad (n\geq1). \] Then, by considering more general $m^{th}$
roots, again assuming the Riemann hypothesis, I shall show that \[
{\sqrt[m]{p_{n+1}}} -{\sqrt[m]{p_n}} < {44\over25 \,e\, (m-2)}; \qquad (n\geq
3;\; m >2). \] In counterpoint, if we limit ourselves to what we can currently
prove unconditionally, then the only explicit Andrica-like results seem to be
variants on the results below: \[ \ln^2 p_{n+1} - \ln^2 p_n < 9; \qquad
(n\geq1). \] \[ \ln^3 p_{n+1} - \ln^3 p_n < 52; \qquad (n\geq1). \] \[ \ln^4
p_{n+1} - \ln^4 p_n < 991; \qquad (n\geq1). \] I shall also slightly update the
region on which Andrica's conjecture is unconditionally verified.
| [
{
"version": "v1",
"created": "Sat, 7 Apr 2018 03:26:57 GMT"
},
{
"version": "v2",
"created": "Tue, 18 Sep 2018 22:05:49 GMT"
},
{
"version": "v3",
"created": "Wed, 28 Nov 2018 20:06:23 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Visser",
"Matt",
"",
"Victoria University of Wellington"
]
] |
1810.12558 | Mahammad Humayoo | Mahammad Humayoo, Gengzhong Zheng, Xiaoqing Dong, Liming Miao, Shuwei
Qiu, Zexun Zhou, Peitao Wang, Zakir Ullah, Naveed Ur Rehman Junejo, Xueqi
Cheng | Relative Importance Sampling for off-Policy Actor-Critic in Deep
Reinforcement Learning | null | null | null | null | cs.LG cs.AI stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Off-policy learning exhibits greater instability when compared to on-policy
learning in reinforcement learning (RL). The difference in probability
distribution between the target policy ($\pi$) and the behavior policy (b) is a
major cause of instability. High variance also originates from distributional
mismatch. The variation between the target policy's distribution and the
behavior policy's distribution can be reduced using importance sampling (IS).
However, importance sampling has high variance, which is exacerbated in
sequential scenarios. We propose a smooth form of importance sampling,
specifically relative importance sampling (RIS), which mitigates variance and
stabilizes learning. To control variance, we alter the value of the smoothness
parameter $\beta\in[0, 1]$ in RIS. We develop the first model-free relative
importance sampling off-policy actor-critic (RIS-off-PAC) algorithms in RL
using this strategy. Our method uses a network to generate the target policy
(actor) and evaluate the current policy ($\pi$) using a value function (critic)
based on behavior policy samples. Our algorithms are trained using behavior
policy action values in the reward function, not target policy ones. Both the
actor and critic are trained using deep neural networks. Our methods performed
better than or equal to several state-of-the-art RL benchmarks on OpenAI Gym
challenges and synthetic datasets.
| [
{
"version": "v1",
"created": "Tue, 30 Oct 2018 07:41:08 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Jan 2019 07:13:50 GMT"
},
{
"version": "v3",
"created": "Wed, 8 May 2019 00:27:31 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Jul 2019 02:42:41 GMT"
},
{
"version": "v5",
"created": "Tue, 16 Jul 2019 01:35:44 GMT"
},
{
"version": "v6",
"created": "Fri, 19 Jul 2019 03:08:45 GMT"
},
{
"version": "v7",
"created": "Wed, 21 Dec 2022 14:31:52 GMT"
},
{
"version": "v8",
"created": "Thu, 22 Dec 2022 14:39:34 GMT"
},
{
"version": "v9",
"created": "Mon, 28 Apr 2025 05:15:26 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Humayoo",
"Mahammad",
""
],
[
"Zheng",
"Gengzhong",
""
],
[
"Dong",
"Xiaoqing",
""
],
[
"Miao",
"Liming",
""
],
[
"Qiu",
"Shuwei",
""
],
[
"Zhou",
"Zexun",
""
],
[
"Wang",
"Peitao",
""
],
[
"Ullah",
"Zakir",
""
],
[
"Junejo",
"Naveed Ur Rehman",
""
],
[
"Cheng",
"Xueqi",
""
]
] |
1812.02762 | Matt Visser | Matt Visser (Victoria University of Wellington) | Strong version of Andrica's conjecture | V1: 8 pages. Sequences: A005250 A002386 A005669 A000101 A107578
A050216 A050216; V2: Now 11 pages; some discussion and references added. V3:
Still 11 pages. Updated to reflect best currently known bound on the location
of the 81st maximal prime gap (somewhere above 2^{64}) | International Mathematical Forum, Vol. 14, no. 4, (2019) 181 - 188 | 10.12988/imf.2019.9729 | null | math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A strong version of Andrica's conjecture can be formulated as follows: Except
for $p_n\in\{3,7,13,23,31,113\}$, that is $n\in\{2,4,6,9,11,30\}$, one
has$\sqrt{p_{n+1}}-\sqrt{p_n} < \frac{1}{2}.$ While a proof is far out of reach
I shall show that this strong version of Andrica's conjecture is
unconditionally and explicitly verified for all primes below the location of
the 81$^{st}$ maximal prime gap, certainly for all primes $p <2^{64}\approx
1.844\times 10^{19}$. Furthermore this strong Andrica conjecture is slightly
stronger than Oppermann's conjecture --- which in turn is slightly stronger
than both the strong and standard Legendre conjectures, and the strong and
standard Brocard conjectures. Thus the Oppermann conjecture, and strong and
standard Legendre conjectures, are all unconditionally and explicitly verified
for all primes $p <2^{64}\approx1.844\times 10^{19}$. Similarly, the strong and
standard Brocard conjectures are unconditionally and explicitly verified for
all primes $p <2^{32} \approx 4.294 \times 10^9$.
| [
{
"version": "v1",
"created": "Thu, 6 Dec 2018 19:23:51 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Feb 2019 06:27:05 GMT"
},
{
"version": "v3",
"created": "Thu, 9 May 2019 23:42:29 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Visser",
"Matt",
"",
"Victoria University of Wellington"
]
] |
1812.11576 | Dmitry Logachev | A. Grishkov, D. Logachev | Lattice of the dual of an Anderson $t$-motive in terms of a map to a
flag variety | 35 pages; some new results are included | null | null | null | math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $M$ be an uniformizable Anderson $t$-motive of rank $r$, $L$ its lattice
and $l_*:=\{l_1,\dots, l_r\}$ its basis. We define a map $\delta$ from the set
of these bases to a flag variety (the present text gives the definition of
$\delta$ only for elements of the maximal Schubert cell, and for few other
cases). If $l_*$ belongs to the maximal Schubert cell then $\delta(l_*)$ is
described as a set of matrices parametrized by integer points of a tetrahedron;
they are called the Siegel element of $l_*$. We give explicit formulas for a
Siegel element for $M'$ -- the dual of $M$. As a by-product, we get another
proof of the theorem that the lattice of $M'$ is the dual of the lattice of
$M$, independent of the proof obtained by U. Hartl and A.-K. Juschka.
Generalizations of this result to non-maximal Schubert cells, other tensor
operations (tensor product and Hom) and t-motives having non-trivial
endomorphism rings, are subjects of further research.
| [
{
"version": "v1",
"created": "Sun, 30 Dec 2018 17:54:28 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Jan 2019 21:26:34 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Sep 2022 16:05:09 GMT"
},
{
"version": "v4",
"created": "Sat, 31 Dec 2022 22:38:14 GMT"
},
{
"version": "v5",
"created": "Thu, 5 Dec 2024 02:44:21 GMT"
},
{
"version": "v6",
"created": "Sun, 27 Apr 2025 23:48:58 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Grishkov",
"A.",
""
],
[
"Logachev",
"D.",
""
]
] |
1904.00499 | Matt Visser | Matt Visser (Victoria University of Wellington) | Verifying the Firoozbakht, Nicholson, and Farhadian conjectures up to
the 81st maximal prime gap | V1: 9 pages; V2: now 10 pages. Updated to reflect the fact that an
exhaustive search of prime gaps has (as of September 2018) verified that the
81st maximal prime gap occurs at some p > 2^{64}. Minor changes of
presentation, typos fixed. Two references added | Mathematics Vol 7, No 8, (2019) 691 | 10.3390/math7080691 | null | math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Firoozbakht, Nicholoson, and Farhadian conjectures can be phrased in
terms of increasingly powerful conjectured bounds on the prime gaps $g_n :=
p_{n+1}-p_n$. \[ g_n \leq p_n \left(p_n^{1/n} -1 \right)\qquad\qquad\qquad (n
\geq 1; \; Firoozbakht). \] \[ g_n \leq p_n \left((n\ln n)^{1/n} -1
\right)\qquad\qquad (n>4; \; Nicholson). \] \[ g_n \leq p_n \left( \left(p_n
{\ln n\over\ln p_n}\right)^{1/n} -1 \right)\qquad (n>4; \; Farhadian). \] While
a general proof of any of these conjectures is far out of reach I shall show
that all three of these conjectures are unconditionally and explicitly verified
for all primes below the location of the 81$^{st}$ maximal prime gap, certainly
for all primes $p <2^{64}$. For the Firoozbakht conjecture this is a very minor
improvement on currently known results, for the Nicholson and Farhadian
conjectures this may be more interesting.
| [
{
"version": "v1",
"created": "Sun, 31 Mar 2019 22:33:31 GMT"
},
{
"version": "v2",
"created": "Mon, 8 Apr 2019 06:07:48 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Visser",
"Matt",
"",
"Victoria University of Wellington"
]
] |
1904.07045 | Laurent Decreusefond | L. Coutin (IMT), Laurent Decreusefond (INFRES, LTCI, DIG) | Donsker's theorem in {Wasserstein}-1 distance | null | null | 10.1214/20-ECP308 | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We compute the Wassertein-1 (or Kolmogorov-Rubinstein) distance between a
random walk in $R^d$ and the Brownian motion. The proof is based on a new
estimate of the Lipschitz modulus of the solution of the Stein's equation. As
an application, we can evaluate the rate of convergence towards the local time
at 0 of the Brownian motion.
| [
{
"version": "v1",
"created": "Mon, 15 Apr 2019 13:48:13 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Coutin",
"L.",
"",
"IMT"
],
[
"Decreusefond",
"Laurent",
"",
"INFRES, LTCI, DIG"
]
] |
1905.00077 | Vladimir Manuilov | R. Eskandari, M. Frank, V. M. Manuilov, M. S. Moslehian | Extensions of the Lax-Milgram theorem to Hilbert C*-modules | 13 pages | Positivity 24 (2020), no. 4, 1169--1180 | 10.1007/s11117-019-00726-9 | null | math.OA math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present three versions of the Lax-Milgram theorem in the framework of
Hilbert C*-modules, two for those over W*-algebras and one for those over
C*-algebras of compact operators. It is remarkable that while the Riesz theorem
is not valid for certain Hilbert C*-modules over C*-algebras of compact
operators, our Lax-Milgram theorem turns out to be valid for all of them. We
also give several examples to illustrate our results, in particular, we show
that the main theorem is not true for Hilbert modules over arbitrary
C*-algebras.
| [
{
"version": "v1",
"created": "Tue, 30 Apr 2019 19:51:29 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Eskandari",
"R.",
""
],
[
"Frank",
"M.",
""
],
[
"Manuilov",
"V. M.",
""
],
[
"Moslehian",
"M. S.",
""
]
] |
2007.10761 | Yuriy Golovaty | Yuriy Golovaty | 2D Schr\"{o}dinger operators with singular potentials concentrated near
curves | 21 pages, 3 figures | Applicable Analysis (2020), 101(13), 4512-4532 | 10.1080/00036811.2020.1859496 | null | math.SP math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the Schr\"{o}dinger operators $H_\varepsilon=-\Delta
+W+V_\varepsilon$ in $\mathbb{R}^2$ with the short-range potentials
$V_\varepsilon$ which are localized around a smooth closed curve $\gamma$. The
operators $H_\varepsilon$ can be viewed as an approximation of the heuristic
Hamiltonian $H=-\Delta+W+a\partial_\nu\delta_\gamma+b\delta_\gamma$, where
$\delta_\gamma$ is Dirac's $\delta$-function supported on $\gamma$ and
$\partial_\nu\delta_\gamma$ is its normal derivative on $\gamma$. Assuming that
the operator $-\Delta +W$ has only discrete spectrum, we analyze the asymptotic
behaviour of eigenvalues and eigenfunctions of $H_\varepsilon$. The
transmission conditions on $\gamma$ for the eigenfunctions $u^+=\alpha u^-$,
$\alpha\, \partial_\nu u^+-\partial_\nu u^-=\beta u^-$, which arise in the
limit as $\varepsilon\to 0$, reveal a nontrivial connection between spectral
properties of $H_\varepsilon$ and the geometry of $\gamma$.
| [
{
"version": "v1",
"created": "Tue, 21 Jul 2020 12:50:03 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Golovaty",
"Yuriy",
""
]
] |
2008.04050 | Joseph Wilcox | J. A. Wilcox (1), M. J. Grant (1), L. Malone (1), C. Putzke (1), D.
Kaczorowski (2), T. Wolf (3), F. Hardy (3), C. Meingast (3), J.G. Analytis (4
and 5), J.-H. Chu (4 and 5), I. R. Fisher (4 and 5), A. Carrington (1) ((1)
H. H. Wills Physics Laboratory, University of Bristol, Bristol, United
Kingdom, (2) Institute of Low Temeprature and Structure Research, Polish
Academy of Sciences, Wroclaw, Poland, (3) Institute for Quantum Materials and
Technologies, Karlsruhe Institute of Technology, Karlsruhe, Germany, (4)
Geballe Laboratory for Advanced Materials and Department of Applied Physics,
Stanford University, California, USA, (5) Stanford Institute for Materials
and Energy Sciences, SLAC National Accelerator Laboratory, California, USA) | Observation of the Non-linear Meissner Effect | null | Nature Communications 13, 1201 (2022) | 10.1038/s41467-022-28790-y | null | cond-mat.supr-con | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A long-standing theoretical prediction is that in clean, nodal unconventional
superconductors the magnetic penetration depth $\lambda$, at zero temperature,
varies linearly with magnetic field. This non-linear Meissner effect is an
equally important manifestation of the nodal state as the well studied
linear-in-$T$ dependence of $\lambda$, but has never been convincingly
experimentally observed. Here we present measurements of the nodal
superconductors CeCoIn$_5$ and LaFePO which clearly show this non-linear
Meissner effect. We further show how the effect of a small dc magnetic field on
$\lambda(T)$ can be used to distinguish gap nodes from non-nodal deep gap
minima. Our measurements of KFe$_2$As$_2$ suggest that this material has such a
non-nodal state.
| [
{
"version": "v1",
"created": "Mon, 10 Aug 2020 12:23:40 GMT"
},
{
"version": "v2",
"created": "Thu, 22 Apr 2021 18:35:33 GMT"
},
{
"version": "v3",
"created": "Mon, 14 Jun 2021 14:56:13 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Wilcox",
"J. A.",
"",
"4\n and 5"
],
[
"Grant",
"M. J.",
"",
"4\n and 5"
],
[
"Malone",
"L.",
"",
"4\n and 5"
],
[
"Putzke",
"C.",
"",
"4\n and 5"
],
[
"Kaczorowski",
"D.",
"",
"4\n and 5"
],
[
"Wolf",
"T.",
"",
"4\n and 5"
],
[
"Hardy",
"F.",
"",
"4\n and 5"
],
[
"Meingast",
"C.",
"",
"4\n and 5"
],
[
"Analytis",
"J. G.",
"",
"4\n and 5"
],
[
"Chu",
"J. -H.",
"",
"4 and 5"
],
[
"Fisher",
"I. R.",
"",
"4 and 5"
],
[
"Carrington",
"A.",
""
]
] |
2010.11750 | Fan Yang | Fan Yang, Hongyang R. Zhang, Sen Wu, Christopher R\'e, Weijie J. Su | Precise High-Dimensional Asymptotics for Quantifying Heterogeneous
Transfers | 70 pages, 6 figures. Final version accepted by the Journal of Machine
Learning Research | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The problem of learning one task with samples from another task is central to
transfer learning (TL). In this paper, we examine a fundamental question: When
does combining the data samples from a source task and a target task perform
better than single-task learning with the target task alone? This question is
motivated by an intriguing phenomenon known as negative transfer often observed
in the TL literature. Precise quantification of TL effects -- even within
simple statistical models -- has remained elusive in the statistical learning
literature. A critical challenge is that to compare TL to single-task learning,
we would need to compare the risks between two different estimators in a very
precise way. In particular, the comparative advantage of one estimator over
another would depend on the specific distribution shifts between the two tasks.
This paper applies recent developments in the random matrix theory literature
to tackle this challenge in a high-dimensional linear regression setting with
two tasks. We provide precise high-dimensional asymptotics for the bias and
variance of hard parameter sharing (HPS) estimators in the proportional limit,
when the sample sizes of both tasks increase proportionally with dimension at
fixed ratios. The precise asymptotics are expressed as a function of the sample
sizes of both tasks, the covariate shift between their feature population
covariate matrices, and the model shift. We provide illustrative examples of
our results in a random-effects model to determine positive and negative
transfers. For example, we can identify a phase transition in the
high-dimensional linear regression setting from positive transfer to negative
transfer under a model shift between the source and target tasks. The finding
regarding phase transition can be extended to a multiple-task learning setting
where the feature covariates are shared across all tasks.
| [
{
"version": "v1",
"created": "Thu, 22 Oct 2020 14:14:20 GMT"
},
{
"version": "v2",
"created": "Sun, 22 Aug 2021 16:59:23 GMT"
},
{
"version": "v3",
"created": "Fri, 11 Aug 2023 01:45:43 GMT"
},
{
"version": "v4",
"created": "Mon, 28 Apr 2025 08:57:58 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Yang",
"Fan",
""
],
[
"Zhang",
"Hongyang R.",
""
],
[
"Wu",
"Sen",
""
],
[
"Ré",
"Christopher",
""
],
[
"Su",
"Weijie J.",
""
]
] |
2011.02249 | Kai Zehmisch | Wolfgang Schmaltz, Stefan Suhr, Kai Zehmisch | Non-fillability of overtwisted contact manifolds via polyfolds | Accepted for publication in Arnold Math. J | Arnold Math. J. 11 (2025), 98-193 | 10.56994/ARMJ.011.001.005 | null | math.SG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We prove that any weakly symplectically fillable contact manifold is tight.
Furthermore we verify the strong Weinstein conjecture for contact manifolds
that appear as the concave boundary of a directed symplectic cobordism whose
positive boundary satisfies the weak-filling condition and is overtwisted.
Similar results are obtained in the presence of bordered Legendrian open books
whose binding-complement has vanishing second Stiefel-Whitney class. The
results are obtained via polyfolds.
| [
{
"version": "v1",
"created": "Wed, 4 Nov 2020 12:13:40 GMT"
},
{
"version": "v2",
"created": "Sat, 2 Sep 2023 17:48:54 GMT"
},
{
"version": "v3",
"created": "Wed, 11 Sep 2024 10:18:44 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Schmaltz",
"Wolfgang",
""
],
[
"Suhr",
"Stefan",
""
],
[
"Zehmisch",
"Kai",
""
]
] |
2012.07903 | Jie Wang | Victor Magron and Jie Wang | SONC Optimization and Exact Nonnegativity Certificates via Second-Order
Cone Programming | 29 pages, 7 tables, 6 figures, extended version of the article
published in the proceedings of ISSAC 2020. arXiv admin note: text overlap
with arXiv:1906.06179 | null | null | null | math.OC cs.SC math.AG | http://creativecommons.org/licenses/by/4.0/ | The second-order cone (SOC) is a class of simple convex cones and optimizing
over them can be done more efficiently than with semidefinite programming. It
is interesting both in theory and in practice to investigate which convex cones
admit a representation using SOCs, given that they have a strong expressive
ability. In this paper, we prove constructively that the cone of sums of
nonnegative circuits (SONC) admits a SOC representation. Based on this, we give
a new algorithm for unconstrained polynomial optimization via SOC programming.
We also provide a hybrid numeric-symbolic scheme which combines the numerical
procedure with a rounding-projection algorithm to obtain exact nonnegativity
certificates. Numerical experiments demonstrate the efficiency of our algorithm
for polynomials with fairly large degree and number of variables.
| [
{
"version": "v1",
"created": "Mon, 14 Dec 2020 19:32:45 GMT"
},
{
"version": "v2",
"created": "Thu, 19 Aug 2021 15:17:33 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Aug 2021 08:06:18 GMT"
},
{
"version": "v4",
"created": "Mon, 28 Apr 2025 15:56:37 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Magron",
"Victor",
""
],
[
"Wang",
"Jie",
""
]
] |
2012.08558 | Jose Risomar Sousa Mr. | Jose Risomar Sousa | A Reformulation of the Riemann Hypothesis | Added an actually valid proof, improved the text and fixed some minor
inaccuracies | null | null | null | math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present some novelties on the Riemann zeta function. Using an extended
formula created for the polylogarithm in a previous paper,
$\mathrm{Li}_{k}(e^{z})$, the zeta function's Dirichlet series is analytically
continued from $\Re(k)>1$ to the right half-plane, $\Re(k)>0$, by means of the
Dirichlet eta function. More strikingly, we offer a reformulation of the
Riemann hypothesis through a zeta's cousin, $\varphi(k)$, a pole-free function
defined on the entire complex plane whose non-trivial zeros coincide with those
of the zeta function.
| [
{
"version": "v1",
"created": "Tue, 15 Dec 2020 19:09:27 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Jan 2021 03:38:27 GMT"
},
{
"version": "v3",
"created": "Sat, 10 Jul 2021 04:31:12 GMT"
},
{
"version": "v4",
"created": "Mon, 28 Apr 2025 08:45:04 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sousa",
"Jose Risomar",
""
]
] |
2101.10059 | Yuriy Golovaty | Yuriy Golovaty | Membranes with thin and heavy inclusions: asymptotics of spectra | 24 pages, 4 figures | Asymptotic Analysis. 2021;130(1-2):23-51 | 10.3233/ASY-211743 | null | math.SP math.AP | http://creativecommons.org/licenses/by/4.0/ | We study the asymptotic behaviour of eigenvalues and eigenfunctions of 2D
vibrating systems with mass density perturbed in a vicinity of closed curves.
The threshold case in which resonance frequencies of the membrane and thin
inclusion coincide or closely situated is investigated. The perturbed
eigenvalue problem can be realized as a family of self-adjoint operators acting
on varying Hilbert spaces. However the so-called limit operator which is
ultimately responsible for the asymptotics of eigenvalues and eigenfunctions is
non-self-adjoint and possesses the Jordan chains of length $2$. Apart from the
lack of self-adjointness, the operator has non-compact resolvent. As a
consequence, its spectrum has a complicated structure, for instance, the
spectrum contains a countable set of eigenvalues with infinite multiplicity.
| [
{
"version": "v1",
"created": "Mon, 25 Jan 2021 13:17:41 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Jan 2022 14:13:40 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Golovaty",
"Yuriy",
""
]
] |
2102.02515 | Yuanyuan Chen | Yuanyuan Chen, Boyang Li, Han Yu, Pengcheng Wu, Chunyan Miao | HYDRA: Hypergradient Data Relevance Analysis for Interpreting Deep
Neural Networks | null | Proceedings of the AAAI Conference on Artificial Intelligence 2021 | 10.1609/aaai.v35i8.16871 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | The behaviors of deep neural networks (DNNs) are notoriously resistant to
human interpretations. In this paper, we propose Hypergradient Data Relevance
Analysis, or HYDRA, which interprets the predictions made by DNNs as effects of
their training data. Existing approaches generally estimate data contributions
around the final model parameters and ignore how the training data shape the
optimization trajectory. By unrolling the hypergradient of test loss w.r.t. the
weights of training data, HYDRA assesses the contribution of training data
toward test data points throughout the training trajectory. In order to
accelerate computation, we remove the Hessian from the calculation and prove
that, under moderate conditions, the approximation error is bounded.
Corroborating this theoretical claim, empirical results indicate the error is
indeed small. In addition, we quantitatively demonstrate that HYDRA outperforms
influence functions in accurately estimating data contribution and detecting
noisy data labels. The source code is available at
https://github.com/cyyever/aaai_hydra_8686.
| [
{
"version": "v1",
"created": "Thu, 4 Feb 2021 10:00:13 GMT"
},
{
"version": "v2",
"created": "Mon, 1 Mar 2021 09:48:05 GMT"
},
{
"version": "v3",
"created": "Wed, 24 Mar 2021 12:46:06 GMT"
},
{
"version": "v4",
"created": "Sat, 5 Feb 2022 13:50:34 GMT"
},
{
"version": "v5",
"created": "Fri, 30 Dec 2022 04:14:24 GMT"
},
{
"version": "v6",
"created": "Sun, 20 Apr 2025 13:29:13 GMT"
},
{
"version": "v7",
"created": "Sun, 27 Apr 2025 07:17:36 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Chen",
"Yuanyuan",
""
],
[
"Li",
"Boyang",
""
],
[
"Yu",
"Han",
""
],
[
"Wu",
"Pengcheng",
""
],
[
"Miao",
"Chunyan",
""
]
] |
2103.12444 | Jie Wang | Jie Wang and Victor Magron | Exploiting Sparsity in Complex Polynomial Optimization | 21 pages | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | In this paper, we study the sparsity-adapted complex moment-Hermitian sum of
squares (moment-HSOS) hierarchy for complex polynomial optimization problems,
where the sparsity includes correlative sparsity and term sparsity. We compare
the strengths of the sparsity-adapted complex moment-HSOS hierarchy with the
sparsity-adapted real moment-SOS hierarchy on either randomly generated complex
polynomial optimization problems or the AC optimal power flow problem. The
results of numerical experiments show that the sparsity-adapted complex
moment-HSOS hierarchy provides a trade-off between the computational cost and
the quality of obtained bounds for large-scale complex polynomial optimization
problems.
| [
{
"version": "v1",
"created": "Tue, 23 Mar 2021 10:50:28 GMT"
},
{
"version": "v2",
"created": "Mon, 28 Apr 2025 10:16:38 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Wang",
"Jie",
""
],
[
"Magron",
"Victor",
""
]
] |
2104.02496 | Simon Wiegrebe | P. Schulze, S. Wiegrebe, P. W. Thurner, C. Heumann, M. A{\ss}enmacher | A Bayesian approach to modeling topic-metadata relationships | 13 pages, 1 table, 5 figures | null | null | null | cs.CL cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | The objective of advanced topic modeling is not only to explore latent
topical structures, but also to estimate relationships between the discovered
topics and theoretically relevant metadata. Methods used to estimate such
relationships must take into account that the topical structure is not directly
observed, but instead being estimated itself in an unsupervised fashion,
usually by common topic models. A frequently used procedure to achieve this is
the method of composition, a Monte Carlo sampling technique performing multiple
repeated linear regressions of sampled topic proportions on metadata
covariates. In this paper, we propose two modifications of this approach:
First, we substantially refine the existing implementation of the method of
composition from the R package stm by replacing linear regression with the more
appropriate Beta regression. Second, we provide a fundamental enhancement of
the entire estimation framework by substituting the current blending of
frequentist and Bayesian methods with a fully Bayesian approach. This allows
for a more appropriate quantification of uncertainty. We illustrate our
improved methodology by investigating relationships between Twitter posts by
German parliamentarians and different metadata covariates related to their
electoral districts, using the Structural Topic Model to estimate topic
proportions.
| [
{
"version": "v1",
"created": "Tue, 6 Apr 2021 13:28:04 GMT"
},
{
"version": "v2",
"created": "Mon, 28 Apr 2025 07:49:29 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Schulze",
"P.",
""
],
[
"Wiegrebe",
"S.",
""
],
[
"Thurner",
"P. W.",
""
],
[
"Heumann",
"C.",
""
],
[
"Aßenmacher",
"M.",
""
]
] |
2106.11298 | S. Ishmam Mohtashim | Shah Ishmam Mohtashim, Arnav Das, Turbasu Chatterjee and Farhan Tanvir
Chowdhury | A near-term quantum simulation of the transverse field Ising model hints
at Glassy Dynamics | 12 pages, 5 figures, 1 table, accepted in special issue on "New
Trends in Quantum Computing Technology" in EPJ ST | Eur. Phys. J. Spec. Top. (2025) | 10.1140/epjs/s11734-025-01630-y | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | We demonstrate quantum circuit simulations of the transverse field Ising
model with longitudinal fields, displaying salient features of glassy dynamics.
The energy landscape and spin configurations of toy models are considered,
using the Variational Quantum Eigensolver (VQE) to obtain the ground-state
energies and corresponding eigenstates for a $6 \times 6$ Ising lattice using
36 qubits and a 1-dimensional Ising chain of length 25 using 25 qubits. The
former showed disordered spin configurations for a specific mixture of values
of the two fields. These insights mirror catalytic processes, where disorder
within a catalyst can lead to inefficient reaction mechanisms. Results obtained
from our proof-of-principle implementation make the case for kick-starting more
concentrated efforts in harnessing existing quantum computational tools for
computationally probing complex dynamical behaviour arising in quantum matter.
Our aim is to leverage tools from quantum information processing to bring about
a more nuanced understanding of the dynamics and structure of glassy systems,
ultimately informing the development of novel materials and technology.
| [
{
"version": "v1",
"created": "Mon, 21 Jun 2021 17:56:00 GMT"
},
{
"version": "v2",
"created": "Sun, 9 Apr 2023 21:37:40 GMT"
},
{
"version": "v3",
"created": "Mon, 7 Apr 2025 16:15:37 GMT"
},
{
"version": "v4",
"created": "Mon, 21 Apr 2025 18:18:00 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Mohtashim",
"Shah Ishmam",
""
],
[
"Das",
"Arnav",
""
],
[
"Chatterjee",
"Turbasu",
""
],
[
"Chowdhury",
"Farhan Tanvir",
""
]
] |
2107.07050 | Andr\'es Dar\'io Berm\'udez Manjarres | A. D. Berm\'udez Manjarres | Hamiltonian point of view of quantum perturbation theory | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We explore the relation of Van Vleck-Primas perturbation theory of quantum
mechanics with the Lie-series-based perturbation theory of Hamiltonian systems
in classical mechanics. In contrast to previous works on the relation of
quantum and classical perturbation theories, our approach is not based on the
conceptual similarities between the two methods. Instead, we show that for
quantum systems with a finite-dimensional Hilbert space, the Van Vleck-Primas
procedure can be recast exactly into a classical perturbation problem. As a
non-obvious consequence, this approach gives a new way of calculating the
geometric phase of quantum systems using tools from the theory of classical
canonical transformations
| [
{
"version": "v1",
"created": "Thu, 15 Jul 2021 00:02:55 GMT"
},
{
"version": "v2",
"created": "Fri, 16 Jul 2021 14:39:54 GMT"
},
{
"version": "v3",
"created": "Mon, 28 Apr 2025 02:00:49 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Manjarres",
"A. D. Bermúdez",
""
]
] |
2107.07701 | Brandon T. Shapiro | Maru Sarazola and Brandon T. Shapiro | Additivity and Fiber Sequences for Combinatorial K-Theory | 47 pages + 17 page technical appendix. This paper previously titled
"A Gillet--Waldhausen Theorem for Chain Complexes of Finite Sets" is being
split into two: this one with new examples and applications, and another with
the previous title covering the Gillet--Waldhausen Theorem for finite sets
and more general extensive categories, content which can still be found in v2 | null | null | null | math.KT math.AT math.CT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The (A)CGW categories of Campbell and Zakharevich show how finite sets and
varieties behave like the objects of an exact category for the purpose of
algebraic $K$-theory. These structures admit a well-behaved Q-construction akin
to Quillen's, and satisfy analogues of the D\'evissage and Localization
theorems. In this work, we modify Campbell and Zakharevich's axioms to obtain a
framework called ECGW categories that allows for an $S_\bullet$-construction
akin to Waldhausen's, and show how it produces a K-theory spectrum which
satisfies an analogue of the Additivity Theorem. We also define a notion of
``relative ECGW categories'' which have weak equivalences determined by a
subcategory of acyclic objects satisfying minimal conditions; these satisfy
analogues of the Fibration and Localization Theorems that generalize previous
versions in the literature. We illustrate these results with examples including
exact categories, extensive categories, algebraic varieties, and polytopes up
to scissors congruence.
| [
{
"version": "v1",
"created": "Fri, 16 Jul 2021 04:19:10 GMT"
},
{
"version": "v2",
"created": "Tue, 30 Nov 2021 18:27:15 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Apr 2025 18:23:27 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sarazola",
"Maru",
""
],
[
"Shapiro",
"Brandon T.",
""
]
] |
2108.06181 | Hrishav Bakul Barua | Hrishav Bakul Barua, Theint Haythi Mg, Pradip Pramanick, Chayan Sarkar | Detecting socially interacting groups using f-formation: A survey of
taxonomy, methods, datasets, applications, challenges, and future research
directions | Preprint submitted for Review | null | 10.1145/3682072 | null | cs.AI cs.CV cs.HC cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Robots in our daily surroundings are increasing day by day. Their usability
and acceptability largely depend on their explicit and implicit interaction
capability with fellow human beings. As a result, social behavior is one of the
most sought-after qualities that a robot can possess. However, there is no
specific aspect and/or feature that defines socially acceptable behavior and it
largely depends on the situation, application, and society. In this article, we
investigate one such social behavior for collocated robots. Imagine a group of
people is interacting with each other and we want to join the group. We as
human beings do it in a socially acceptable manner, i.e., within the group, we
do position ourselves in such a way that we can participate in the group
activity without disturbing/obstructing anybody. To possess such a quality,
first, a robot needs to determine the formation of the group and then determine
a position for itself, which we humans do implicitly. The theory of f-formation
can be utilized for this purpose. As the types of formations can be very
diverse, detecting the social groups is not a trivial task. In this article, we
provide a comprehensive survey of the existing work on social interaction and
group detection using f-formation for robotics and other applications. We also
put forward a novel holistic survey framework combining all the possible
concerns and modules relevant to this problem. We define taxonomies based on
methods, camera views, datasets, detection capabilities and scale, evaluation
approaches, and application areas. We discuss certain open challenges and
limitations in current literature along with possible future research
directions based on this framework. In particular, we discuss the existing
methods/techniques and their relative merits and demerits, applications, and
provide a set of unsolved but relevant problems in this domain.
| [
{
"version": "v1",
"created": "Fri, 13 Aug 2021 11:51:17 GMT"
},
{
"version": "v2",
"created": "Wed, 13 Oct 2021 13:28:16 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Barua",
"Hrishav Bakul",
""
],
[
"Mg",
"Theint Haythi",
""
],
[
"Pramanick",
"Pradip",
""
],
[
"Sarkar",
"Chayan",
""
]
] |
2108.12443 | Bruce E. Sagan | Sergi Elizalde (Dartmouth College), Matthew Plante (University of
Connecticut), Tom Roby (University of Connecticut), Bruce Sagan (Michigan
State University) | Rowmotion on fences | 26 pages, 7 figures | Algebraic Combin., 6 (2023), no. 1, 17-36, | null | null | math.CO math.GR | http://creativecommons.org/licenses/by/4.0/ | A fence is a poset with elements F = {x_1, x_2, ..., x_n} and covers x_1 <
x_2 < ... < x_a > x_{a+1} > ... > x_b < x_{b+1} < ... where a, b, ... are
positive integers. We investigate rowmotion on antichains and ideals of F. In
particular, we show that orbits of antichains can be visualized using tilings.
This permits us to prove various homomesy results for the number of elements of
an antichain or ideal in an orbit. Rowmotion on fences also exhibits a new
phenomenon, which we call homometry, where the value of a statistic is constant
on orbits of the same size. Along the way, we prove a general homomesy result
for all self-dual posets. We end with some conjectures and avenues for future
research.
| [
{
"version": "v1",
"created": "Fri, 27 Aug 2021 18:02:37 GMT"
},
{
"version": "v2",
"created": "Sat, 26 Apr 2025 22:46:04 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Elizalde",
"Sergi",
"",
"Dartmouth College"
],
[
"Plante",
"Matthew",
"",
"University of\n Connecticut"
],
[
"Roby",
"Tom",
"",
"University of Connecticut"
],
[
"Sagan",
"Bruce",
"",
"Michigan\n State University"
]
] |
2109.08843 | Yujian Feng | Feng Chen, Fei Wu, Qi Wu, Zhiguo Wan | Memory Regulation and Alignment toward Generalizer RGB-Infrared Person | This paper is withdrawn due to an internal decision among the authors
to discontinue its publication | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The domain shift, coming from unneglectable modality gap and non-overlapped
identity classes between training and test sets, is a major issue of
RGB-Infrared person re-identification. A key to tackle the inherent issue --
domain shift -- is to enforce the data distributions of the two domains to be
similar. However, RGB-IR ReID always demands discriminative features, leading
to over-rely feature sensitivity of seen classes, \textit{e.g.}, via
attention-based feature alignment or metric learning. Therefore, predicting the
unseen query category from predefined training classes may not be accurate and
leads to a sub-optimal adversarial gradient. In this paper, we uncover it in a
more explainable way and propose a novel multi-granularity memory regulation
and alignment module (MG-MRA) to solve this issue. By explicitly incorporating
a latent variable attribute, from fine-grained to coarse semantic granularity,
into intermediate features, our method could alleviate the over-confidence of
the model about discriminative features of seen classes. Moreover, instead of
matching discriminative features by traversing nearest neighbor, sparse
attributes, \textit{i.e.}, global structural pattern, are recollected with
respect to features and assigned to measure pair-wise image similarity in
hashing. Extensive experiments on RegDB \cite{RegDB} and SYSU-MM01 \cite{SYSU}
show the superiority of the proposed method that outperforms existing
state-of-the-art methods. Our code is available in
https://github.com/Chenfeng1271/MGMRA.
| [
{
"version": "v1",
"created": "Sat, 18 Sep 2021 05:55:06 GMT"
},
{
"version": "v2",
"created": "Mon, 28 Apr 2025 04:35:02 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Chen",
"Feng",
""
],
[
"Wu",
"Fei",
""
],
[
"Wu",
"Qi",
""
],
[
"Wan",
"Zhiguo",
""
]
] |
2109.09060 | Chun Tao | Chun Tao, Deboleena Roy, Indranil Chakraborty, Kaushik Roy | On the Noise Stability and Robustness of Adversarially Trained Networks
on NVM Crossbars | 13 pages, 14 figures | IEEE Transactions on Very Large Scale Integration (VLSI) Systems (
Volume: 30, Issue: 10, October 2022) | 10.1109/TVLSI.2022.3193312 | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Applications based on Deep Neural Networks (DNNs) have grown exponentially in
the past decade. To match their increasing computational needs, several
Non-Volatile Memory (NVM) crossbar based accelerators have been proposed.
Recently, researchers have shown that apart from improved energy efficiency and
performance, such approximate hardware also possess intrinsic robustness for
defense against adversarial attacks. Prior works quantified this intrinsic
robustness for vanilla DNNs trained on unperturbed inputs. However, adversarial
training of DNNs is the benchmark technique for robustness, and sole reliance
on intrinsic robustness of the hardware may not be sufficient. In this work, we
explore the design of robust DNNs through the amalgamation of adversarial
training and intrinsic robustness of NVM crossbar-based analog hardware. First,
we study the noise stability of such networks on unperturbed inputs and observe
that internal activations of adversarially trained networks have lower
Signal-to-Noise Ratio (SNR), and are sensitive to noise compared to vanilla
networks. As a result, they suffer on average 2x performance degradation due to
the approximate computations on analog hardware. Noise stability analyses show
the instability of adversarially trained DNNs. On the other hand, for
adversarial images generated using Square Black Box attacks, ResNet-10/20
adversarially trained on CIFAR-10/100 display a robustness gain of 20-30%. For
adversarial images generated using Projected-Gradient-Descent (PGD) White-Box
attacks, adversarially trained DNNs present a 5-10% gain in robust accuracy due
to underlying NVM crossbar when $\epsilon_{attack}$ is greater than
$\epsilon_{train}$. Our results indicate that implementing adversarially
trained networks on analog hardware requires careful calibration between
hardware non-idealities and $\epsilon_{train}$ for optimum robustness and
performance.
| [
{
"version": "v1",
"created": "Sun, 19 Sep 2021 04:59:39 GMT"
},
{
"version": "v2",
"created": "Thu, 18 May 2023 21:42:49 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Tao",
"Chun",
""
],
[
"Roy",
"Deboleena",
""
],
[
"Chakraborty",
"Indranil",
""
],
[
"Roy",
"Kaushik",
""
]
] |
2110.03723 | Lorenzo Bonazzi | Lorenzo Bonazzi | On the cut-set of the Gruenberg-Kegel graph of a finite solvable group | This a substatially revised form of "Finite solvable groups whose
Gruenberg-Kegel graph has a cut set" to appear in the Czechoslovak
Mathematical Journal | null | null | null | math.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $\Gamma(G)$ be the Gruenberg-Kegel graph of a finite group $G$. We prove
that if $G$ is solvable and $\sigma$ is a cut-set for $\Gamma(G)$, then $G$ has
a $\sigma$-series of length $5$ whose factors are controlled. As a consequence,
we prove that if $G$ is a solvable group and $\Gamma(G)$ has a cut-vertex $p$,
then the Fitting length $\ell_F(G)$ of $G$ is bounded and the bound obtained is
the best possible. A cut-set is said \emph{minimal} if it does not contain any
other proper subset that is a cut-set for the graph. For a finite solvable
group $G$, we give a geometrical description of $\Gamma(G)$ when it has a
minimal cut-set of size $2$, for a finite solvable group $G$.
| [
{
"version": "v1",
"created": "Thu, 7 Oct 2021 18:23:26 GMT"
},
{
"version": "v2",
"created": "Fri, 4 Nov 2022 17:53:54 GMT"
},
{
"version": "v3",
"created": "Sun, 27 Apr 2025 19:20:36 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Bonazzi",
"Lorenzo",
""
]
] |
2110.08420 | Kawin Ethayarajh | Kawin Ethayarajh, Yejin Choi, Swabha Swayamdipta | Understanding Dataset Difficulty with $\mathcal{V}$-Usable Information | ICML 2022 (Outstanding Paper) | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Estimating the difficulty of a dataset typically involves comparing
state-of-the-art models to humans; the bigger the performance gap, the harder
the dataset is said to be. However, this comparison provides little
understanding of how difficult each instance in a given distribution is, or
what attributes make the dataset difficult for a given model. To address these
questions, we frame dataset difficulty -- w.r.t. a model $\mathcal{V}$ -- as
the lack of $\mathcal{V}$-$\textit{usable information}$ (Xu et al., 2019),
where a lower value indicates a more difficult dataset for $\mathcal{V}$. We
further introduce $\textit{pointwise $\mathcal{V}$-information}$ (PVI) for
measuring the difficulty of individual instances w.r.t. a given distribution.
While standard evaluation metrics typically only compare different models for
the same dataset, $\mathcal{V}$-$\textit{usable information}$ and PVI also
permit the converse: for a given model $\mathcal{V}$, we can compare different
datasets, as well as different instances/slices of the same dataset.
Furthermore, our framework allows for the interpretability of different input
attributes via transformations of the input, which we use to discover
annotation artefacts in widely-used NLP benchmarks.
| [
{
"version": "v1",
"created": "Sat, 16 Oct 2021 00:21:42 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Jun 2022 00:47:15 GMT"
},
{
"version": "v3",
"created": "Sun, 27 Apr 2025 01:03:12 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Ethayarajh",
"Kawin",
""
],
[
"Choi",
"Yejin",
""
],
[
"Swayamdipta",
"Swabha",
""
]
] |
2110.12915 | Li-Hsin Cheng | Li-Hsin Cheng, Pablo B.J. Bosch, Rutger F.H. Hofman, Timo B.
Brakenhoff, Eline F. Bruggemans, Rob J. van der Geest, Eduard R. Holman | Revealing unforeseen diagnostic image features with deep learning by
detecting cardiovascular diseases from apical four-chamber ultrasounds | null | null | 10.1161/JAHA.121.024168 | null | eess.IV cs.AI cs.CV physics.med-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Background. With the rise of highly portable, wireless, and low-cost
ultrasound devices and automatic ultrasound acquisition techniques, an
automated interpretation method requiring only a limited set of views as input
could make preliminary cardiovascular disease diagnoses more accessible. In
this study, we developed a deep learning (DL) method for automated detection of
impaired left ventricular (LV) function and aortic valve (AV) regurgitation
from apical four-chamber (A4C) ultrasound cineloops and investigated which
anatomical structures or temporal frames provided the most relevant information
for the DL model to enable disease classification.
Methods and Results. A4C ultrasounds were extracted from 3,554
echocardiograms of patients with either impaired LV function (n=928), AV
regurgitation (n=738), or no significant abnormalities (n=1,888). Two
convolutional neural networks (CNNs) were trained separately to classify the
respective disease cases against normal cases. The overall classification
accuracy of the impaired LV function detection model was 86%, and that of the
AV regurgitation detection model was 83%. Feature importance analyses
demonstrated that the LV myocardium and mitral valve were important for
detecting impaired LV function, while the tip of the mitral valve anterior
leaflet, during opening, was considered important for detecting AV
regurgitation.
Conclusion. The proposed method demonstrated the feasibility of a 3D CNN
approach in detection of impaired LV function and AV regurgitation using A4C
ultrasound cineloops. The current research shows that DL methods can exploit
large training data to detect diseases in a different way than conventionally
agreed upon methods, and potentially reveal unforeseen diagnostic image
features.
| [
{
"version": "v1",
"created": "Mon, 25 Oct 2021 12:53:45 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Cheng",
"Li-Hsin",
""
],
[
"Bosch",
"Pablo B. J.",
""
],
[
"Hofman",
"Rutger F. H.",
""
],
[
"Brakenhoff",
"Timo B.",
""
],
[
"Bruggemans",
"Eline F.",
""
],
[
"van der Geest",
"Rob J.",
""
],
[
"Holman",
"Eduard R.",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.