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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2505.10158 | Triveni Magadum | Manya Pandit, Triveni Magadum, Harshit Mittal, Omkar Kushwaha | Digital Natives, Digital Activists: Youth, Social Media and the Rise of
Environmental Sustainability Movements | null | null | null | null | cs.SE cs.CY | http://creativecommons.org/licenses/by/4.0/ | The research examines the challenges revolving around young people's social
movements, activism regarding sustainability, as well as the accompanying
social media aspect, and how social media impacts environmental action. This
study focuses on the environmental craze on social media platforms and its
impact on young activists aged 16-25. With the advancement of social media, new
avenues have opened for participation in sustainability issues, especially for
the marginalized, as information moved through transnational networks at
lightning speed. Along with specific Formative Visual Storytelling methods, the
young leaders of the movement deploy hashtags and other online tools to capture
the attention of their peers and decision makers. Challenges persist with
"clicktivism" fatigue from the internet, and site limitations. This article
contributes to insights on emerging forms of civic activism by explaining how
digital natives adapt technology to reframe green activism. The research
suggests that effective digital environmental movements integrate online and
offline action, make it simple for individuals to get involved, and promote
tolerance to algorithmic modifications and climate care among participants.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:36:24 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Pandit",
"Manya",
""
],
[
"Magadum",
"Triveni",
""
],
[
"Mittal",
"Harshit",
""
],
[
"Kushwaha",
"Omkar",
""
]
] |
2505.10159 | Yuzuru Mitsui | Yuzuru Mitsui, Hiroshi Kori | Waveform proportionality and Taylor's law in coupled Lorenz systems | null | null | null | null | nlin.AO nlin.CD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Taylor's law (TL) describes a power-law relationship between the mean and
variance of a quantity, a phenomenon observed across diverse scientific
disciplines. Despite its ubiquity, the underlying mechanisms responsible for TL
are not yet fully elucidated. In particular, the frequent empirical observation
of TL with an exponent 2 warrants further investigation. In a previous study
[Phys. Rev. Lett. 134, 167202 (2025)], we hypothesized that synchronization
contributes to the emergence of TL with an exponent 2. We validated this
hypothesis using coupled oscillator models derived from a food chain model, the
R\"ossler system, the Brusselator, and the Lorenz system. Our analytical and
numerical results demonstrated that strong coupling leads to a form of
synchronization wherein time series become proportional to each other,
consequently resulting in TL with an exponent 2. Here, we extend our previous
findings for the coupled Lorenz system and provide detailed calculations.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:40:45 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Mitsui",
"Yuzuru",
""
],
[
"Kori",
"Hiroshi",
""
]
] |
2505.10160 | Yannis Montreuil | Yannis Montreuil, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi | One-Stage Top-$k$ Learning-to-Defer: Score-Based Surrogates with
Theoretical Guarantees | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce the first one-stage Top-$k$ Learning-to-Defer framework, which
unifies prediction and deferral by learning a shared score-based model that
selects the $k$ most cost-effective entities-labels or experts-per input. While
existing one-stage L2D methods are limited to deferring to a single expert, our
approach jointly optimizes prediction and deferral across multiple entities
through a single end-to-end objective. We define a cost-sensitive loss and
derive a novel convex surrogate that is independent of the cardinality
parameter $k$, enabling generalization across Top-$k$ regimes without
retraining. Our formulation recovers the Top-1 deferral policy of prior
score-based methods as a special case, and we prove that our surrogate is both
Bayes-consistent and $\mathcal{H}$-consistent under mild assumptions. We
further introduce an adaptive variant, Top-$k(x)$, which dynamically selects
the number of consulted entities per input to balance predictive accuracy and
consultation cost. Experiments on CIFAR-10 and SVHN confirm that our one-stage
Top-$k$ method strictly outperforms Top-1 deferral, while Top-$k(x)$ achieves
superior accuracy-cost trade-offs by tailoring allocations to input complexity.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:41:16 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Montreuil",
"Yannis",
""
],
[
"Carlier",
"Axel",
""
],
[
"Ng",
"Lai Xing",
""
],
[
"Ooi",
"Wei Tsang",
""
]
] |
2505.10161 | Abdallah Slaoui | Hanan Saidi, Abdallah Slaoui, Hanane El Hadfi, and Rachid Ahl Laamara | Multiple phase estimation with photon-added multi-mode coherent states
of GHZ-type | 15 pages, 16 figures | null | null | null | quant-ph math-ph math.MP | http://creativecommons.org/publicdomain/zero/1.0/ | This paper explores multiparameter quantum metrology using
Greenberger-Horne-Zeilinger (GHZ)-type photon-added coherent states (PACS) and
investigates both independent and simultaneous parameter estimation with linear
and non-linear protocols, highlighting the significant potential of quantum
resources to enhance precision in multiparameter scenarios. To provide a
comprehensive analysis, we explicitly derive analytical expressions for the
quantum Cram\'er-Rao bound (QCRB) for each protocol. Additionally, we compare
the two estimation strategies, examining the behavior of their QCRBs and
offering insights into the advantages and limitations of these quantum states
in various contexts. Our results show that simultaneous estimation generally
outperforms independent estimation, particularly in non-linear protocols.
Furthermore, we analyze how the QCRB varies with the coherent state amplitude
$|\alpha|^2$, the number of estimated parameters $d$, and the photon excitation
order $n$ across three protocols. The results indicate that increasing
$|\alpha|^2$ and decreasing $d$ improves estimation precision. For low $n$, the
variation in the QCRB is similar for both symmetric and antisymmetric cases;
however, at higher $n$, the antisymmetric case exhibits slightly better
precision. The dependence on $d$ is comparable for both types of states. We
also compare PACS-based GHZ states with NOON states and entangled coherent
states, demonstrating the relative performance of each. Finally, we conclude
with an analysis of homodyne detection in the context of a linear protocol,
discussing its impact on estimation accuracy.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:42:59 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Saidi",
"Hanan",
""
],
[
"Slaoui",
"Abdallah",
""
],
[
"Hadfi",
"Hanane El",
""
],
[
"Laamara",
"Rachid Ahl",
""
]
] |
2505.10162 | Louis Paletta | Louis Paletta, Anthony Leverrier, Mazyar Mirrahimi and Christophe
Vuillot | High-performance local automaton decoders for defect matching in 1D | null | null | null | null | quant-ph cond-mat.stat-mech | http://creativecommons.org/licenses/by/4.0/ | Local automaton decoders offer a promising path toward real-time quantum
error correction by replacing centralized classical decoding, with inherent
hardware constraints, by a natively parallel and streamlined architecture from
a simple local transition rule. We propose two new types of local decoders for
the quantum repetition code in one dimension. The signal-rule decoders
interpret odd parities between neighboring qubits as defects, attracted to each
other via the exchange of classical point-like excitations, represented by a
few bits of local memory. We prove the existence of a threshold in the
code-capacity model and present numerical evidence of exponential logical error
suppression under a phenomenological noise model, with data and measurement
errors at each error correction cycle. Compared to previously known local
decoders that suffer from sub-optimal threshold and scaling, our construction
significantly narrows the gap with global decoders for practical system sizes
and error rates. Implementation requirements can be further reduced by
eliminating the need for local classical memories, with a new rule defined on
two rows of qubits. This shearing-rule works well at relevant system sizes
making it an appealing short-term solution. When combined with biased-noise
qubits, such as cat qubits, these decoders enable a fully local quantum memory
in one dimension.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:43:08 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Paletta",
"Louis",
""
],
[
"Leverrier",
"Anthony",
""
],
[
"Mirrahimi",
"Mazyar",
""
],
[
"Vuillot",
"Christophe",
""
]
] |
2505.10163 | Jeong-Hyuck Park | Jeong-Hyuck Park | Gravitational Core of Double Field Theory: Lecture Notes | 1+49 pages & 1+4 figures | null | null | null | gr-qc hep-th | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Double Field Theory (DFT) has emerged as a comprehensive framework for
gravity, presenting a testable and robust alternative to General Relativity
(GR), rooted in the $\mathbf{O}(D,D)$ symmetry principle of string theory.
These lecture notes aim to provide an accessible introduction to DFT,
structured in a manner similar to traditional GR courses. Key topics include
doubled-yet-gauged coordinates, Riemannian versus non-Riemannian
parametrisations of fundamental fields, covariant derivatives, curvatures, and
the $\mathbf{O}(D,D)$-symmetric augmentation of the Einstein field equation,
identified as the unified field equation for the closed string massless sector.
By offering a novel perspective, DFT addresses unresolved questions in GR and
enables the exploration of diverse physical phenomena, paving the way for
significant future research.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:43:30 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Park",
"Jeong-Hyuck",
""
]
] |
2505.10164 | Mikhail Piotrovich | N.R. Ikhsanov, M.Yu. Piotrovich and N.G. Beskrovnaya | MAD-Accretion Scenario in High-Mass X-Ray Binaries with Neutron Stars | 14 pages | null | null | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | The wind-fed accretion of gas with an intrinsic magnetic field onto neutron
stars in high-mass X-ray binaries (HMXBs) is discussed. The criteria for the
implementation of quasi-spherical accretion in the system, the formation of a
Keplerian accretion disk and a non-Keplerian magnetically arrested disk
(MAD-accretion scenario) are formulated. It is shown that the period of the
equilibrium rotation of a neutron star in the MAD-accretion scenario, in
addition to the parameters of the binary system, is also determined by the
velocity, temperature, and degree of magnetization of the stellar wind of its
massive companion. Within the hypothesis of the equilibrium rotation, we have
estimated the parameters of the stellar wind for the most fully-studied
quasi-equilibrium accreting pulsars in the HMXBs corresponding to the criteria
for the MAD-accretion scenario implementation. We show that the period of the
equilibrium rotation of a neutron star in the system, all other things being
equal, increases with the growth of the magnetization degree of the stellar
wind component of its massive companion flowing out in the plane of the
system's orbit. Therefore, the periods of accreting pulsars in systems with
similar parameters can differ from each other by several orders of magnitude
and, under favorable conditions, reach values of several thousand and even tens
of thousands of seconds.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:45:07 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ikhsanov",
"N. R.",
""
],
[
"Piotrovich",
"M. Yu.",
""
],
[
"Beskrovnaya",
"N. G.",
""
]
] |
2505.10165 | Yacheng Kang | Yacheng Kang, Jin-Ping Zhu, Yu-Han Yang, Ziming Wang, Eleonora Troja,
Bing Zhang, Lijing Shao, Zhuo Li | Shared Properties of Merger-driven Long-duration Gamma-Ray Bursts | 9 pages, 3 figures, accepted for publication in A&A | null | null | null | astro-ph.HE astro-ph.CO gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The recent detections of bright optical/infrared kilonova signals following
two long-duration gamma-ray bursts (LGRBs), GRB 211211A and GRB 230307A, have
significantly challenged the traditional classification of GRBs. These
merger-driven LGRBs may represent a distinct GRB population. Since traditional
GRB classification methods often struggle to distinguish merger-driven LGRBs
from traditional merger-driven short-duration GRBs resulting from compact
object mergers and collapse-driven LGRBs produced by massive stars, this work
aims to explore the shared properties in terms of hardness, energy, and
duration among currently observed merger-driven LGRB events, thereby
identifying their observed differences from the traditional GRB population. We
collect a sample of merger-driven LGRBs with known redshifts, including
observed information on their main emission (ME) and whole emission (WE)
phases. Treating ME and WE properties as two independent sets of information,
we apply several GRB classification methodologies to explore their potential
shared properties. Using the phenomenologically defined Energy-Hardness (EH)
parameter, we identify a probable universal linear correlation across
merger-driven LGRBs, regardless of whether their ME or WE phases are
considered. We propose that such shared properties of merger-driven LGRBs are
unlikely to arise from the low-redshift selection effect and they become
particularly intriguing when compared with the relatively weak correlations or
lack of correlation observed in traditional merger-driven short-duration GRBs
(with or without extended emissions) and collapse-driven LGRBs. Our newly
proposed correlation highlights the necessity for further investigation into
the observations of merger-driven LGRBs and the physical mechanisms underlying
the empirical correlation.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:50:36 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Kang",
"Yacheng",
""
],
[
"Zhu",
"Jin-Ping",
""
],
[
"Yang",
"Yu-Han",
""
],
[
"Wang",
"Ziming",
""
],
[
"Troja",
"Eleonora",
""
],
[
"Zhang",
"Bing",
""
],
[
"Shao",
"Lijing",
""
],
[
"Li",
"Zhuo",
""
]
] |
2505.10166 | Hang Liu | Hang Liu, Francesco Troisi, Hannes H\"ubener, Simone Latini, Angel
Rubio | Cavity-Mediated Electron-Electron Interactions: Renormalizing Dirac
States in Graphene | 20 pages, 10 figures | null | null | null | cond-mat.mtrl-sci physics.optics quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Embedding materials in optical cavities has emerged as a strategy for tuning
material properties. Accurate simulations of electrons in materials interacting
with quantum photon fluctuations of a cavity are crucial for understanding and
predicting cavity-induced phenomena. In this article, we develop a
non-perturbative quantum electrodynamical approach based on a photon-free
self-consistent Hartree-Fock framework to model the coupling between electrons
and cavity photons in crystalline materials. We apply this theoretical approach
to investigate graphene coupled to the vacuum field fluctuations of cavity
photon modes with different types of polarizations. The cavity photons
introduce nonlocal electron-electron interactions, originating from the quantum
nature of light, that lead to significant renormalization of the Dirac bands.
In contrast to the case of graphene coupled to a classical circularly polarized
light field, where a topological Dirac gap emerges, the nonlocal interactions
induced by a quantum linearly polarized photon mode give rise to the formation
of flat bands and the opening of a topologically trivial Dirac gap. When two
symmetric cavity photon modes are introduced, Dirac cones remain gapless, but a
Fermi velocity renormalization yet indicates the relevant role of nonlocal
interactions. These effects disappear in the classical limit for coherent
photon modes. This new self-consistent theoretical framework paves the way for
the simulation of non-perturbative quantum effects in strongly coupled
light-matter systems, and allows for a more comprehensive discovery of novel
cavity-induced quantum phenomena.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:50:54 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Liu",
"Hang",
""
],
[
"Troisi",
"Francesco",
""
],
[
"Hübener",
"Hannes",
""
],
[
"Latini",
"Simone",
""
],
[
"Rubio",
"Angel",
""
]
] |
2505.10167 | Mostafizur Rahman | Saikat Barua, Mostafizur Rahman, Shehenaz Khaled, Md Jafor Sadek,
Rafiul Islam, Shahnewaz Siddique | QuXAI: Explainers for Hybrid Quantum Machine Learning Models | 16 pages, 6 figures, 7 equations | null | null | null | cs.LG cs.AI quant-ph | http://creativecommons.org/licenses/by/4.0/ | The emergence of hybrid quantum-classical machine learning (HQML) models
opens new horizons of computational intelligence but their fundamental
complexity frequently leads to black box behavior that undermines transparency
and reliability in their application. Although XAI for quantum systems still in
its infancy, a major research gap is evident in robust global and local
explainability approaches that are designed for HQML architectures that employ
quantized feature encoding followed by classical learning. The gap is the focus
of this work, which introduces QuXAI, an framework based upon Q-MEDLEY, an
explainer for explaining feature importance in these hybrid systems. Our model
entails the creation of HQML models incorporating quantum feature maps, the use
of Q-MEDLEY, which combines feature based inferences, preserving the quantum
transformation stage and visualizing the resulting attributions. Our result
shows that Q-MEDLEY delineates influential classical aspects in HQML models, as
well as separates their noise, and competes well against established XAI
techniques in classical validation settings. Ablation studies more
significantly expose the virtues of the composite structure used in Q-MEDLEY.
The implications of this work are critically important, as it provides a route
to improve the interpretability and reliability of HQML models, thus promoting
greater confidence and being able to engage in safer and more responsible use
of quantum-enhanced AI technology.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:51:34 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Barua",
"Saikat",
""
],
[
"Rahman",
"Mostafizur",
""
],
[
"Khaled",
"Shehenaz",
""
],
[
"Sadek",
"Md Jafor",
""
],
[
"Islam",
"Rafiul",
""
],
[
"Siddique",
"Shahnewaz",
""
]
] |
2505.10168 | Magnus Appel | Magnus Appel, Joe Alexandersen | Space-Time Multigrid Methods Suitable for Topology Optimisation of
Transient Heat Conduction | 30 pages, 13 figures | null | null | null | cs.CE | http://creativecommons.org/licenses/by-sa/4.0/ | This paper presents Space-Time MultiGrid (STMG) methods which are suitable
for performing topology optimisation of transient heat conduction problems. The
proposed methods use a pointwise smoother and uniform Cartesian space-time
meshes. For problems with high contrast in the diffusivity, it was found that
it is beneficial to define a coarsening strategy based on the geometric mean of
the minimum and maximum diffusivity. However, other coarsening strategies may
be better for other smoothers. Several methods of discretising the coarse
levels were tested. Of these, it was best to use a method which averages the
thermal resistivities on the finer levels. However, this was likely a
consequence of the fact that only one spatial dimension was considered for the
test problems. A second coarsening strategy was proposed which ensures spatial
resolution on the coarse grids. Mixed results were found for this strategy. The
proposed STMG methods were used as a solver for a one-dimensional topology
optimisation problem. In this context, the adjoint problem was also solved
using the STMG methods. The STMG methods were sufficiently robust for this
application, since they converged during every optimisation cycle. It was found
that the STMG methods also work for the adjoint problem when the prolongation
operator only sends information forwards in time, even although the direction
of time for the adjoint problem is backwards.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:54:11 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Appel",
"Magnus",
""
],
[
"Alexandersen",
"Joe",
""
]
] |
2505.10169 | Matthias K\"ummerer | Matthias K\"ummerer, Harneet Khanuja, Matthias Bethge | Modeling Saliency Dataset Bias | null | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in image-based saliency prediction are approaching gold
standard performance levels on existing benchmarks. Despite this success, we
show that predicting fixations across multiple saliency datasets remains
challenging due to dataset bias. We find a significant performance drop (around
40%) when models trained on one dataset are applied to another. Surprisingly,
increasing dataset diversity does not resolve this inter-dataset gap, with
close to 60% attributed to dataset-specific biases. To address this remaining
generalization gap, we propose a novel architecture extending a mostly
dataset-agnostic encoder-decoder structure with fewer than 20 dataset-specific
parameters that govern interpretable mechanisms such as multi-scale structure,
center bias, and fixation spread. Adapting only these parameters to new data
accounts for more than 75% of the generalization gap, with a large fraction of
the improvement achieved with as few as 50 samples. Our model sets a new
state-of-the-art on all three datasets of the MIT/Tuebingen Saliency Benchmark
(MIT300, CAT2000, and COCO-Freeview), even when purely generalizing from
unrelated datasets, but with a substantial boost when adapting to the
respective training datasets. The model also provides valuable insights into
spatial saliency properties, revealing complex multi-scale effects that combine
both absolute and relative sizes.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:55:47 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Kümmerer",
"Matthias",
""
],
[
"Khanuja",
"Harneet",
""
],
[
"Bethge",
"Matthias",
""
]
] |
2505.10170 | Ranendu Adhikary | Ranendu Adhikary and Mriganka Mandal | Multipartite Hardy paradox unlocks device-independent key sharing | 7 pages, 2 figures, comments are welcome | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | We introduce a device-independent quantum key distribution protocol for N
parties, using the multipartite Hardy paradox to certify genuine multipartite
nonlocality. Unlike traditional multipartite protocols that extract the key
from measurement outcomes, our approach generates the shared secret key
directly from the parties' choices of measurement settings. This settings-based
method, certified by the maximal violation of the multipartite Hardy paradox,
achieves a positive key rate and offers a fresh perspective on secure key
distribution. Notably, the Hardy paradox enables any two parties to create a
secret key with a rate much higher than the N-party key, due to more robust
pairwise correlations. This unique capability, inherent to the multipartite
Hardy paradox, allows for tailored key distribution within the group, enhancing
flexibility. Our work establishes a new paradigm for device-independent
conference key agreement, where keys are generated directly from measurement
settings using non-maximally entangled states. This approach ensures robust
security in untrusted quantum networks and enables pairwise key rates that
surpass the N-party rate, offering unprecedented flexibility in key
distribution. By challenging conventional methods, it paves the way for
scalable, noise-resilient multiparty quantum communication systems.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 10:58:52 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Adhikary",
"Ranendu",
""
],
[
"Mandal",
"Mriganka",
""
]
] |
2505.10171 | Abhinav Naga | Abhinav Naga, Xitong Zhang, Junyu Yang, Halim Kusumaatmaja | Modeling droplet-particle interactions on solid surfaces by coupling the
lattice Boltzmann and discrete element methods | 24 pages, 11 figures | null | null | null | physics.flu-dyn cond-mat.soft | http://creativecommons.org/licenses/by/4.0/ | We introduce a numerical method for investigating interfacial flows coupled
with frictional solid particles. Our method combines the lattice Boltzmann
method (LBM) to model the dynamics of a two-component fluid and the discrete
element method (DEM) to model contact forces (normal reaction, sliding
friction, rolling friction) between solid particles and between solid particles
and flat solid surfaces. To couple the fluid and particle dynamics, we (1) use
the momentum exchange method to transfer hydrodynamic forces between the fluids
and particles, (2) account for different particle wettability using a geometric
boundary condition, and (3) explicitly account for capillary forces between
particles and liquid-fluid interfaces using a 3D capillary force model. We
benchmark the contact forces by investigating the dynamics of a particle
bouncing off a solid surface and rolling down an inclined plane. To benchmark
the hydrodynamic and capillary forces, we investigate the Segr\`e-Silberberg
effect and measure the force required to detach a particle from a liquid-fluid
interface, respectively. Motivated by the self-cleaning properties of the lotus
leaf, we apply our method to investigate how drops remove contaminant particles
from surfaces and quantify the forces acting on particles during removal. Our
method makes it possible to investigate the influence of various parameters
that are often difficult to tune independently in experiments, including
contact angles, surface tension, viscosity, and coefficient of friction between
the surface and particles. Our results highlight that friction plays a crucial
role when drops remove particles from surfaces.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:00:10 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Naga",
"Abhinav",
""
],
[
"Zhang",
"Xitong",
""
],
[
"Yang",
"Junyu",
""
],
[
"Kusumaatmaja",
"Halim",
""
]
] |
2505.10172 | Zeyan Li | Zeyan Li, Libing Chen, Yin Tang | Does Scaling Law Apply in Time Series Forecasting? | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Rapid expansion of model size has emerged as a key challenge in time series
forecasting. From early Transformer with tens of megabytes to recent
architectures like TimesNet with thousands of megabytes, performance gains have
often come at the cost of exponentially increasing parameter counts. But is
this scaling truly necessary? To question the applicability of the scaling law
in time series forecasting, we propose Alinear, an ultra-lightweight
forecasting model that achieves competitive performance using only k-level
parameters. We introduce a horizon-aware adaptive decomposition mechanism that
dynamically rebalances component emphasis across different forecast lengths,
alongside a progressive frequency attenuation strategy that achieves stable
prediction in various forecasting horizons without incurring the computational
overhead of attention mechanisms. Extensive experiments on seven benchmark
datasets demonstrate that Alinear consistently outperforms large-scale models
while using less than 1% of their parameters, maintaining strong accuracy
across both short and ultra-long forecasting horizons. Moreover, to more fairly
evaluate model efficiency, we propose a new parameter-aware evaluation metric
that highlights the superiority of ALinear under constrained model budgets. Our
analysis reveals that the relative importance of trend and seasonal components
varies depending on data characteristics rather than following a fixed pattern,
validating the necessity of our adaptive design. This work challenges the
prevailing belief that larger models are inherently better and suggests a
paradigm shift toward more efficient time series modeling.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:04:39 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Li",
"Zeyan",
""
],
[
"Chen",
"Libing",
""
],
[
"Tang",
"Yin",
""
]
] |
2505.10173 | Vineet Rawat | V. Rawat (1 and 2), M. R. Samal (1), A. Zavagno (3 and 4), Sami Dib
(5), Davide Elia (6), J. Jose (7), D.K. Ojha (8), K. Srivastav1 (1 and 2)
((1) Physical Research Laboratory, Ahmedabad, India, (2) Indian Institute of
Technology Gandhinagar, India, (3) Aix Marseille Univ, CNRS, CNES, LAM
Marseille, France, (4) Institut Universitaire de France, France, (5)
Max-Planck-Institut f\"ur Astronomie, Heidelberg, Germany, (6) Istituto di
Astrofisica e Planetologia Spaziali, INAF, Roma, Italy, (7) Indian Institute
of Science Education and Research (IISER) Tirupati, India, (8) Department of
Astronomy and Astrophysics, Tata Institute of Fundamental Research, Mumbai,
India) | Star formation efficiency and scaling relations in parsec-scale
cluster-forming clumps | 19 pages, 14 figures | null | null | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Numerical simulations predict that clumps ($\sim$1 pc) should form stars at
high efficiency to produce bound star clusters. We conducted a statistical
study of 17 nearby cluster-forming clumps to examine the star formation rate
and gas mass surface density relations (i.e. $\Sigma_{\rm{SFR}}$ vs.
$\Sigma_{\rm{gas}}$) at the clump scale. Using near-infrared point sources and
Herschel dust continuum analysis, we obtained the radius, age, and stellar mass
for most clusters in the ranges 0.5$-$1.6 pc, 0.5$-$1.5 Myr, 40$-$500
M$_\odot$, respectively, and also found that they are associated with
$\Sigma_{\rm{gas}}$ values ranging from 80$-$600 M$_\odot$ pc$^{-2}$. We
obtained the best-fit scaling relations as $\Sigma_{\rm{SFR}}$ $\propto$
$\Sigma_{\rm{gas}}^{1.46}$ and $\Sigma_{\rm{SFR}}$ $\propto$
$(\Sigma_{\rm{gas}}/t_{\rm{ff}})^{0.80}$ for the studied sample of clumps.
Comparing our results with existing scaling relations at cloud and
extragalactic scales, we found that while the power-law exponent obtained in
this work is similar to those found at these scales, the star formation rate
surface densities are relatively higher for similar gas mass surface densities.
From this work, we obtained instantaneous median star formation efficiency
(SFE) and efficiency per free-fall time ($\epsilon_{\rm{ff}}$) of $\sim$20% and
$\sim$13%, respectively, for the studied clumps. We discuss the cause of the
obtained high SFE and $\epsilon_{\rm{ff}}$ in the studied clumps and also
discuss the results in the context of bound cluster formation within molecular
clouds. We conclude that our results do not favour a universal scaling law with
a constant value of $\epsilon_{\rm{ff}}$ in star-forming systems across
different scales.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:05:04 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Rawat",
"V.",
"",
"1 and 2"
],
[
"Samal",
"M. R.",
"",
"3 and 4"
],
[
"Zavagno",
"A.",
"",
"3 and 4"
],
[
"Dib",
"Sami",
"",
"1 and 2"
],
[
"Elia",
"Davide",
"",
"1 and 2"
],
[
"Jose",
"J.",
"",
"1 and 2"
],
[
"Ojha",
"D. K.",
"",
"1 and 2"
],
[
"Srivastav1",
"K.",
"",
"1 and 2"
]
] |
2505.10174 | Jingbo Zhao | Jingbo Zhao and Zhaoming Lu and J. Andrew Zhang and Jiaxi Zhou and
Weicai Li and Tao Gu | Subspace-Based Super-Resolution Sensing for Bi-Static ISAC with Clock
Asynchronism | 13 pages, 9 figures. This work has been submitted to the IEEE for
possible publication | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Bi-static sensing is an attractive configuration for integrated sensing and
communications (ISAC) systems; however, clock asynchronism between widely
separated transmitters and receivers introduces time-varying time offsets (TO)
and phase offsets (PO), posing significant challenges. This paper introduces a
signal-subspace-based framework that estimates decoupled angles, delays, and
complex gain sequences (CGS)-- the target-reflected signals -- for multiple
dynamic target paths. The proposed framework begins with a novel TO alignment
algorithm, leveraging signal subspace or covariance, to mitigate TO variations
across temporal snapshots, enabling coherent delay-domain analysis.
Subsequently, subspace-based methods are developed to compensate for TO
residuals and to perform joint angle-delay estimation. Finally, leveraging the
high resolution in the joint angle-delay domain, the framework compensates for
the PO and estimates the CGS for each target. The framework can be applied to
both single-antenna and multi-antenna systems. Extensive simulations and
experiments using commercial Wi-Fi devices demonstrate that the proposed
framework significantly surpasses existing solutions in parameter estimation
accuracy and delay resolution. Notably, it uniquely achieves a super-resolution
in the delay domain, with a probability-of-resolution curve tightly approaching
that in synchronized systems.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:07:47 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Zhao",
"Jingbo",
""
],
[
"Lu",
"Zhaoming",
""
],
[
"Zhang",
"J. Andrew",
""
],
[
"Zhou",
"Jiaxi",
""
],
[
"Li",
"Weicai",
""
],
[
"Gu",
"Tao",
""
]
] |
2505.10175 | Francesco Mattesini | Francesco Mattesini, Felix Otto | From Combinatorics to Partial Differential Equations | Comments very welcome! | null | null | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The optimal matching of point clouds in $\mathbb{R}^d$ is a combinatorial
problem; applications in statistics motivate to consider random point clouds,
like the Poisson point process. There is a crucial dependance on dimension $d$,
with $d=2$ being the critical dimension. This is revealed by adopting an
analytical perspective, connecting e.\,g.~to Optimal Transportation. These
short notes provide an introduction to the subject. The material presented here
is based on a series of lectures held at the International Max Planck Research
School during the summer semester 2022. Recordings of the lectures are
available at
https://www.mis.mpg.de/events/event/imprs-ringvorlesung-summer-semester-2022.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:08:19 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Mattesini",
"Francesco",
""
],
[
"Otto",
"Felix",
""
]
] |
2505.10176 | Xiang He | Xiang He, Dongcheng Zhao, Yang Li, Qingqun Kong, Xin Yang, Yi Zeng | Incorporating brain-inspired mechanisms for multimodal learning in
artificial intelligence | The manuscript is under review and the code is available at
https://github.com/Brain-Cog-Lab/IEMF | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimodal learning enhances the perceptual capabilities of cognitive systems
by integrating information from different sensory modalities. However, existing
multimodal fusion research typically assumes static integration, not fully
incorporating key dynamic mechanisms found in the brain. Specifically, the
brain exhibits an inverse effectiveness phenomenon, wherein weaker unimodal
cues yield stronger multisensory integration benefits; conversely, when
individual modal cues are stronger, the effect of fusion is diminished. This
mechanism enables biological systems to achieve robust cognition even with
scarce or noisy perceptual cues. Inspired by this biological mechanism, we
explore the relationship between multimodal output and information from
individual modalities, proposing an inverse effectiveness driven multimodal
fusion (IEMF) strategy. By incorporating this strategy into neural networks, we
achieve more efficient integration with improved model performance and
computational efficiency, demonstrating up to 50% reduction in computational
cost across diverse fusion methods. We conduct experiments on audio-visual
classification, continual learning, and question answering tasks to validate
our method. Results consistently demonstrate that our method performs
excellently in these tasks. To verify universality and generalization, we also
conduct experiments on Artificial Neural Networks (ANN) and Spiking Neural
Networks (SNN), with results showing good adaptability to both network types.
Our research emphasizes the potential of incorporating biologically inspired
mechanisms into multimodal networks and provides promising directions for the
future development of multimodal artificial intelligence. The code is available
at https://github.com/Brain-Cog-Lab/IEMF.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:08:50 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"He",
"Xiang",
""
],
[
"Zhao",
"Dongcheng",
""
],
[
"Li",
"Yang",
""
],
[
"Kong",
"Qingqun",
""
],
[
"Yang",
"Xin",
""
],
[
"Zeng",
"Yi",
""
]
] |
2505.10177 | Fabrice Baudoin Dr | Fabrice Baudoin, Li Chen, Meng Yang | Korevaar-Schoen and heat kernel characterizations of Sobolev and BV
spaces on local trees | null | null | null | null | math.AP math.FA math.MG math.PR | http://creativecommons.org/licenses/by/4.0/ | We study Sobolev and BV spaces on local trees which are metric spaces locally
isometric to real trees. Such spaces are equipped with a Radon measure
satisfying a locally uniform volume growth condition. Using the intrinsic
geodesic structure, we define weak gradients and develop from it a coherent
theory of Sobolev and BV spaces. We provide two main characterizations: one via
Korevaar-Schoen-type energy functionals and another via the heat kernel
associated with the natural Dirichlet form. Applications include interpolation
results for Besov-Lipschitz spaces, critical exponents computations, and a Nash
inequality. In globally tree-like settings we also establish $L^p$ gradient
bounds for the heat semigroup.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:08:58 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Baudoin",
"Fabrice",
""
],
[
"Chen",
"Li",
""
],
[
"Yang",
"Meng",
""
]
] |
2505.10178 | Thomas Mejstrik | Thomas Mejstrik | The finiteness conjecture for $3 \times 3$ binary matrices | null | Dolomites Research Notes on Approximation, 15(5), 24-38 (2022) | 10.14658/PUPJ-DRNA-2022-5-3 | null | math.NA cs.NA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The invariant polytope algorithm was a breakthrough in the joint spectral
radius computation, allowing to find the exact value of the joint spectral
radius for most matrix families~\cite{GP2013,GP2016}. This algorithm found many
applications in problems of functional analysis, approximation theory,
combinatorics, etc. In this paper we propose a modification of the invariant
polytope algorithm enlarging the class of problems to which it is applicable.
Precisely, we introduce mixed numeric and symbolic computations. A further
minor modification of augmenting the input set with additional matrices speeds
up the algorithm in certain cases. With this modifications we are able to
automatically prove the finiteness conjecture for all pairs of binary $3\times
3$ matrices and sign $2\times 2$ matrices.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:22:40 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Mejstrik",
"Thomas",
""
]
] |
2505.10179 | Chongjun Ouyang | Chongjun Ouyang, Zhaolin Wang, Yuanwei Liu, and Zhiguo Ding | Rate Region of ISAC for Pinching-Antenna Systems | submit to possible IEEE journal | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Pinching-Antenna SyStem (PASS) reconstructs wireless channels through
\emph{pinching beamforming}, wherein the activated positions of pinching
antennas along dielectric waveguides are optimized to shape the radiation
pattern. The aim of this article is to analyze the performance limits of
employing PASS in integrated sensing and communications (ISAC). Specifically, a
PASS-assisted ISAC system is considered, where a pinched waveguide is utilized
to simultaneously communicate with a user and sense a target. Closed-form
expressions for the achievable communication rate (CR) and sensing rate (SR)
are derived to characterize the information-theoretic limits of this
dual-functional operation. \romannumeral1) For the single-pinch case,
closed-form solutions for the optimal pinching antenna location are derived
under \emph{sensing-centric (S-C)}, \emph{communications-centric (C-C)}, and
\emph{Pareto-optimal} designs. On this basis, the CR-SR trade-off is
characterized by deriving the full CR-SR rate region, which is shown to
encompass that of conventional fixed-antenna systems. \romannumeral2) For the
multiple-pinch case, an antenna location refinement method is applied to obtain
the optimal C-C and S-C pinching beamformers. As a further advance, inner and
outer bounds on the achievable CR-SR region are derived using an element-wise
alternating optimization technique and by invoking Cauchy-Schwarz and
Karamata's inequalities, respectively. Numerical results demonstrate that:
\romannumeral1) the derived bounds closely approximate the true CR-SR region;
and \romannumeral2) PASS can achieve a significantly larger rate region than
conventional-antenna systems.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:27:10 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ouyang",
"Chongjun",
""
],
[
"Wang",
"Zhaolin",
""
],
[
"Liu",
"Yuanwei",
""
],
[
"Ding",
"Zhiguo",
""
]
] |
2505.10180 | Wu Zhaoxin | Xiangrui Hou, Zhaoxin Wu, Fangyu Wang, Shiyao Zhu, Bo Yan, Zhaoju Yang | Quantum boomerang effect of light | null | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The quantum boomerang effect is a counterintuitive phenomenon where a wave
packet, despite having an initial momentum, returns to its starting position in
a disordered medium. However, up to now, the experimental exploration of this
effect remains largely unexplored. Here, we report the experimental observation
of the quantum boomerang effect of light. Our experiment is based on a
one-dimensional disordered photonic lattice, which is composed of on-chip
optical waveguides with engineered on-site random potential. We first
characterize this optical disordered system by demonstrating the static
Anderson localization of light beams. Next, through launching a kinetic light
beam into the system, we observe that the light beam first moves away from its
starting point, arrives at a maximum value, reverses its direction, and returns
to its original position over time, confirming the observation of the quantum
boomerang effect of light. Surprisingly, we find that optical loss, usually
considered to be detrimental to optical experiments, can enhance the quantum
boomerang effect by accelerating the light back to its original position. Our
work provides new insights into the light-matter interactions in disordered
medium and opens an avenue for future study of this phenomenon in nonlinear and
many-photon contexts.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:28:02 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Hou",
"Xiangrui",
""
],
[
"Wu",
"Zhaoxin",
""
],
[
"Wang",
"Fangyu",
""
],
[
"Zhu",
"Shiyao",
""
],
[
"Yan",
"Bo",
""
],
[
"Yang",
"Zhaoju",
""
]
] |
2505.10181 | Ciriaco Goddi | Ciriaco Goddi (1 and 2), Douglas F. Carlos (1) (for the EHT
Collaboration, (1) Universidade de S\~ao Paulo, Instituto de Astronomia,
Geof\'isica e Ci\^encias Atmosf\'ericas, Departamento de Astronomia, S\~ao
Paulo, Brazil, (2) Dipartimento di Fisica, Universit\'a degli Studi di
Cagliari, SP Monserrato-Sestu, Monserrato, Italy) | First polarization study of the M87 jet and active galactic nuclei at
submillimeter wavelengths with ALMA | 17 pages, accepted for publication in A&A | null | null | null | astro-ph.GA astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | We present full-polarization observations at $\lambda = 0.87$ mm (345 GHz)
conducted with the Atacama Large Millimeter/submillimeter Array (ALMA) toward
Messier 87 (M87) and seven other radio-loud active galactic nuclei (AGN). We
determine polarization and Faraday properties of the targets, with linear
polarization (LP) fractions spanning 1\% to 17\%, and Faraday rotation measures
(RMs) exceeding $10^{5}$ rad m$^{-2}$. These RM values, 1-2 orders of magnitude
higher than typically observed at $\lambda > 3$ mm, suggest denser Faraday
screens or stronger magnetic fields in the (sub)millimeter emission regions.
Additionally, we present the first submillimeter polarized images of the M87
jet and the observed AGN. In the M87 jet, we identify RM gradients and sign
reversals, suggestive of a helical magnetic field structure extending over
kiloparsec scales. These observations were part of the ALMA Band~7 Very Long
Baseline Interferometry (VLBI) commissioning during the April 2021 Event
Horizon Telescope (EHT) campaign. The results provide essential constraints for
calibrating, analyzing, and interpreting VLBI data from the EHT at 345 GHz,
representing a critical step toward submillimeter VLBI imaging.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:28:50 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Goddi",
"Ciriaco",
"",
"1 and 2"
],
[
"Carlos",
"Douglas F.",
""
]
] |
2505.10182 | Yoichi Ishibashi | Yoichi Ishibashi, Taro Yano, Masafumi Oyamada | Mining Hidden Thoughts from Texts: Evaluating Continual Pretraining with
Synthetic Data for LLM Reasoning | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) have demonstrated significant improvements in
reasoning capabilities through supervised fine-tuning and reinforcement
learning. However, when training reasoning models, these approaches are
primarily applicable to specific domains such as mathematics and programming,
which imposes fundamental constraints on the breadth and scalability of
training data. In contrast, continual pretraining (CPT) offers the advantage of
not requiring task-specific signals. Nevertheless, how to effectively
synthesize training data for reasoning and how such data affect a wide range of
domains remain largely unexplored. This study provides a detailed evaluation of
Reasoning CPT, a form of CPT that uses synthetic data to reconstruct the hidden
thought processes underlying texts, based on the premise that texts are the
result of the author's thinking process. Specifically, we apply Reasoning CPT
to Gemma2-9B using synthetic data with hidden thoughts derived from STEM and
Law corpora, and compare it to standard CPT on the MMLU benchmark. Our analysis
reveals that Reasoning CPT consistently improves performance across all
evaluated domains. Notably, reasoning skills acquired in one domain transfer
effectively to others; the performance gap with conventional methods widens as
problem difficulty increases, with gains of up to 8 points on the most
challenging problems. Furthermore, models trained with hidden thoughts learn to
adjust the depth of their reasoning according to problem difficulty.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:29:01 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ishibashi",
"Yoichi",
""
],
[
"Yano",
"Taro",
""
],
[
"Oyamada",
"Masafumi",
""
]
] |
2505.10183 | Wanyu Wang | Jieke Lin, Wanyu Wang, Longxiang Yin, Yinhe Han | KAITIAN: A Unified Communication Framework for Enabling Efficient
Collaboration Across Heterogeneous Accelerators in Embodied AI Systems | 9 pages, 4 figures. Jieke Lin and Wanyu Wang contributed equally to
this work | null | null | null | cs.DC cs.AI | http://creativecommons.org/licenses/by/4.0/ | Embodied Artificial Intelligence (AI) systems, such as autonomous robots and
intelligent vehicles, are increasingly reliant on diverse heterogeneous
accelerators (e.g., GPGPUs, NPUs, FPGAs) to meet stringent real-time processing
and energy-efficiency demands. However, the proliferation of vendor-specific
proprietary communication libraries creates significant interoperability
barriers, hindering seamless collaboration between different accelerator types
and leading to suboptimal resource utilization and performance bottlenecks in
distributed AI workloads. This paper introduces KAITIAN, a novel distributed
communication framework designed to bridge this gap. KAITIAN provides a unified
abstraction layer that intelligently integrates vendor-optimized communication
libraries for intra-group efficiency with general-purpose communication
protocols for inter-group interoperability. Crucially, it incorporates a
load-adaptive scheduling mechanism that dynamically balances computational
tasks across heterogeneous devices based on their real-time performance
characteristics. Implemented as an extension to PyTorch and rigorously
evaluated on a testbed featuring NVIDIA GPUs and Cambricon MLUs, KAITIAN
demonstrates significant improvements in resource utilization and scalability
for distributed training tasks. Experimental results show that KAITIAN can
accelerate training time by up to 42% compared to baseline homogeneous systems,
while incurring minimal communication overhead (2.8--4.3%) and maintaining
model accuracy. KAITIAN paves the way for more flexible and powerful
heterogeneous computing in complex embodied AI applications.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:29:43 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Lin",
"Jieke",
""
],
[
"Wang",
"Wanyu",
""
],
[
"Yin",
"Longxiang",
""
],
[
"Han",
"Yinhe",
""
]
] |
2505.10184 | Axel Lemoine | Axel Lemoine | The Tangent Space Attack | 26 pages without references and appendix; 31 in total | null | null | null | cs.IT cs.CR math.IT | http://creativecommons.org/licenses/by/4.0/ | We propose a new method for retrieving the algebraic structure of a generic
alternant code given an arbitrary generator matrix, provided certain conditions
are met. We then discuss how this challenges the security of the McEliece
cryptosystem instantiated with this family of codes. The central object of our
work is the quadratic hull related to a linear code, defined as the
intersection of all quadrics passing through the columns of a given generator
or parity-check matrix, where the columns are considered as points in the
affine or projective space. The geometric properties of this object reveal
important information about the internal algebraic structure of the code. This
is particularly evident in the case of generalized Reed-Solomon codes, whose
quadratic hull is deeply linked to a well-known algebraic variety called the
rational normal curve. By utilizing the concept of Weil restriction of affine
varieties, we demonstrate that the quadratic hull of a generic dual alternant
code inherits many interesting features from the rational normal curve, on
account of the fact that alternant codes are subfield-subcodes of generalized
Reed-Solomon codes. If the rate of the generic alternant code is sufficiently
high, this allows us to construct a polynomial-time algorithm for retrieving
the underlying generalized Reed-Solomon code from which the alternant code is
defined, which leads to an efficient key-recovery attack against the McEliece
cryptosystem when instantiated with this class of codes. Finally, we discuss
the generalization of this approach to Algebraic-Geometry codes and Goppa
codes.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:30:46 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Lemoine",
"Axel",
""
]
] |
2505.10185 | Seongyun Lee | Seongyun Lee, Seungone Kim, Minju Seo, Yongrae Jo, Dongyoung Go,
Hyeonbin Hwang, Jinho Park, Xiang Yue, Sean Welleck, Graham Neubig, Moontae
Lee, Minjoon Seo | The CoT Encyclopedia: Analyzing, Predicting, and Controlling how a
Reasoning Model will Think | Work in progress | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Long chain-of-thought (CoT) is an essential ingredient in effective usage of
modern large language models, but our understanding of the reasoning strategies
underlying these capabilities remains limited. While some prior works have
attempted to categorize CoTs using predefined strategy types, such approaches
are constrained by human intuition and fail to capture the full diversity of
model behaviors. In this work, we introduce the CoT Encyclopedia, a bottom-up
framework for analyzing and steering model reasoning. Our method automatically
extracts diverse reasoning criteria from model-generated CoTs, embeds them into
a semantic space, clusters them into representative categories, and derives
contrastive rubrics to interpret reasoning behavior. Human evaluations show
that this framework produces more interpretable and comprehensive analyses than
existing methods. Moreover, we demonstrate that this understanding enables
performance gains: we can predict which strategy a model is likely to use and
guide it toward more effective alternatives. Finally, we provide practical
insights, such as that training data format (e.g., free-form vs.
multiple-choice) has a far greater impact on reasoning behavior than data
domain, underscoring the importance of format-aware model design.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:31:02 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Lee",
"Seongyun",
""
],
[
"Kim",
"Seungone",
""
],
[
"Seo",
"Minju",
""
],
[
"Jo",
"Yongrae",
""
],
[
"Go",
"Dongyoung",
""
],
[
"Hwang",
"Hyeonbin",
""
],
[
"Park",
"Jinho",
""
],
[
"Yue",
"Xiang",
""
],
[
"Welleck",
"Sean",
""
],
[
"Neubig",
"Graham",
""
],
[
"Lee",
"Moontae",
""
],
[
"Seo",
"Minjoon",
""
]
] |
2505.10186 | Julian Siber | Mishel Carelli, Bernd Finkbeiner, Julian Siber | Closure and Complexity of Temporal Causality | Fortieth Annual ACM/IEEE Symposium on Logic in Computer Science (LICS
2025) | null | null | null | cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Temporal causality defines what property causes some observed temporal
behavior (the effect) in a given computation, based on a counterfactual
analysis of similar computations. In this paper, we study its closure
properties and the complexity of computing causes. For the former, we establish
that safety, reachability, and recurrence properties are all closed under
causal inference: If the effect is from one of these property classes, then the
cause for this effect is from the same class. We also show that persistence and
obligation properties are not closed in this way. These results rest on a
topological characterization of causes which makes them applicable to a wide
range of similarity relations between computations. Finally, our complexity
analysis establishes improved upper bounds for computing causes for safety,
reachability, and recurrence properties. We also present the first lower bounds
for all of the classes.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:37:15 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Carelli",
"Mishel",
""
],
[
"Finkbeiner",
"Bernd",
""
],
[
"Siber",
"Julian",
""
]
] |
2505.10187 | Takahiro Suzuki | Takahiro Suzuki, Michele Aleandri, Stefano Moretti | Mill's canons meet social ranking: A characterization of plurality | null | null | null | null | econ.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In his book entitled ''A System of Logic, Ratiocinative and Inductive''
(1843), John Stuart Mill proposed principles of inductive reasoning in the form
of five canons. To date, these canons are classic methods for causal reasoning:
they are intended to single out the circumstances that are connected to the
phenomenon under focus. The present paper reinterprets Mill's canons in the
modern theory of social ranking solutions, which aims to estimate the power of
individuals based on teams' performances. We first apply Mill's canons to
determine the key success factors in cooperative performances and then
characterize plurality using a strong version of Mill's first canon. Plurality
is also compatible with most other canons. Thus, our results demonstrated a
hidden link between classical causal reasoning and the theory of social ranking
solutions.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:40:08 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Suzuki",
"Takahiro",
""
],
[
"Aleandri",
"Michele",
""
],
[
"Moretti",
"Stefano",
""
]
] |
2505.10188 | Felix Liedeker | Felix Liedeker, Olivia Sanchez-Graillet, Moana Seidler, Christian
Brandt, J\"org Wellmer, Philipp Cimiano | A User Study Evaluating Argumentative Explanations in Diagnostic
Decision Support | Presented at 'The First Workshop on Natural Language Argument-Based
Explanations', co-located with ECAI 2024 | null | null | null | cs.AI | http://creativecommons.org/licenses/by-sa/4.0/ | As the field of healthcare increasingly adopts artificial intelligence, it
becomes important to understand which types of explanations increase
transparency and empower users to develop confidence and trust in the
predictions made by machine learning (ML) systems. In shared decision-making
scenarios where doctors cooperate with ML systems to reach an appropriate
decision, establishing mutual trust is crucial. In this paper, we explore
different approaches to generating explanations in eXplainable AI (XAI) and
make their underlying arguments explicit so that they can be evaluated by
medical experts. In particular, we present the findings of a user study
conducted with physicians to investigate their perceptions of various types of
AI-generated explanations in the context of diagnostic decision support. The
study aims to identify the most effective and useful explanations that enhance
the diagnostic process. In the study, medical doctors filled out a survey to
assess different types of explanations. Further, an interview was carried out
post-survey to gain qualitative insights on the requirements of explanations
incorporated in diagnostic decision support. Overall, the insights gained from
this study contribute to understanding the types of explanations that are most
effective.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:42:24 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Liedeker",
"Felix",
""
],
[
"Sanchez-Graillet",
"Olivia",
""
],
[
"Seidler",
"Moana",
""
],
[
"Brandt",
"Christian",
""
],
[
"Wellmer",
"Jörg",
""
],
[
"Cimiano",
"Philipp",
""
]
] |
2505.10189 | Wlodek Bednarek | W. Bednarek, J. Sitarek, M. Ulatowski | Dips in the sub-TeV gamma-ray light curves from central parts of
galaxies due to transiting luminous stars | 8 pages, 5 figures, accepted by Astronomy & Astrophysics, For the
purpose of Open Access, the author has applied a CC-BY public copyright
licence to any Author Accepted Manuscript (AAM) version arising from this
submission | null | null | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | The GeV-TeV {\gamma}-ray emission is observed from the direction of the
source Sgr A^*, which is identified with the SMBH in the centre of our Galaxy.
According to some models this {\gamma}-ray emission might originate in the very
compact, central region identified with the direct surrounding of the SMBH. Sgr
A^* is surrounded by a massive nuclear star cluster. Occasionally these stars
might pass close to the line of sight of the observer, resulting in partial
absorption of the sub-TeV {\gamma}-ray emission. We investigate the conditions
at which an absorption feature appears in the {\gamma}-ray light curves from
the Galactic Centre or nuclei of other galaxies containing SMBHs. The detection
of such features would allow to obtain constraints on the emission site of
{\gamma} rays in active galaxies. We calculate the optical depths for {\gamma}
rays in the radiation of individual massive stars, or from the whole population
of stars for different parameters of the star cluster. We show that the
observer with a line of sight close to the orbital plane of the star can
register a {\gamma}-ray absorption dip lasting from a fraction of a day up to a
few tens of days. The combined effect of the bulk absorption on the whole
population of stars instead can produce a flickering of the observed emission
of a red-noise type in the power spectrum of the emission. Predicted absorption
features in the sub-TeV {\gamma}-ray light curves from galaxies with SMBHs
should be easily detectable by the Large-Sized Telescopes of the Cherenkov
Telescope Array Observatory. The discovery of such absorption features provides
a unique indication that the {\gamma}-ray production is occurring in a compact
region, close to the horizon of the SMBH.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:42:32 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Bednarek",
"W.",
""
],
[
"Sitarek",
"J.",
""
],
[
"Ulatowski",
"M.",
""
]
] |
2505.10190 | Stefan Ivkovic | Otmane Benchiheb, Stefan Ivkovic, Noureddine Karim, Marko Kostic | Luh hypercyclic vector for composition operator | null | null | null | null | math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we deal with the construction of holomorphic functions on a
simply connected domain satisfying that all its derivatives and antiderivatives
under a composition operator have a dense orbit. Such functions will be called
Luh hypercyclic vectors for the respective composition operator. We show that
there is a dense linear manifold of Luh hypercyclic vectors. Moreover, we study
the dynamics of cosine operator function generated by weighted composition
operators on solid Banach function spaces, in particular on Orlicz and Morrey
spaces, and we give sufficient conditions for supercyclicity of such cosine
operator functions in terms of the corresponding weight function. Also, we give
concrete examples of weighted translations satisfying these sufficient
conditions.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:47:18 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Benchiheb",
"Otmane",
""
],
[
"Ivkovic",
"Stefan",
""
],
[
"Karim",
"Noureddine",
""
],
[
"Kostic",
"Marko",
""
]
] |
2505.10191 | Qingyu Zheng | Qingyu Zheng, Qi Shao, Guijun Han, Wei Li, Hong Li and Xuan Wang | LanTu: Dynamics-Enhanced Deep Learning for Eddy-Resolving Ocean
Forecasting | 22 pages, 6 figures | null | null | null | physics.ao-ph cs.AI cs.LG nlin.CD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mesoscale eddies dominate the spatiotemporal multiscale variability of the
ocean, and their impact on the energy cascade of the global ocean cannot be
ignored. Eddy-resolving ocean forecasting is providing more reliable protection
for fisheries and navigational safety, but also presents significant scientific
challenges and high computational costs for traditional numerical models.
Artificial intelligence (AI)-based weather and ocean forecasting systems are
becoming powerful tools that balance forecast performance with computational
efficiency. However, the complex multiscale features in the ocean dynamical
system make AI models still face many challenges in mesoscale eddy forecasting
(especially regional modelling). Here, we develop LanTu, a regional
eddy-resolving ocean forecasting system based on dynamics-enhanced deep
learning. We incorporate cross-scale interactions into LanTu and construct
multiscale physical constraint for optimising LanTu guided by knowledge of eddy
dynamics in order to improve the forecasting skill of LanTu for mesoscale
evolution. The results show that LanTu outperforms the existing advanced
operational numerical ocean forecasting system (NOFS) and AI-based ocean
forecasting system (AI-OFS) in temperature, salinity, sea level anomaly and
current prediction, with a lead time of more than 10 days. Our study highlights
that dynamics-enhanced deep learning (LanTu) can be a powerful paradigm for
eddy-resolving ocean forecasting.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:47:54 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Zheng",
"Qingyu",
""
],
[
"Shao",
"Qi",
""
],
[
"Han",
"Guijun",
""
],
[
"Li",
"Wei",
""
],
[
"Li",
"Hong",
""
],
[
"Wang",
"Xuan",
""
]
] |
2505.10192 | Prashant Shinde | Prashant P. Shinde, Priyadarshini P. Pai, Shashishekar P. Adiga, K.
Subramanya Mayya, Yongbeom Seo, Myungsoo Hwang, Heeyoung Go, and Changmin
Park | Defect Detection in Photolithographic Patterns Using Deep Learning
Models Trained on Synthetic Data | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the photolithographic process vital to semiconductor manufacturing,
various types of defects appear during EUV pattering. Due to ever-shrinking
pattern size, these defects are extremely small and cause false or missed
detection during inspection. Specifically, the lack of defect-annotated quality
data with good representation of smaller defects has prohibited deployment of
deep learning based defect detection models in fabrication lines. To resolve
the problem of data unavailability, we artificially generate scanning electron
microscopy (SEM) images of line patterns with known distribution of defects and
autonomously annotate them. We then employ state-of-the-art object detection
models to investigate defect detection performance as a function of defect
size, much smaller than the pitch width. We find that the real-time object
detector YOLOv8 has the best mean average precision of 96% as compared to
EfficientNet, 83%, and SSD, 77%, with the ability to detect smaller defects. We
report the smallest defect size that can be detected reliably. When tested on
real SEM data, the YOLOv8 model correctly detected 84.6% of Bridge defects and
78.3% of Break defects across all relevant instances. These promising results
suggest that synthetic data can be used as an alternative to real-world data in
order to develop robust machine-learning models.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:50:02 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Shinde",
"Prashant P.",
""
],
[
"Pai",
"Priyadarshini P.",
""
],
[
"Adiga",
"Shashishekar P.",
""
],
[
"Mayya",
"K. Subramanya",
""
],
[
"Seo",
"Yongbeom",
""
],
[
"Hwang",
"Myungsoo",
""
],
[
"Go",
"Heeyoung",
""
],
[
"Park",
"Changmin",
""
]
] |
2505.10193 | Thomas Weber | Antonio Del Donno, Emanuele Latini, Thomas Weber | Gauge transformations on quantum principal bundles | 28 pages. Submitted to PoS, as a contribution to the Proceedings of
Corfu Summer Institute 2024 "Workshop on Noncommutative and Generalized
Geometry in String theory, Gauge theory and Related Physical Models", 17-24
September 2024 | null | null | null | math.QA math-ph math.MP math.RA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We understand quantum principal bundle as faithfully flat Hopf--Galois
extensions, with a structure Hopf algebra coacting on a total space algebra and
with base algebra given by the coinvariant elements. To endow such bundles with
a compatible differential structure, one requires the coaction to extend as a
morphism of differential graded algebras. This leads to an exact noncommutative
Atiyah sequence, a graded Hopf--Galois extension of differential forms and a
canonical braiding on total space forms such that the latter are graded-braided
commutative. We recall this approach to noncommutative differential geometry
and further discuss the extension of quantum gauge transformations, in the
sense of Brzezi\'nski, to differential forms. In this way we obtain an action
of quantum gauge transformations on connections of the quantum principal bundle
and their curvature. Explicit examples, such as the noncommutative 2-torus, the
quantum Hopf fibration and smash product algebras are discussed.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:50:05 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Del Donno",
"Antonio",
""
],
[
"Latini",
"Emanuele",
""
],
[
"Weber",
"Thomas",
""
]
] |
2505.10194 | Raffael Poeggel | Raffael Poeggel (1), Maximilian Stahlke (1), Jonas Pirkl (1), Jonathan
Ott (1), George Yammine (1), Tobias Feigl (1 and 2), Christopher Mutschler
((1) Fraunhofer Institute for Integrated Circuits (IIS), (2)
Friedrich-Alexander-Universitaet Erlangen-Nuernberg (FAU)) | Passive Channel Charting: Locating Passive Targets using a UWB Mesh | null | null | null | null | eess.SP | http://creativecommons.org/licenses/by/4.0/ | Fingerprint-based passive localization enables high localization accuracy
using low-cost UWB IoT radio sensors. However, fingerprinting demands extensive
effort for data acquisition. The concept of channel charting reduces this
effort by modeling and projecting the manifold of \ac{csi} onto a 2D coordinate
space. So far, researchers only applied this concept to active radio
localization, where a mobile device intentionally and actively emits a specific
signal. In this paper, we apply channel charting to passive localization. We
use a pedestrian dead reckoning (PDR) system to estimate a target's velocity
and derive a distance matrix from it. We then use this matrix to learn a
distance-preserving embedding in 2D space, which serves as a fingerprinting
model. In our experiments, we deploy six nodes in a fully connected
ultra-wideband (UWB) mesh network to show that our method achieves high
localization accuracy, with an average error of just 0.24\,m, even when we
train and test on different targets.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:50:41 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Poeggel",
"Raffael",
"",
"1 and 2"
],
[
"Stahlke",
"Maximilian",
"",
"1 and 2"
],
[
"Pirkl",
"Jonas",
"",
"1 and 2"
],
[
"Ott",
"Jonathan",
"",
"1 and 2"
],
[
"Yammine",
"George",
"",
"1 and 2"
],
[
"Feigl",
"Tobias",
"",
"1 and 2"
],
[
"Mutschler",
"Christopher",
""
]
] |
2505.10195 | Sreenivasa Chary Thatikonda | Sreenivasa chary Thatikonda, F. N. De Oliveira-Lopes, A. Mustonen, K.
Pommois, D. Told, F.Jenko | Verification of a hybrid gyrokinetic model using the advanced
semi-Lagrange code ssV | null | null | null | null | physics.plasm-ph | http://creativecommons.org/licenses/by/4.0/ | The super simple Vlasov (ssV) code was developed to study instabilities,
turbulence, and reconnection in weakly magnetized plasmas, such as the solar
wind in the dissipation range and the edge of fusion plasmas. The ssV code
overcomes the limitations of standard gyrokinetic theory by using a hybrid
model that incorporates fully kinetic ions and gyrokinetic electrons. This
hybrid gyrokinetic model enables accurate modeling in regimes characterized by
steep gradients and high-frequency dynamics. To achieve this, ssV implements a
set of semi-Lagrangian numerical schemes, including Positive Flux Conservative
(PFC), Flux Conservative fifth-order (FCV), FCV with Umeda limiters, and a
Semi-Lagrangian Monotonicity-Preserving fifth-order scheme (SLMP5). Benchmark
problems such as Landau damping, ion-acoustic waves, ion Bernstein waves, and
kinetic Alfven waves were employed to evaluate the schemes. The SLMP5 scheme
consistently delivered the best overall accuracy and numerical stability
performance. The code also addresses well-known electromagnetic gyrokinetic
simulation issues, such as the Ampere cancellation problem, using carefully
chosen velocity-space resolutions and accurate integral evaluation.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:51:40 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Thatikonda",
"Sreenivasa chary",
""
],
[
"De Oliveira-Lopes",
"F. N.",
""
],
[
"Mustonen",
"A.",
""
],
[
"Pommois",
"K.",
""
],
[
"Told",
"D.",
""
],
[
"Jenko",
"F.",
""
]
] |
2505.10196 | Hui Hu | Hui Hu, Zhao Liu, and Xia-Ji Liu | Unconventional superconductivity of an altermagnetic metal: Polarized
BCS and inhomogeneous Fulde-Ferrell-Larkin-Ovchinnikov states | 10 pages, 8 figures | null | null | null | cond-mat.supr-con cond-mat.quant-gas cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the superconductivity of two-dimensional spin-1/2 Fermi
systems with $d$-wave altermagnetism under external magnetic field near zero
temperature. At large altermagnetic coupling without magnetic field, we show
that altermagnetism drives a second-order phase transition from the standard
Bardeen-Cooper-Schrieffer (BCS) state to an inhomogeneous
Fulde-Ferrell-Larkin-Ovchinnikov (FFLO) state. The inclusion of magnetic field
turns the BCS state into a long-sought polarized BCS superconductor with
spin-population imbalance. It also shrinks the parameter window of the FFLO
state and eventually leads to a nontrivial quantum tri-critical Lifshitz point,
where two second-order phase transition lines between the polarized BCS, FFLO
and normal states intersect. At small altermagnetic coupling, we find the usual
route to the FFLO state driven by magnetic field. The presence of the
altermagnetic coupling narrows the phase window of the FFLO state and creates
another quantum Lifshitz point, where a first-order transition curve meets a
second-order transition line. Between the two Lifshitz points, the transition
from the polarized BCS state to the normal state is smooth. Our predicted rich
phase diagram is relevant to some recently discovered unconventional magnets,
including RuO$_{2}$ that exhibits a relatively high superconducting temperature
in the thin film limit under applied strain. Our results of unconventional
superfluidity are also testable in ultracold atom laboratories, where a
spin-1/2 altermagnetic Fermi gas might be realizable upon loading into
two-dimensional Hubbard lattices.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:53:13 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Hu",
"Hui",
""
],
[
"Liu",
"Zhao",
""
],
[
"Liu",
"Xia-Ji",
""
]
] |
2505.10197 | Anjali De Silva | Anjali de Silva, Gang Chen, Hui Ma, Seyed Mohammad Nekooei and
Xingquan Zuo | Advancing Community Detection with Graph Convolutional Neural Networks:
Bridging Topological and Attributive Cohesion | This paper has been accepted by IJCAI (International Joint Conference
on Artificial Intelligence) 2025 | null | null | null | cs.SI cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Community detection, a vital technology for real-world applications, uncovers
cohesive node groups (communities) by leveraging both topological and attribute
similarities in social networks. However, existing Graph Convolutional Networks
(GCNs) trained to maximize modularity often converge to suboptimal solutions.
Additionally, directly using human-labeled communities for training can
undermine topological cohesiveness by grouping disconnected nodes based solely
on node attributes. We address these issues by proposing a novel Topological
and Attributive Similarity-based Community detection (TAS-Com) method. TAS-Com
introduces a novel loss function that exploits the highly effective and
scalable Leiden algorithm to detect community structures with global optimal
modularity. Leiden is further utilized to refine human-labeled communities to
ensure connectivity within each community, enabling TAS-Com to detect community
structures with desirable trade-offs between modularity and compliance with
human labels. Experimental results on multiple benchmark networks confirm that
TAS-Com can significantly outperform several state-of-the-art algorithms.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:53:33 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"de Silva",
"Anjali",
""
],
[
"Chen",
"Gang",
""
],
[
"Ma",
"Hui",
""
],
[
"Nekooei",
"Seyed Mohammad",
""
],
[
"Zuo",
"Xingquan",
""
]
] |
2505.10198 | Luciano Martinez-Rau | Mariano Ferrero, Jos\'e Omar Chelotti, Luciano Sebasti\'an
Martinez-Rau, Leandro Vignolo, Mart\'in Pires, Julio Ricardo Galli, Leonardo
Luis Giovanini, Hugo Leonardo Rufiner | A multi-head deep fusion model for recognition of cattle foraging events
using sound and movement signals | Preprint submitted to Engineering Applications of Artificial
Intelligence | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Monitoring feeding behaviour is a relevant task for efficient herd management
and the effective use of available resources in grazing cattle. The ability to
automatically recognise animals' feeding activities through the identification
of specific jaw movements allows for the improvement of diet formulation, as
well as early detection of metabolic problems and symptoms of animal
discomfort, among other benefits. The use of sensors to obtain signals for such
monitoring has become popular in the last two decades. The most frequently
employed sensors include accelerometers, microphones, and cameras, each with
its own set of advantages and drawbacks. An unexplored aspect is the
simultaneous use of multiple sensors with the aim of combining signals in order
to enhance the precision of the estimations. In this direction, this work
introduces a deep neural network based on the fusion of acoustic and inertial
signals, composed of convolutional, recurrent, and dense layers. The main
advantage of this model is the combination of signals through the automatic
extraction of features independently from each of them. The model has emerged
from an exploration and comparison of different neural network architectures
proposed in this work, which carry out information fusion at different levels.
Feature-level fusion has outperformed data and decision-level fusion by at
least a 0.14 based on the F1-score metric. Moreover, a comparison with
state-of-the-art machine learning methods is presented, including traditional
and deep learning approaches. The proposed model yielded an F1-score value of
0.802, representing a 14% increase compared to previous methods. Finally,
results from an ablation study and post-training quantization evaluation are
also reported.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:55:16 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ferrero",
"Mariano",
""
],
[
"Chelotti",
"José Omar",
""
],
[
"Martinez-Rau",
"Luciano Sebastián",
""
],
[
"Vignolo",
"Leandro",
""
],
[
"Pires",
"Martín",
""
],
[
"Galli",
"Julio Ricardo",
""
],
[
"Giovanini",
"Leonardo Luis",
""
],
[
"Rufiner",
"Hugo Leonardo",
""
]
] |
2505.10199 | Yehu Izhak Maimon | Yehu I. Maimon, Alex B. Nielsen and Ofek Birnholtz | Constraining Dirty Black Holes and pseudo-complex General Relativity
with the Gravitational Waves Transient Catalog 3.0 | null | null | null | null | gr-qc | http://creativecommons.org/licenses/by/4.0/ | We use data from the Gravitational Wave Transient Catalog 3.0 to update
constraints on parameterized deviations from General Relativity, as encountered
in pseudo-complex general relativity (pcGR) theory and models of dirty black
holes. The pcGR framework extends Einstein's theory of general relativity by
introducing additional parameters that diverge from standard predictions in the
strong-field regime, potentially excluding black hole horizons for specific
parameter choices. We analyze gravitational wave signals from coalescing
compact objects to obtain new bounds on these parameters. Our results modify
existing constraints and identify previously unexplored regions of parameter
space, exploring the observational viability of dirty black holes and
horizonless solutions in pcGR. We confirm the exclusion of 1PN deviations
sufficient to avoid a horizon, and for the first time rule out 1.5PN as well.
We also discuss implications for current and future gravitational wave
observations in refining these constraints
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:55:28 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Maimon",
"Yehu I.",
""
],
[
"Nielsen",
"Alex B.",
""
],
[
"Birnholtz",
"Ofek",
""
]
] |
2505.10200 | Henrik Van Impel | Henrik van Impel, David Steuer, Volker Schulz-von der Gathen, Marc
B\"oke, Judith Golda | Analyzing atomic oxygen product evolution in Micro Cavity Plasma Arrays
by a combination of a Multi-PMT OES Setup and a 0-D Chemical Model | null | null | null | null | physics.plasm-ph | http://creativecommons.org/licenses/by/4.0/ | Dielectric barrier discharges (DBDs) are widely used in applications such as
ozone generation and volatile organic compound treatment, where performance can
be enhanced through catalyst integration. A fundamental understanding of
reactive species generation is essential for advancing these technologies.
However, temporally resolving reactive species production, especially during
the initial discharges, remains a challenge, despite its importance for
controlling production rates and energy efficiency. This study examines atomic
oxygen production as a model system for reactive species production in a
micro-cavity plasma array, a custom surface DBD confined to micrometer-sized
cavities. Optical emission spectroscopy was employed to investigate
plasma-chemical processes in helium with 0.1-0.25$\%$ molecular oxygen
admixture at atmospheric pressure. The discharge, powered by a 15$\,$kHz,
600$\,$V amplitude triangular voltage, achieved near-complete oxygen
dissociation (up to 100$\%$), as determined via helium state-enhanced
actinometry (SEA). A novel multi-photomultiplier system enabled precise
temporal tracking of atomic oxygen density and dissociation dynamics. To ensure
measurement accuracy, a basic 0D chemical model was developed, reinforcing the
reliability of the experimental results.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:56:08 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"van Impel",
"Henrik",
""
],
[
"Steuer",
"David",
""
],
[
"der Gathen",
"Volker Schulz-von",
""
],
[
"Böke",
"Marc",
""
],
[
"Golda",
"Judith",
""
]
] |
2505.10201 | Johannes Schmidt | Victor Lagerkvist, Mohamed Maizia, Johannes Schmidt | A Fine-Grained Complexity View on Propositional Abduction -- Algorithms
and Lower Bounds | null | null | null | null | cs.CC cs.AI | http://creativecommons.org/licenses/by/4.0/ | The Boolean satisfiability problem (SAT) is a well-known example of monotonic
reasoning, of intense practical interest due to fast solvers, complemented by
rigorous fine-grained complexity results. However, for non-monotonic reasoning,
e.g., abductive reasoning, comparably little is known outside classic
complexity theory. In this paper we take a first step of bridging the gap
between monotonic and non-monotonic reasoning by analyzing the complexity of
intractable abduction problems under the seemingly overlooked but natural
parameter n: the number of variables in the knowledge base. We obtain several
positive results for $\Sigma^P_2$- as well as NP- and coNP-complete fragments,
which implies the first example of beating exhaustive search for a
$\Sigma^P_2$-complete problem (to the best of our knowledge). We complement
this with lower bounds and for many fragments rule out improvements under the
(strong) exponential-time hypothesis.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:56:19 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Lagerkvist",
"Victor",
""
],
[
"Maizia",
"Mohamed",
""
],
[
"Schmidt",
"Johannes",
""
]
] |
2505.10202 | Hongyi Huang | Jintian Shao, Hongyi Huang, Jiayi Wu, YiMing Cheng, ZhiYu Wu, You
Shan, MingKai Zheng | VQ-Logits: Compressing the Output Bottleneck of Large Language Models
via Vector Quantized Logits | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have achieved remarkable success but face
significant computational and memory challenges, particularly due to their
extensive output vocabularies. The final linear projection layer, mapping
hidden states to vocabulary-sized logits, often constitutes a substantial
portion of the model's parameters and computational cost during inference.
Existing methods like adaptive softmax or hierarchical softmax introduce
structural complexities. In this paper, we propose VQ-Logits, a novel approach
that leverages Vector Quantization (VQ) to drastically reduce the parameter
count and computational load of the LLM output layer. VQ-Logits replaces the
large V * dmodel output embedding matrix with a small, shared codebook of K
embedding vectors (K << V ). Each token in the vocabulary is mapped to one of
these K codebook vectors. The LLM predicts logits over this compact codebook,
which are then efficiently "scattered" to the full vocabulary space using the
learned or preassigned mapping. We demonstrate through extensive experiments on
standard language modeling benchmarks (e.g., WikiText-103, C4) that VQ-Logits
can achieve up to 99% parameter reduction in the output layer and 6x speedup in
logit computation, with only a marginal 4% increase in perplexity compared to
full softmax baselines. We further provide detailed ablation studies on
codebook size, initialization, and learning strategies, showcasing the
robustness and effectiveness of our approach.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 11:58:04 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Shao",
"Jintian",
""
],
[
"Huang",
"Hongyi",
""
],
[
"Wu",
"Jiayi",
""
],
[
"Cheng",
"YiMing",
""
],
[
"Wu",
"ZhiYu",
""
],
[
"Shan",
"You",
""
],
[
"Zheng",
"MingKai",
""
]
] |
2505.10203 | Deokkeun An | Deokkeun An, Young Sun Lee, Yutaka Hirai, Timothy C. Beers | Deciphering the Milky Way's Star Formation at Cosmic Noon with High
Proper-Motion Stars: A Precursor to the Merger-Driven Starburst | 24 pages, 14 figures, submitted to A&A | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Evidence suggests that the Milky Way (MW) underwent a major collision with
the Gaia-Sausage/Enceladus (GSE) dwarf galaxy around cosmic noon. While GSE has
since been fully disrupted, it brought in ex-situ stars and dynamically heated
in-situ stars into the halo. In addition, the gas-rich merger may have
triggered a burst of in-situ star formation, potentially giving rise to a
chemically distinct stellar component. We investigate the region of phase space
where stars formed during the GSE merger likely reside, and retain distinct
chemical and dynamical signatures. Building on our previous investigation of
metallicity ([Fe/H]) and vertical angular momentum ($L_Z$) distributions, we
analyse spectroscopic samples from GALAH, APOGEE, SDSS, and LAMOST, combined
with Gaia kinematics. We focus on high proper-motion stars as effective tracers
of the phase-space volume likely influenced by the GSE merger. To correct for
selection effects, we incorporate metallicity estimates derived from SDSS and
SMSS photometry. Our analysis reveals that low-$\alpha$ stars with GSE-like
kinematics exhibit bimodality in [Na/Fe] and [Al/Fe] at $-1.0 \lesssim {\rm
[Fe/H]} \lesssim -0.4$. One group follows the low light-element abundances of
GSE stars, while another exhibits enhanced values. These low-$\alpha$, high-Na
stars have eccentric orbits, but are more confined to the inner MW. A subset of
these stars overlaps with the Eos population, representing only the
high-eccentricity portion of a broader structure. After correcting for sampling
biases, we find that the population comprises $\sim7\%$ relative to the GSE
debris. These results suggest that the low-$\alpha$, high-Na stars formed in a
compact region, likely fueled by gas from the GSE progenitor, analogous to
clumpy star-forming clouds seen in high-redshift galaxies. Such stars may trace
the first sparks of more extensive merger-driven starburst activity.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:02:00 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"An",
"Deokkeun",
""
],
[
"Lee",
"Young Sun",
""
],
[
"Hirai",
"Yutaka",
""
],
[
"Beers",
"Timothy C.",
""
]
] |
2505.10204 | Sascha Chur | Sascha Chur, Robin Minke, Youfan He, M\'at\'e Vass, Thomas
Mussenbrock, Ralf Peter Brinkmann, Efe Kemaneci, Lars Sch\"ucke, Volker
Schulz-von der Gathen, Andrew R. Gibson, Marc B\"oke, Judith Golda | Dynamics of reactive oxygen species produced by the COST microplasma jet | null | null | null | null | physics.plasm-ph cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study is focused on measuring the densities of the excited molecular
oxygen species, O$_{2}(\text{a}^{1}\Delta_{\text{g}})$ and
O$_{2}(\text{b}^{1}\Sigma_{\text{g}}^{+})$, produced in a COST atmospheric
pressure plasma jet using a helium-oxygen mixture. Knowledge of the ozone
density is critical for measurements because of its high quenching rate of
these species. Additionally O$_{2}(\text{a}^{1}\Delta_{\text{g}})$ is difficult
to measure, due to its low emission intensity and sensitivity to background
interference in the plasma region. Therefore a flow cell was used to enhance
signal detection in the effluent region. To validate the measurements and
improve understanding of reaction mechanisms, results were compared with two
simulation models: a pseudo-1D plug flow simulation and a 2D fluid simulation.
The plug flow simulation provided an effective means for estimating species
densities, with a fast computation time. The 2D simulation offered a more
realistic description of the flow dynamics, which proved critical to correctly
describe the experimental trends. However, it requires long computation times
to reach an equilibrium state in the flow cell. Otherwise, it leads to
discrepancies to the experimental data. Further discrepancies arose, from an
overestimation of the ozone density from the models, as validated from the
O$_{2}(\text{b}^{1}\Sigma_{\text{g}}^{+})$ density measurements. Optimizing the
reaction rate coefficients for the effluent region might improve the agreement
with the experimental results. Despite these limitations both simulations
aligned reasonably well with experimental data, showcasing the well validated
plasma chemistry of the models, even for complicated effluent geometries.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:02:07 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Chur",
"Sascha",
""
],
[
"Minke",
"Robin",
""
],
[
"He",
"Youfan",
""
],
[
"Vass",
"Máté",
""
],
[
"Mussenbrock",
"Thomas",
""
],
[
"Brinkmann",
"Ralf Peter",
""
],
[
"Kemaneci",
"Efe",
""
],
[
"Schücke",
"Lars",
""
],
[
"der Gathen",
"Volker Schulz-von",
""
],
[
"Gibson",
"Andrew R.",
""
],
[
"Böke",
"Marc",
""
],
[
"Golda",
"Judith",
""
]
] |
2505.10205 | Umair Haroon | Umair Haroon, Ahmad AlMughrabi, Thanasis Zoumpekas, Ricardo Marques,
and Petia Radeva | VolE: A Point-cloud Framework for Food 3D Reconstruction and Volume
Estimation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Accurate food volume estimation is crucial for medical nutrition management
and health monitoring applications, but current food volume estimation methods
are often limited by mononuclear data, leveraging single-purpose hardware such
as 3D scanners, gathering sensor-oriented information such as depth
information, or relying on camera calibration using a reference object. In this
paper, we present VolE, a novel framework that leverages mobile device-driven
3D reconstruction to estimate food volume. VolE captures images and camera
locations in free motion to generate precise 3D models, thanks to AR-capable
mobile devices. To achieve real-world measurement, VolE is a reference- and
depth-free framework that leverages food video segmentation for food mask
generation. We also introduce a new food dataset encompassing the challenging
scenarios absent in the previous benchmarks. Our experiments demonstrate that
VolE outperforms the existing volume estimation techniques across multiple
datasets by achieving 2.22 % MAPE, highlighting its superior performance in
food volume estimation.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:03:05 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Haroon",
"Umair",
""
],
[
"AlMughrabi",
"Ahmad",
""
],
[
"Zoumpekas",
"Thanasis",
""
],
[
"Marques",
"Ricardo",
""
],
[
"Radeva",
"Petia",
""
]
] |
2505.10206 | Kendall Shepherd | Kendall G. Shepherd, Guglielmo Costa, Cristiano Ugolini, Guglielmo
Volpato, Diego Bossini, Cecilia Sgalletta, Francesco Addari, Alessandro
Bressan, Leo Girardi, and Mario Spera | Enhanced Mass Loss of Very Massive Stars: Impact on the Evolution,
Binary Processes, and Remnant Mass Spectrum | Submitted to A&A Section 7. Comments are welcome | null | null | null | astro-ph.SR astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | Very massive stars (VMS) play a fundamental role in astrophysics due to their
winds and supernovae (SN), and their role as massive black hole (BH)
progenitors. However, their origin and evolution remain a significant
challenge. Recent theoretical work and observations suggest that VMS
approaching the Eddington limit may experience mass loss above the standard
wind predictions. This study investigates how enhanced winds influence single
and binary VMS evolution, observable properties, and resulting BH populations.
New stellar wind prescriptions, sensitive to the Eddington parameter
($\Gamma_e$) and the luminosity-to-mass ratio, were implemented into the
stellar evolution code PARSEC v2.0. These updated single-star tracks (100 - 600
M$_{\odot}$ at Z=0.006) were used to model the VMS population in the Tarantula
Nebula and integrated into the SEVN binary evolution code. The
$\Gamma_e$-enhanced single-star tracks match observed VMS properties better
than standard models. Explaining the most massive star, R136a1, through a
single-star origin suggests a zero-age main sequence (ZAMS) mass limit of $<$
400 M$_{\odot}$ regardless of the wind recipe used. However, binary stellar
mergers also offer a suitable origin for R136a1 and other observed VMS,
potentially lowering the upper ZAMS mass limit by ~100 M$_{\odot}$. In
binaries, enhanced winds inhibit main-sequence stellar mergers and limit BH
production above the pair-instability mass gap's lower edge (~50 M$_{\odot}$).
Binary BHs merging in a Hubble time with enhanced winds yield more primary BHs
above 30 M$_{\odot}$ and enable secondary BHs between 30-40 solar masses, a
range not found with standard stellar winds at LMC metallicity. This study
highlights the crucial role that stellar winds and binary interactions play in
VMS evolution and offers predictions relevant for interpreting VMS observations
and gravitational wave source origins.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:03:50 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Shepherd",
"Kendall G.",
""
],
[
"Costa",
"Guglielmo",
""
],
[
"Ugolini",
"Cristiano",
""
],
[
"Volpato",
"Guglielmo",
""
],
[
"Bossini",
"Diego",
""
],
[
"Sgalletta",
"Cecilia",
""
],
[
"Addari",
"Francesco",
""
],
[
"Bressan",
"Alessandro",
""
],
[
"Girardi",
"Leo",
""
],
[
"Spera",
"Mario",
""
]
] |
2505.10207 | Minh Hang Nguyen | Allen Ibiapina, Minh Hang Nguyen, Mika\"el Rabie, Cl\'eoph\'ee Robin | How to Color Temporal Graphs to Ensure Proper Transitions | 20 pages, 9 figures | null | null | null | cs.DM | http://creativecommons.org/licenses/by/4.0/ | Graph Coloring consists in assigning colors to vertices ensuring that two
adjacent vertices do not have the same color. In dynamic graphs, this notion is
not well defined, as we need to decide if different colors for adjacent
vertices must happen all the time or not, and how to go from a coloring in one
time to the next one.
In this paper, we define a coloring notion for Temporal Graphs where at each
step, the coloring must be proper. It uses a notion of compatibility between
two consecutive snapshots that implies that the coloring stays proper while the
transition happens. Given a graph, the minimum number of colors needed to
ensure that such coloring exists is the \emph{Temporal Chromatic Number} of
this graph.
With those notions, we provide some lower and upper bounds for the temporal
chromatic number in the general case. We then dive into some specific classes
of graphs such as trees, graphs with bounded degree or bounded degeneracy.
Finally, we consider temporal graphs where grow pace is one, that is, a single
edge can be added and a single other one can be removed between two time steps.
In that case, we consider bipartite and bounded degree graphs.
Even though the problem is defined with full knowledge of the temporal graph,
our results also work in the case where future snapshots are given online: we
need to choose the coloring of the next snapshot after having computed the
current one, not knowing what
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:03:56 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ibiapina",
"Allen",
""
],
[
"Nguyen",
"Minh Hang",
""
],
[
"Rabie",
"Mikaël",
""
],
[
"Robin",
"Cléophée",
""
]
] |
2505.10208 | Ekrem O\u{g}uzhan Ang\"uner | Ekrem O\u{g}uzhan Ang\"uner | Prediction of radio-quiet gamma-ray pulsar distances using the
Fundamental Plane relation | Published in The Astrophysical Journal, 36 pages, 13 figures, 10
tables. Associated data and notebooks are available at Zenodo:
https://zenodo.org/records/15293227 | E. O. Ang\"uner 2025, ApJ, 985, 58 | 10.3847/1538-4357/adc914 | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | The fundamental plane (FP) relation connects gamma-ray luminosity to
intrinsic pulsar properties, offering the potential to estimate distances for
radio-quiet (RQ) gamma-ray pulsars, where direct measurements are often
unavailable. The Fermi Third Pulsar Catalog presents spectral data for 294
gamma-ray pulsars, including 72 RQ pulsars, of which only 9 have known
distances. This study investigates the FP relation's potential to predict
distances for RQ gamma-ray pulsars using machine learning (ML) approaches.
Ordinary least-squares regression was employed alongside ML methods, including
random forest and support vector regression, to predict RQ gamma-ray pulsar
distances. To ensure robustness, the analysis considered spectral cutoff
significance and outlier data. Results confirm that the FP relation is valid
only for pulsars exhibiting significant gamma-ray cutoffs, with FP exponents
closely matching theoretical expectations. It was shown that gamma-ray emission
origins significantly favor the curvature radiation regime over the synchrotron
regime at the 4.2 sigma level. The distances for 62 RQ pulsars, for which no
traditional measurements exist, are predicted for the first time. The
luminosity and spatial distributions of the considered RQ pulsar population
match known pulsar distributions. Predictions for RQ pulsars with known
distances are consistent with measured data. This study provides a proof of
principle that the FP relation, combined with ML methods, is a promising tool
for distance estimation in RQ gamma-ray pulsars, using gamma-ray emission
directly for the first time in pulsar distance estimates. Future improvements
can be achieved with larger data sets, precise spectral cutoff determinations,
and corrections for key factors like the beaming factor.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:07:09 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Angüner",
"Ekrem Oğuzhan",
""
]
] |
2505.10209 | Qianfeng Li | Dian Hu, Qianfeng Li, Yongqian Zhang | An Inverse Problem for Multi-Dimensional Piston Models with Large
Velocity Variations | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | When a circular symmetric piston suddenly expands into a still gas, a leading
shock wave is generated. This paper investigates an inverse problem of
reconstructing the trajectory of the piston from the given leading shock front
and the given initial flow conditions. We observe that in piston models, as the
initial density goes to zero, the piston approaches the shock front; however,
in the region between the piston and the shock front, the strict hyperbolicity
of the system degenerates. By applying asymptotic analysis, we provide
quantitative characterizations of the distance between the piston and the shock
front, and the degeneration of strict hyperbolicity. Consequently, by designing
appropriate a priori assumptions to balance the benefits and drawbacks arising
as the initial density approaches zero, we employ the method of characteristics
to prove the global-in-time existence of the piecewise smooth solution for this
inverse problem. In particular, the resulting flow structure exhibits
significant velocity variations.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:10:17 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Hu",
"Dian",
""
],
[
"Li",
"Qianfeng",
""
],
[
"Zhang",
"Yongqian",
""
]
] |
2505.10210 | Suresh C. Jaryal | Ayan Chatterjee, Suresh C. Jaryal, Akshay Kumar | Spherical trapped surfaces in n-dimensional general relativity | Published in Physical Review D | Phys. Rev. D \textbf{110}, no.10, 104059 (2024) | 10.1103/PhysRevD.110.104059 | null | gr-qc | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In this paper, we examine gravitational collapse of matter fields in
$n$-dimensional general relativity. The matter energy-momentum tensor under
consideration includes dust, perfect fluids with equations of state and matter
admitting bulk and shear viscosity. By adjusting various parameters of the
matter energy-momentum tensor, we determine the trapped region and spherical
marginally trapped surfaces in homogeneous and inhomogeneous models of
collapse. We show that, as expected, the time development of marginally trapped
tube is intricately related to the initial velocity and density profiles of the
collapsing matter configuration. This study clarifies the role of initial data
in the formation of spacetime singularity during gravitational collapse and
implies that, under generic conditions on the matter profiles, the central
spacetime singularity is always covered by a horizon.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:11:46 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Chatterjee",
"Ayan",
""
],
[
"Jaryal",
"Suresh C.",
""
],
[
"Kumar",
"Akshay",
""
]
] |
2505.10211 | Thomas Chuna | Alexander Benedix Robles, Phil-Alexander Hofmann, Thomas Chuna, Tobias
Dornheim, Michael Hecht | PyLIT: Reformulation and implementation of the analytic continuation
problem using kernel representation methods | 29 pages, 11 figures, submitted to Computer Physics Communications | null | null | null | physics.comp-ph physics.plasm-ph | http://creativecommons.org/licenses/by-sa/4.0/ | Path integral Monte Carlo (PIMC) simulations are a cornerstone for studying
quantum many-body systems. The analytic continuation (AC) needed to estimate
dynamic quantities from these simulations is an inverse Laplace transform,
which is ill-conditioned. If this inversion were surmounted, then dynamical
observables (e.g. dynamic structure factor (DSF) $S(q,\omega)$) could be
extracted from the imaginary-time correlation functions estimates.
Although of important, the AC problem remains challenging due to its
ill-posedness. To address this challenge, we express the DSF as a linear
combination of kernel functions with known Laplace transforms that have been
tailored to satisfy its physical constraints. We use least-squares optimization
regularized with a Bayesian prior to determine the coefficients of this linear
combination. We explore various regularization term, such as the commonly used
entropic regularizer, as well as the Wasserstein distance and $L^2$-distance as
well as techniques for setting the regularization weight. A key outcome is the
open-source package PyLIT (\textbf{Py}thon \textbf{L}aplace \textbf{I}nverse
\textbf{T}ransform), which leverages Numba and unifies the presented
formulations. PyLIT's core functionality is kernel construction and
optimization.
In our applications, we find PyLIT's DSF estimates share qualitative features
with other more established methods. We identify three key findings. Firstly,
independent of the regularization choice, utilizing non-uniform grid point
distributions reduced the number of unknowns and thus reduced our space of
possible solutions. Secondly, the Wasserstein distance, a previously unexplored
regularizer, performs as good as the entropic regularizer while benefiting from
its linear gradient. Thirdly, future work can meaningfully combine regularized
and stochastic optimization.
(text cut for char. limit)
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:13:17 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Robles",
"Alexander Benedix",
""
],
[
"Hofmann",
"Phil-Alexander",
""
],
[
"Chuna",
"Thomas",
""
],
[
"Dornheim",
"Tobias",
""
],
[
"Hecht",
"Michael",
""
]
] |
2505.10212 | Dario Di Palma | Dario Di Palma, Felice Antonio Merra, Maurizio Sfilio, Vito Walter
Anelli, Fedelucio Narducci, Tommaso Di Noia | Do LLMs Memorize Recommendation Datasets? A Preliminary Study on
MovieLens-1M | null | null | 10.1145/3726302.3730178 | null | cs.IR cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Large Language Models (LLMs) have become increasingly central to
recommendation scenarios due to their remarkable natural language understanding
and generation capabilities. Although significant research has explored the use
of LLMs for various recommendation tasks, little effort has been dedicated to
verifying whether they have memorized public recommendation dataset as part of
their training data. This is undesirable because memorization reduces the
generalizability of research findings, as benchmarking on memorized datasets
does not guarantee generalization to unseen datasets. Furthermore, memorization
can amplify biases, for example, some popular items may be recommended more
frequently than others.
In this work, we investigate whether LLMs have memorized public
recommendation datasets. Specifically, we examine two model families (GPT and
Llama) across multiple sizes, focusing on one of the most widely used dataset
in recommender systems: MovieLens-1M. First, we define dataset memorization as
the extent to which item attributes, user profiles, and user-item interactions
can be retrieved by prompting the LLMs. Second, we analyze the impact of
memorization on recommendation performance. Lastly, we examine whether
memorization varies across model families and model sizes. Our results reveal
that all models exhibit some degree of memorization of MovieLens-1M, and that
recommendation performance is related to the extent of memorization. We have
made all the code publicly available at:
https://github.com/sisinflab/LLM-MemoryInspector
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:16:36 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Di Palma",
"Dario",
""
],
[
"Merra",
"Felice Antonio",
""
],
[
"Sfilio",
"Maurizio",
""
],
[
"Anelli",
"Vito Walter",
""
],
[
"Narducci",
"Fedelucio",
""
],
[
"Di Noia",
"Tommaso",
""
]
] |
2505.10213 | Alireza Moradi | Mohammadmahdi Ghasemloo, Alireza Moradi | Informed Forecasting: Leveraging Auxiliary Knowledge to Boost LLM
Performance on Time Series Forecasting | null | null | null | null | cs.LG stat.AP | http://creativecommons.org/licenses/by-nc-sa/4.0/ | With the widespread adoption of Large Language Models (LLMs), there is a
growing need to establish best practices for leveraging their capabilities
beyond traditional natural language tasks. In this paper, a novel cross-domain
knowledge transfer framework is proposed to enhance the performance of LLMs in
time series forecasting -- a task of increasing relevance in fields such as
energy systems, finance, and healthcare. The approach systematically infuses
LLMs with structured temporal information to improve their forecasting
accuracy. This study evaluates the proposed method on a real-world time series
dataset and compares it to a naive baseline where the LLM receives no auxiliary
information. Results show that knowledge-informed forecasting significantly
outperforms the uninformed baseline in terms of predictive accuracy and
generalization. These findings highlight the potential of knowledge transfer
strategies to bridge the gap between LLMs and domain-specific forecasting
tasks.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:17:52 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ghasemloo",
"Mohammadmahdi",
""
],
[
"Moradi",
"Alireza",
""
]
] |
2505.10214 | Derrick Carr | D. S. Carr, S. J. Kappannan, Z. L. Hutchens, M. S. Polimera, M. A.
Norris, K. D. Eckert, A. J. Moffett | Using Machine Learning to Estimate NUV Magnitudes and Probe Quenching
Mechanisms of $z$=0 Nuggets in the RESOLVE and ECO Surveys | 20 pages, 7 figures | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | We present a $z=0$ census of nuggets -- compact galaxies that form via
gas-rich violent disk instability -- within the luminosity- and volume-limited
REsolved Spectroscopy Of a Local VolumE (RESOLVE) and Environmental COntext
(ECO) surveys. We use random forest (RF) models to predict near-ultraviolet
(NUV) magnitudes for ECO galaxies that lack high-quality NUV magnitudes,
thereby doubling the number of ECO galaxies with reliable extinction-corrected
star formation rates (SFRs) and red/green/blue classifications based on
specific SFRs (sSFRs). The resulting RF-enhanced RESOLVE+ECO nugget sample
allows us to analyze rare subpopulations -- green nuggets and nuggets with
active galactic nuclei (AGN) -- likely associated with quenching. Green nuggets
are more similar to red nuggets than to blue nuggets in halo mass
($M_{\text{halo}}$) distribution, with both red and green nuggets being found
mainly at $M_{\text{halo}} \ge 10^{11.4} M_\odot$, where permanent halo
quenching is predicted. At these masses, the AGN frequency for green nuggets is
higher ($\text{48.2\%}^{+5.3\%}_{-5.3\%}$) than for either blue
($\text{39.2\%}^{+2.9\%}_{-2.8\%}$) or red ($\text{29.3\%}^{+3.0\%}_{-2.8\%}$)
nuggets. Between $M_{halo} = 10^{11.4}-10^{12} M_\odot$, at the onset of
permanent quenching, the AGN frequency for green nuggets is nearly double the
frequency for blue or red nuggets, implying AGN are associated with this
transition. At $M_{halo} < 10^{11.4} M_\odot$, where temporary cyclic quenching
is expected, the AGN frequency for blue nuggets
($\text{7.5\%}^{+1.4\%}_{-1.2\%}$) is lower than for either green
($\text{31.3\%}^{+8.7\%}_{-7.5\%}$) or red ($\text{18.8\%}^{+11.5\%}_{-7.8\%}$)
nuggets. At all masses, nuggets with AGN have reduced sSFRs and likely also
atomic gas content compared to nuggets without AGN, but the quenching is more
extreme below $M_{halo} = 10^{11.4} M_\odot$.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:19:30 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Carr",
"D. S.",
""
],
[
"Kappannan",
"S. J.",
""
],
[
"Hutchens",
"Z. L.",
""
],
[
"Polimera",
"M. S.",
""
],
[
"Norris",
"M. A.",
""
],
[
"Eckert",
"K. D.",
""
],
[
"Moffett",
"A. J.",
""
]
] |
2505.10215 | Andrea Colagrossi Dr. | Xiao-Ting Huang, Peng-Nan Sun, Hong-Guan Lyu, Andrea Colagrossi, A-Man
Zhang | Extension of the consistent $\delta^{+}$-SPH model for multiphase flows
considering the compressibility of different phases | 49 pages, 33 figures | Journal of Computational Physics 535 (2025) 114031 | 10.1016/j.jcp.2025.114031 | null | physics.flu-dyn physics.comp-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In hydrodynamic problems involving wave impact on structures, air
compressibility is crucial for accurate pressure prediction when an air bubble
is entrapped. In this work, the consistent $\delta^{+}$-SPH model, originally
developed for single-phase scenarios, is extended to multiphase contexts.
Although the consistent $\delta^{+}$-SPH model shows good performance for
single phase and viscous flow simulations, extending it to multiphase scenarios
presents challenges, such as proper implementation of particle shifting for
multiphase interfaces. Therefore, within the framework of the consistent
$\delta^{+}$-SPH, we introduce the following enhancements: firstly, new
strategy for handling $\delta \boldmath{u}$-terms given by the particle
shifting technique at multiphase interfaces are proposed to maintain stability
and conservation. Secondly, for modeling of incompressible phases, like water,
an acoustic damper term is introduced to alleviate acoustic waves resulting
from the weakly-compressible assumption, which is expected to achieve smooth
pressure field comparable to truly-incompressible hypothesis, thereby reducing
the nonphysical pressure wave during the violent impact state; for modeling
compressible phases like air, a physical sound speed is adopted in the equation
of state to accurately model real gas phase compressibility. To test and
validate the present multiphase SPH model, simulations were conducted for six
scenarios. In particular, except for sloshing with two-layer liquids, the other
scenarios fully consider air pressure oscillations when air is entrapped,
compressed, or expanded by surrounding flows. The results demonstrate
significant advantages of the present SPH model in simulating multiphase
problems involving strong liquid impact and different phase compressibility.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:20:01 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Huang",
"Xiao-Ting",
""
],
[
"Sun",
"Peng-Nan",
""
],
[
"Lyu",
"Hong-Guan",
""
],
[
"Colagrossi",
"Andrea",
""
],
[
"Zhang",
"A-Man",
""
]
] |
2505.10216 | Denis Garoli | Yanqiu Zou, Luca Mattarozzi, Huaizhou Jin, Qifei Ma, Sandro Cattarin,
Shukun Weng, Ali Douaki, German Lanzavecchia, Karol Ko{\l}\k{a}taj, Corduri
Nicco, Ben Johns, Nicol\`o Maccaferri, Guillermo Acuna, Zhenrong Zheng,
Shangzhong Jin and Denis Garoli | UV-SERS monitoring of plasmons photodegradation of biomolecules on
Aluminum platforms decorated with Rhodium nanoparticles | null | null | null | null | physics.app-ph | http://creativecommons.org/licenses/by/4.0/ | In the search for novel nanostructured materials for UV plasmonics a limited
number of choices can be done. Materials such as aluminum, rhodium, gallium and
few others can be used. One of the most interesting application for UV
plasmonics is Surface Enhanced Raman Spectroscopy. It can be extended to this
spectral range to explore spectral properties of biomolecules that have only a
small cross section in the visible spectral range. We have recently reported on
a functional substrates based on nanoporous aluminum decorated with rhodium
nanoparticles. This system showed an interesting behavior for UV excitation at
266 nm, with an unexpected decreasing Raman intensity for increasing rhodium
nanoparticles concentrations. We proposed that this effect can be due to the
difficult access to the hot spots for the molecules deposited via thermal
evaporation. Here we extend this study exploring the performance of the system
at another UV excitation wavelengths (325 nm) reporting on experimental results
obtained using a deposition process that can bring the molecules at the
hot-spots in a more efficient way. Extensive spectroscopic acquisitions,
combined with 3D maps, allow to shade a more clear view on the performance of
this plasmonic platform. In particular, the photodegration and the potential
oxidation of biomolecules driven by the hot-electron/hot-holes produced by the
rhodium nanoparticles will be reported.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:20:55 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Zou",
"Yanqiu",
""
],
[
"Mattarozzi",
"Luca",
""
],
[
"Jin",
"Huaizhou",
""
],
[
"Ma",
"Qifei",
""
],
[
"Cattarin",
"Sandro",
""
],
[
"Weng",
"Shukun",
""
],
[
"Douaki",
"Ali",
""
],
[
"Lanzavecchia",
"German",
""
],
[
"Kołątaj",
"Karol",
""
],
[
"Nicco",
"Corduri",
""
],
[
"Johns",
"Ben",
""
],
[
"Maccaferri",
"Nicolò",
""
],
[
"Acuna",
"Guillermo",
""
],
[
"Zheng",
"Zhenrong",
""
],
[
"Jin",
"Shangzhong",
""
],
[
"Garoli",
"Denis",
""
]
] |
2505.10217 | Ramon Nou | Petar Andri\'c and Aaron Call and Ramon Nou | Enabling Syscall Intercept for RISC-V | RISC-V summit 2025 accepted | null | null | null | cs.OS cs.AR | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The European Union technological sovereignty strategy centers around the
RISC-V Instruction Set Architecture, with the European Processor Initiative
leading efforts to build production-ready processors. Focusing on realizing a
functional RISC-V ecosystem, the BZL initiative (www.bzl.es) is making an
effort to create a software stack along with the hardware. In this work, we
detail the efforts made in porting a widely used syscall interception library,
mainly used on AdHocFS (i.e., DAOS, GekkoFS), to RISC-V and how we overcame
some of the limitations encountered.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:21:16 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Andrić",
"Petar",
""
],
[
"Call",
"Aaron",
""
],
[
"Nou",
"Ramon",
""
]
] |
2505.10218 | Bowen Wu | Zongsheng Wang, Kaili Sun, Bowen Wu, Qun Yu, Ying Li, Baoxun Wang | RAIDEN-R1: Improving Role-awareness of LLMs via GRPO with Verifiable
Reward | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Role-playing conversational agents (RPCAs) face persistent challenges in
maintaining role consistency. To address this, we propose RAIDEN-R1, a novel
reinforcement learning framework that integrates Verifiable Role-Awareness
Reward (VRAR). The method introduces both singular and multi-term mining
strategies to generate quantifiable rewards by assessing role-specific keys.
Additionally, we construct a high-quality, role-aware Chain-of-Thought dataset
through multi-LLM collaboration, and implement experiments to enhance reasoning
coherence. Experiments on the RAIDEN benchmark demonstrate RAIDEN-R1's
superiority: our 14B-GRPO model achieves 88.04% and 88.65% accuracy on
Script-Based Knowledge and Conversation Memory metrics, respectively,
outperforming baseline models while maintaining robustness. Case analyses
further reveal the model's enhanced ability to resolve conflicting contextual
cues and sustain first-person narrative consistency. This work bridges the
non-quantifiability gap in RPCA training and provides insights into role-aware
reasoning patterns, advancing the development of RPCAs.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:22:10 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Wang",
"Zongsheng",
""
],
[
"Sun",
"Kaili",
""
],
[
"Wu",
"Bowen",
""
],
[
"Yu",
"Qun",
""
],
[
"Li",
"Ying",
""
],
[
"Wang",
"Baoxun",
""
]
] |
2505.10219 | Theo Gruner | Maximilian T\"olle, Theo Gruner, Daniel Palenicek, Tim Schneider,
Jonas G\"unster, Joe Watson, Davide Tateo, Puze Liu, Jan Peters | Towards Safe Robot Foundation Models Using Inductive Biases | 14 pages, 5 figures | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Safety is a critical requirement for the real-world deployment of robotic
systems. Unfortunately, while current robot foundation models show promising
generalization capabilities across a wide variety of tasks, they fail to
address safety, an important aspect for ensuring long-term operation. Current
robot foundation models assume that safe behavior should emerge by learning
from a sufficiently large dataset of demonstrations. However, this approach has
two clear major drawbacks. Firstly, there are no formal safety guarantees for a
behavior cloning policy trained using supervised learning. Secondly, without
explicit knowledge of any safety constraints, the policy may require an
unreasonable number of additional demonstrations to even approximate the
desired constrained behavior. To solve these key issues, we show how we can
instead combine robot foundation models with geometric inductive biases using
ATACOM, a safety layer placed after the foundation policy that ensures safe
state transitions by enforcing action constraints. With this approach, we can
ensure formal safety guarantees for generalist policies without providing
extensive demonstrations of safe behavior, and without requiring any specific
fine-tuning for safety. Our experiments show that our approach can be
beneficial both for classical manipulation tasks, where we avoid unwanted
collisions with irrelevant objects, and for dynamic tasks, such as the robot
air hockey environment, where we can generate fast trajectories respecting
complex tasks and joint space constraints.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:22:21 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Tölle",
"Maximilian",
""
],
[
"Gruner",
"Theo",
""
],
[
"Palenicek",
"Daniel",
""
],
[
"Schneider",
"Tim",
""
],
[
"Günster",
"Jonas",
""
],
[
"Watson",
"Joe",
""
],
[
"Tateo",
"Davide",
""
],
[
"Liu",
"Puze",
""
],
[
"Peters",
"Jan",
""
]
] |
2505.10220 | Peilan Wang | Peilan Wang, Yu Xue, Weidong Mei, Jun Fang, and Rui Zhang | UAV-Enabled Passive 6DMA for ISAC: Joint Location, Orientation, and
Reflection Optimization | 5 pages | null | null | null | eess.SP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Improving the fundamental performance trade-off in integrated sensing and
communication (ISAC) systems has been deemed as one of the most significant
challenges. To address it, we propose in this letter a novel ISAC system that
leverages an unmanned aerial vehicle (UAV)-mounted intelligent reflecting
surface (IRS) and the UAV's maneuverability in six-dimensional (6D) space,
i.e., three-dimensional (3D) location and 3D rotation, thus referred to as
passive 6D movable antenna (6DMA). We aim to maximize the signal-to-noise ratio
(SNR) for sensing a single target while ensuring a minimum SNR at a
communication user equipment (UE), by jointly optimizing the transmit
beamforming at the ISAC base station (BS), the 3D location and orientation as
well as the reflection coefficients of the IRS. To solve this challenging
non-convex optimization problem, we propose a two-stage approach. In the first
stage, we aim to optimize the IRS's 3D location, 3D orientation, and reflection
coefficients to enhance both the channel correlations and power gains for
sensing and communication. Given their optimized parameters, the optimal
transmit beamforming at the ISAC BS is derived in closed form. Simulation
results demonstrate that the proposed passive 6DMA-enabled ISAC system
significantly improves the sensing and communication trade-off by
simultaneously enhancing channel correlations and power gains, and outperforms
other baseline schemes.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:23:12 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Wang",
"Peilan",
""
],
[
"Xue",
"Yu",
""
],
[
"Mei",
"Weidong",
""
],
[
"Fang",
"Jun",
""
],
[
"Zhang",
"Rui",
""
]
] |
2505.10221 | Mustafa Kemal D\"oner | Mustafa Kemal D\"oner | Entanglement generation and dynamical equilibrium within de Broglie Bohm
theory | 9 pages, 2 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we reassess our previous hybrid semiclassical gravity model and
demonstrate its ability to induce entanglement even in the limit of weak
entanglement. By incorporating feedback terms sourced from gravitational
potential interactions between subsystems, based on a refined version of Theory
of (Exclusively) Local Beables developed by Travis Norsen, we show that these
interactions alone can generate entanglement. Building on this insight, we
propose a novel communication scheme and introduce the concept of dynamical
equilibrium, which formalizes a stable strategy for subsystems. This framework
is inspired by the quantum equilibrium hypothesis of de Broglie Bohm theory and
causal interpretation of David Bohm and Jean Pierre Vigier.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:24:33 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Döner",
"Mustafa Kemal",
""
]
] |
2505.10222 | Jintian Shao | Jintian Shao, Hongyi Huang, Jiayi Wu, Beiwen Zhang, ZhiYu Wu, You
Shan, MingKai Zheng | ComplexFormer: Disruptively Advancing Transformer Inference Ability via
Head-Specific Complex Vector Attention | null | null | null | null | cs.LG cs.CL | http://creativecommons.org/licenses/by/4.0/ | Transformer models rely on self-attention to capture token dependencies but
face challenges in effectively integrating positional information while
allowing multi-head attention (MHA) flexibility. Prior methods often model
semantic and positional differences disparately or apply uniform positional
adjustments across heads, potentially limiting representational capacity. This
paper introduces ComplexFormer, featuring Complex Multi-Head Attention-CMHA.
CMHA empowers each head to independently model semantic and positional
differences unified within the complex plane, representing interactions as
rotations and scaling. ComplexFormer incorporates two key improvements: (1) a
per-head Euler transformation, converting real-valued query/key projections
into polar-form complex vectors for head-specific complex subspace operation;
and (2) a per-head adaptive differential rotation mechanism,
exp[i(Adapt(ASmn,i) + Delta(Pmn),i)], allowing each head to learn distinct
strategies for integrating semantic angle differences (ASmn,i) with relative
positional encodings (Delta(Pmn),i). Extensive experiments on language
modeling, text generation, code generation, and mathematical reasoning show
ComplexFormer achieves superior performance, significantly lower generation
perplexity , and improved long-context coherence compared to strong baselines
like RoPE-Transformers. ComplexFormer demonstrates strong parameter efficiency,
offering a more expressive, adaptable attention mechanism.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:30:33 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Shao",
"Jintian",
""
],
[
"Huang",
"Hongyi",
""
],
[
"Wu",
"Jiayi",
""
],
[
"Zhang",
"Beiwen",
""
],
[
"Wu",
"ZhiYu",
""
],
[
"Shan",
"You",
""
],
[
"Zheng",
"MingKai",
""
]
] |
2505.10223 | Puru Vaish | Puru Vaish, Felix Meister, Tobias Heimann, Christoph Brune, Jelmer M.
Wolterink | Data-Agnostic Augmentations for Unknown Variations: Out-of-Distribution
Generalisation in MRI Segmentation | Accepted at MIDL 2025 | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Medical image segmentation models are often trained on curated datasets,
leading to performance degradation when deployed in real-world clinical
settings due to mismatches between training and test distributions. While data
augmentation techniques are widely used to address these challenges,
traditional visually consistent augmentation strategies lack the robustness
needed for diverse real-world scenarios. In this work, we systematically
evaluate alternative augmentation strategies, focusing on MixUp and Auxiliary
Fourier Augmentation. These methods mitigate the effects of multiple variations
without explicitly targeting specific sources of distribution shifts. We
demonstrate how these techniques significantly improve out-of-distribution
generalization and robustness to imaging variations across a wide range of
transformations in cardiac cine MRI and prostate MRI segmentation. We
quantitatively find that these augmentation methods enhance learned feature
representations by promoting separability and compactness. Additionally, we
highlight how their integration into nnU-Net training pipelines provides an
easy-to-implement, effective solution for enhancing the reliability of medical
segmentation models in real-world applications.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:32:02 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Vaish",
"Puru",
""
],
[
"Meister",
"Felix",
""
],
[
"Heimann",
"Tobias",
""
],
[
"Brune",
"Christoph",
""
],
[
"Wolterink",
"Jelmer M.",
""
]
] |
2505.10224 | Pietro Dardano | Pietro Dardano, Paolo Rocco and David Frisini | Force-Driven Validation for Collaborative Robotics in Automated Avionics
Testing | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-sa/4.0/ | ARTO is a project combining collaborative robots (cobots) and Artificial
Intelligence (AI) to automate functional test procedures for civilian and
military aircraft certification. This paper proposes a Deep Learning (DL) and
eXplainable AI (XAI) approach, equipping ARTO with interaction analysis
capabilities to verify and validate the operations on cockpit components.
During these interactions, forces, torques, and end effector poses are recorded
and preprocessed to filter disturbances caused by low performance force
controllers and embedded Force Torque Sensors (FTS). Convolutional Neural
Networks (CNNs) then classify the cobot actions as Success or Fail, while also
identifying and reporting the causes of failure. To improve interpretability,
Grad CAM, an XAI technique for visual explanations, is integrated to provide
insights into the models decision making process. This approach enhances the
reliability and trustworthiness of the automated testing system, facilitating
the diagnosis and rectification of errors that may arise during testing.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:32:46 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Dardano",
"Pietro",
""
],
[
"Rocco",
"Paolo",
""
],
[
"Frisini",
"David",
""
]
] |
2505.10225 | Ali Ghorashi | Ali Ghorashi, Nicholas Rivera, Ravishankar Sundararaman, Efthimios
Kaxiras, John Joannopoulos, Marin Solja\v{c}i\'c | Alkali Intercalation of Moire Heterostructures for Low-Loss Plasmonics | null | null | null | null | cond-mat.mtrl-sci cond-mat.mes-hall | http://creativecommons.org/licenses/by/4.0/ | Two-dimensional metals generically support gapless plasmons with wavelengths
well below the wavelength of free-space radiation at the same frequency.
Typically, however, this substantial confinement of electromagnetic energy is
associated with commensurately high losses, and mitigating such losses may only
be achieved through judicious band structure engineering near the Fermi level.
In a clean system, an isolated, moderately flat, band at the Fermi level with
sufficiently high carrier density can support a plasmon that is immune to
propagation losses up to some order in the electron-phonon interaction.
However, proposed materials that satisfy these criteria have been
ferromagnetic, structurally unstable, or otherwise difficult to fabricate.
Here, we propose a class of band structure engineered materials that evade
these typical pitfalls -- Moire heterostructures of hexagonal boron nitride
intercalated with alkali atoms. We find that only sodium atoms engender a
sufficiently isolated band with plasmons lossless at first order in the
electron-phonon interaction. We calculate higher order electron-phonon losses
and find that at frequencies of about $1$eV the electron-phonon decay mechanism
is negligible -- leading to a contribution to the decay rate of about 10^7 Hz
in a small frequency range. We next calculate losses from the electron-electron
interaction and find that this is the dominant process -- leading plasmons to
decay to lower frequency plasmons at a rate of around 10^14 Hz.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:33:38 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ghorashi",
"Ali",
""
],
[
"Rivera",
"Nicholas",
""
],
[
"Sundararaman",
"Ravishankar",
""
],
[
"Kaxiras",
"Efthimios",
""
],
[
"Joannopoulos",
"John",
""
],
[
"Soljačić",
"Marin",
""
]
] |
2505.10226 | Yanxiang Wang | Yanxiang Wang, Yihe Yan, Jiawei Hu, Cheng Jiang, Brano Kusy, Ashraf
Uddin, Mahbub Hassan, Wen Hu | Solar-CSK: Decoding Color Coded Visible Light Communications using Solar
Cells | 14 pages, 25 figures | null | null | null | cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visible Light Communication (VLC) provides an energy-efficient wireless
solution by using existing LED-based illumination for high-speed data
transmissions. Although solar cells offer the advantage of simultaneous energy
harvesting and data reception, their broadband nature hinders accurate decoding
of color-coded signals like Color Shift Keying (CSK). In this paper, we propose
a novel approach exploiting the concept of tandem solar cells, multi-layer
devices with partial wavelength selectivity, to capture coarse color
information without resorting to energy-limiting color filters. To address the
residual spectral overlap, we develop a bidirectional LSTM-based machine
learning framework that infers channel characteristics by comparing solar
cells' photovoltaic signals with pilot-based anchor data. Our commercial
off-the-shelf (COTS) solar prototype achieves robust performance across varying
distances and ambient lighting levels, significantly reducing bit error rates
compared to conventional channel estimation methods. These findings mark a step
toward sustainable, high-performance VLC systems powered by the multi-layer
solar technologies.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:33:54 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Wang",
"Yanxiang",
""
],
[
"Yan",
"Yihe",
""
],
[
"Hu",
"Jiawei",
""
],
[
"Jiang",
"Cheng",
""
],
[
"Kusy",
"Brano",
""
],
[
"Uddin",
"Ashraf",
""
],
[
"Hassan",
"Mahbub",
""
],
[
"Hu",
"Wen",
""
]
] |
2505.10227 | Simon Oyarzun | S. Oyarz\'un, C. Gonzalez-Fuentes, Erick Burgos, M. Myronov, M. Jamet,
R. L.Rodr\'iguez-Su\'arez, and F. Pezzoli | Giant spin-to-charge conversion in germanium tin epilayers | 15 pages, 4 figures | null | null | null | cond-mat.mtrl-sci cond-mat.mes-hall | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We report a study of the spin-to-charge current conversion in compressively
strained Ge1-xSnx alloy epilayers as a function of the Sn concentration by
means of the inverse spin Hall effect (ISHE). The spin current is generated by
spin-pumping effect (SPE) from a thin NiFe layer driven into ferromagnetic
resonance (FMR). By simultaneously measuring the magnetic damping of the NiFe
layer and the ISHE-induced charge current we extract two key spintronics
parameters: the spin Hall angle and the effective spin mixing conductance. Our
results reveal a giant spin-to-charge conversion and a non-monotonic dependence
of the charge current signal on the Sn concentration, consistent with the
variation in the magnetic damping observed in FMR. The values of spin Hall
angle are comparable to those reported for heavy metals such as Pt and Ta.
Furthermore, we show that the spin conductivity at the Au/GeSn interface can be
enhanced by tuning the Sn concentration.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:37:30 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Oyarzún",
"S.",
""
],
[
"Gonzalez-Fuentes",
"C.",
""
],
[
"Burgos",
"Erick",
""
],
[
"Myronov",
"M.",
""
],
[
"Jamet",
"M.",
""
],
[
"Rodríguez-Suárez",
"R. L.",
""
],
[
"Pezzoli",
"F.",
""
]
] |
2505.10228 | Anusha Srikanthan | Anusha Srikanthan, Hanli Zhang, Spencer Folk, Vijay Kumar, Nikolai
Matni | Quad-LCD: Layered Control Decomposition Enables Actuator-Feasible
Quadrotor Trajectory Planning | 4 pages, 4 figures | ICRA 2025 Workshop on 25 Years of Aerial Robotics: Challenges and
Opportunities | null | null | cs.RO cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we specialize contributions from prior work on data-driven
trajectory generation for a quadrotor system with motor saturation constraints.
When motors saturate in quadrotor systems, there is an ``uncontrolled drift" of
the vehicle that results in a crash. To tackle saturation, we apply a control
decomposition and learn a tracking penalty from simulation data consisting of
low, medium and high-cost reference trajectories. Our approach reduces crash
rates by around $49\%$ compared to baselines on aggressive maneuvers in
simulation. On the Crazyflie hardware platform, we demonstrate feasibility
through experiments that lead to successful flights. Motivated by the growing
interest in data-driven methods to quadrotor planning, we provide open-source
lightweight code with an easy-to-use abstraction of hardware platforms.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:37:35 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Srikanthan",
"Anusha",
""
],
[
"Zhang",
"Hanli",
""
],
[
"Folk",
"Spencer",
""
],
[
"Kumar",
"Vijay",
""
],
[
"Matni",
"Nikolai",
""
]
] |
2505.10229 | Kun Yin | Kun Yin | Strong and weak convergence rates for fully coupled multiscale
stochastic differential equations driven by $\alpha$-stable processes | 33 pages | null | null | null | math.PR | http://creativecommons.org/licenses/by/4.0/ | We first establish strong convergence rates for multiscale systems driven by
$\alpha$-stable processes, with analyses constructed in two distinct scaling
regimes. When addressing weak convergence rates of this system, we derive four
averaged equations with respect to four scaling regimes. Notably, under
sufficient H\"{o}lder regularity conditions on the time-dependent drifts of
slow process, the strong convergence orders are related to the known optimal
strong convergence order $1-\frac{1}{\alpha}$, and the weak convergence orders
are 1. Our primary approach involves employing nonlocal Poisson equations to
construct ``corrector equations" that effectively eliminate inhomogeneous
terms.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:38:50 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Yin",
"Kun",
""
]
] |
2505.10230 | J\'ozsef Kolumb\'an J. | Borb\'ala Fazekas and J\'ozsef J. Kolumb\'an | Estimating the convex relaxation of the ideal magnetohydrodynamics
equations | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the explicit convex relaxation of the ideal
magnetohydrodynamics equations. We provide a non-trivial lower estimate on the
lamination hull and an upper estimate on the $\Lambda$-convex hull, the latter
providing inequalities which will be satisfied by weak limits of weak solution
of the ideal MHD equations, which serve as a model of averaged turbulent
magnetohydrodynamical flows.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:42:45 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Fazekas",
"Borbála",
""
],
[
"Kolumbán",
"József J.",
""
]
] |
2505.10231 | Haozhe Luo | Haozhe Luo, Ziyu Zhou, Zixin Shu, Aur\'elie Pahud de Mortanges, Robert
Berke, Mauricio Reyes | On the Interplay of Human-AI Alignment,Fairness, and Performance
Trade-offs in Medical Imaging | null | null | null | null | cs.CV cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural networks excel in medical imaging but remain prone to biases,
leading to fairness gaps across demographic groups. We provide the first
systematic exploration of Human-AI alignment and fairness in this domain. Our
results show that incorporating human insights consistently reduces fairness
gaps and enhances out-of-domain generalization, though excessive alignment can
introduce performance trade-offs, emphasizing the need for calibrated
strategies. These findings highlight Human-AI alignment as a promising approach
for developing fair, robust, and generalizable medical AI systems, striking a
balance between expert guidance and automated efficiency. Our code is available
at https://github.com/Roypic/Aligner.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:43:23 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Luo",
"Haozhe",
""
],
[
"Zhou",
"Ziyu",
""
],
[
"Shu",
"Zixin",
""
],
[
"de Mortanges",
"Aurélie Pahud",
""
],
[
"Berke",
"Robert",
""
],
[
"Reyes",
"Mauricio",
""
]
] |
2505.10232 | Gwena\"el Ferrando | Gwena\"el Ferrando, Amit Sever, Elior Urisman | Correlators of Line Defect and Local Operator in Conformal Field
Theories with a Slightly Broken Higher-Spin Symmetry | 47 pages, 3 figures | null | null | BONN-TH-2025-19 | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study three-dimensional conformal field theories with a large-$N$ limit.
Leveraging the framework of slightly broken higher-spin symmetry, we bootstrap
correlation functions between the single-trace, local operators and straight,
conformal line defects with boundaries. These correlation functions, which
depend on a single conformal cross-ratio, encapsulate all bulk-defect operator
product expansion coefficients. Concentrating on the quasi-fermionic theory, we
explicitly compute all correlators involving the spin-zero and spin-one
conserved currents, along with an infinite family of correlators involving the
higher-spin currents. Furthermore, we demonstrate that the dependence of these
correlators on the defect's shape is fully determined by our bootstrap
constraints.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:45:36 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ferrando",
"Gwenaël",
""
],
[
"Sever",
"Amit",
""
],
[
"Urisman",
"Elior",
""
]
] |
2505.10233 | Yucheng Qiu | Sudhakantha Girmohanta, Yu-Cheng Qiu | Quest for a phenomenologically consistent low cutoff theory | 8 pages (including 2 appendices), 6 figures, 1 table | null | null | null | hep-ph hep-th | http://creativecommons.org/licenses/by/4.0/ | The Randall-Sundrum model with the Higgs localized on the IR brane solves the
gauge hierarchy problem. However, the associated low cutoff ($\Lambda \sim 10$
TeV) generically leads to unacceptably rapid nucleon decay and excessively
large Majorana neutrino masses. Achieving consistency while simultaneously
explaining the Yukawa hierarchy requires either a horizontal symmetry or a
discrete gauged symmetry. We demonstrate that eliminating all dangerous
operators within a horizontal symmetry framework must come with large and
unattractive charge assignments, if possible at all. Hence, we consider an
exact discrete gauged $\mathbf{Z}_N$ symmetry, with fermion mass hierarchies
generated via wavefunction overlap. We employ this to reproduce the current CKM
and PMNS structures. Under minimal assumptions, it predicts Dirac neutrinos
with a total mass $\sim 66$ meV. Since the $\mathbf{Z}_N$ charges must be
generation-blind, flavor observables serve as key probes.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:45:53 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Girmohanta",
"Sudhakantha",
""
],
[
"Qiu",
"Yu-Cheng",
""
]
] |
2505.10234 | Arun Ashok | A.Ashok, A.Cabrera, S.Baje, A.Zambanini, K. Allinger, A.Bahr, and S.
van Waasen | Self Clocked Digital LDO for Cryogenic Power Management in 22nm FDSOI
with 98 Percent Efficiency | null | null | null | null | quant-ph cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A universal quantum computer~(QC), though promising ground breaking solutions
to complex problems, still faces several challenges with respect to
scalability. Current state-of-the-art QC use a great quantity of cables to
connect the physical qubits, situated in the cryogenic temperature, to room
temperature electronics. Integrated cryogenic electronics together with
semiconductor spin qubits is one way closer for scalability. Such a scalable
quantum computer can have qubits and the control electronics at 4K stage. Being
at 4K, more thermal dissipation is allowed without overloading the cooling
capability of the fridge. Still, control and power circuitry is expected to be
highly efficient. While commercial CMOS technologies are found to be operatable
at \qty{}{mK}, lack of reliable cryogenic models while designing, increased
mismatches at cryo temperatures makes the design challenging and risky. Using
an FDSOI technology with backgate biasing to compensate for the threshold
voltage drift happening at cryo~(compensating around 200mV) and digital
circuitry is a way to address this challenge. In this work, a self-clocked
digital low dropout regulator (DLDO) is designed in FDSOI for high power
efficient, variation tolerant regulator to supply cryogenic circuits for
Quantum computing. The proposed digital LDO is more resilient to mismatch and
having self clocking and close and fine loops addresses the power efficiency
and faster transient response.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:45:55 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ashok",
"A.",
""
],
[
"Cabrera",
"A.",
""
],
[
"Baje",
"S.",
""
],
[
"Zambanini",
"A.",
""
],
[
"Allinger",
"K.",
""
],
[
"Bahr",
"A.",
""
],
[
"van Waasen",
"S.",
""
]
] |
2505.10235 | Reginald Christian Bernardo | Reginald Christian Bernardo and Seoktae Koh and Gansukh Tumurtushaa | Implications of pulsar timing arrays for Gauss-Bonnet Inflation | 19 pages + refs, 4 figures, comments welcome | null | null | null | astro-ph.CO astro-ph.HE gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Correlated time-of-arrival measurements by pulsar timing arrays (PTAs) have
provided a new means of constraining astrophysical or cosmological models that
produce a gravitational wave (GW) background. For this work, we discuss the
implications of PTA observations for Gauss-Bonnet (GB) inflationary models
through the production and propagation of inflationary GWs. We show that our GB
inflationary scenario is consistent with present PTA and cosmological data. A
blue-tilted tensor power spectrum supported by PTAs can be naturally
accommodated in GB inflation. Using observational constraints, we derive
general conditions for the inflaton potential and the GB coupling function,
suggesting that in GB inflation, the inflaton must climb up the potential
before rolling downhill and reaching the end of inflation. We provide two
concrete GB inflationary models to demonstrate the viability of this mechanism.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:47:51 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Bernardo",
"Reginald Christian",
""
],
[
"Koh",
"Seoktae",
""
],
[
"Tumurtushaa",
"Gansukh",
""
]
] |
2505.10236 | Rob Bemthuis | Rob H. Bemthuis | Lost in Models? Structuring Managerial Decision Support in Process
Mining with Multi-criteria Decision Making | 8 pages, 2 figures, 2 tables; to appear in Intelligent Information
Systems, Lecture Notes in Business Information Processing (LNBIP) Vol. 557,
Springer, 2025. Part of 37th International Conference on Advanced Information
Systems Engineering | null | null | null | cs.CY cs.DB | http://creativecommons.org/licenses/by/4.0/ | Process mining is increasingly adopted in modern organizations, producing
numerous process models that, while valuable, can lead to model overload and
decision-making complexity. This paper explores a multi-criteria
decision-making (MCDM) approach to evaluate and prioritize process models by
incorporating both quantitative metrics (e.g., fitness, precision) and
qualitative factors (e.g., cultural fit). An illustrative logistics example
demonstrates how MCDM, specifically the Analytic Hierarchy Process (AHP),
facilitates trade-off analysis and promotes alignment with managerial
objectives. Initial insights suggest that the MCDM approach enhances
context-sensitive decision-making, as selected models address both operational
metrics and broader managerial needs. While this study is an early-stage
exploration, it provides an initial foundation for deeper exploration of
MCDM-driven strategies to enhance the role of process mining in complex
organizational settings.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:49:56 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Bemthuis",
"Rob H.",
""
]
] |
2505.10237 | Sahab Singh | Sahab Singh, D. Choudhury, B. Maheshwari, R. Roy, K. Yadav, R. Palit,
B. Das, P. Dey, A. Kundu, Md. S. R. Laskar, D. Negi, V. Malik, S. Jadhav, B.
S. Naidu, A. V. Thomas, D. L. Balabanski, A. Dhal, S. Bhattacharya, A. K.
Singh, S. Bhattacharyya, and S. Nag | Competition between the neutron-proton pair break-ups delineating the
level structure of 202Po | null | null | null | null | nucl-ex nucl-th | http://creativecommons.org/licenses/by/4.0/ | High-spin spectroscopic study of $^{202}$Po ($Z$ = 84, $N$ = 118) has been
carried out using the $^{195}$Pt($^{12}$C, 5n)$^{202}$Po fusion-evaporation
reaction. An extended level scheme has been proposed up to an excitation energy
of $E_x\approx$ 8 MeV and angular momentum of 27$\hbar$, with the addition of
57 newly observed $\gamma$-ray transitions, along with the revisions in the
placement of 8 already known transitions and the multipolarities of 4 of these
transitions. The energy of the unobserved 8$^+ \rightarrow 6^+$ transition has
been proposed to be 9.0(5) keV, which resolves the uncertainty in the
excitation energy of the levels above the 6$^{+}$ state. Three new sequences of
$M1$ transitions have also been identified in the high excitation energy regime
and included in the proposed level scheme. The large-scale shell model
calculations for $Z>82$ and $N<126$ valence space have been carried out using
PBPOP interaction which explained the overall level scheme for both the
positive and negative parity states. The calculations successfully reproduced
the purity of the proton $\pi h_{9/2}$ dominated $8^+$ isomeric state, and also
explained the missing $E2$ decay of the ${12}^+$ isomeric state in terms of
changing nucleonic configurations.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:49:57 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Singh",
"Sahab",
""
],
[
"Choudhury",
"D.",
""
],
[
"Maheshwari",
"B.",
""
],
[
"Roy",
"R.",
""
],
[
"Yadav",
"K.",
""
],
[
"Palit",
"R.",
""
],
[
"Das",
"B.",
""
],
[
"Dey",
"P.",
""
],
[
"Kundu",
"A.",
""
],
[
"Laskar",
"Md. S. R.",
""
],
[
"Negi",
"D.",
""
],
[
"Malik",
"V.",
""
],
[
"Jadhav",
"S.",
""
],
[
"Naidu",
"B. S.",
""
],
[
"Thomas",
"A. V.",
""
],
[
"Balabanski",
"D. L.",
""
],
[
"Dhal",
"A.",
""
],
[
"Bhattacharya",
"S.",
""
],
[
"Singh",
"A. K.",
""
],
[
"Bhattacharyya",
"S.",
""
],
[
"Nag",
"S.",
""
]
] |
2505.10238 | Yanbo Ding | Yanbo Ding | MTVCrafter: 4D Motion Tokenization for Open-World Human Image Animation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human image animation has gained increasing attention and developed rapidly
due to its broad applications in digital humans. However, existing methods rely
largely on 2D-rendered pose images for motion guidance, which limits
generalization and discards essential 3D information for open-world animation.
To tackle this problem, we propose MTVCrafter (Motion Tokenization Video
Crafter), the first framework that directly models raw 3D motion sequences
(i.e., 4D motion) for human image animation. Specifically, we introduce 4DMoT
(4D motion tokenizer) to quantize 3D motion sequences into 4D motion tokens.
Compared to 2D-rendered pose images, 4D motion tokens offer more robust
spatio-temporal cues and avoid strict pixel-level alignment between pose image
and character, enabling more flexible and disentangled control. Then, we
introduce MV-DiT (Motion-aware Video DiT). By designing unique motion attention
with 4D positional encodings, MV-DiT can effectively leverage motion tokens as
4D compact yet expressive context for human image animation in the complex 3D
world. Hence, it marks a significant step forward in this field and opens a new
direction for pose-guided human video generation. Experiments show that our
MTVCrafter achieves state-of-the-art results with an FID-VID of 6.98,
surpassing the second-best by 65%. Powered by robust motion tokens, MTVCrafter
also generalizes well to diverse open-world characters (single/multiple,
full/half-body) across various styles and scenarios. Our video demos and code
are provided in the supplementary material and at this anonymous GitHub link:
https://anonymous.4open.science/r/MTVCrafter-1B13.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:50:29 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ding",
"Yanbo",
""
]
] |
2505.10239 | Gokhan Solak | Gokhan Solak, Gustavo J. G. Lahr, Idil Ozdamar, Arash Ajoudani | Context-aware collaborative pushing of heavy objects using
skeleton-based intention prediction | Accepted to be presented at ICRA 2025 conference. Video:
https://youtu.be/qy7l_wGOyzo | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In physical human-robot interaction, force feedback has been the most common
sensing modality to convey the human intention to the robot. It is widely used
in admittance control to allow the human to direct the robot. However, it
cannot be used in scenarios where direct force feedback is not available since
manipulated objects are not always equipped with a force sensor. In this work,
we study one such scenario: the collaborative pushing and pulling of heavy
objects on frictional surfaces, a prevalent task in industrial settings. When
humans do it, they communicate through verbal and non-verbal cues, where body
poses, and movements often convey more than words. We propose a novel
context-aware approach using Directed Graph Neural Networks to analyze
spatio-temporal human posture data to predict human motion intention for
non-verbal collaborative physical manipulation. Our experiments demonstrate
that robot assistance significantly reduces human effort and improves task
efficiency. The results indicate that incorporating posture-based context
recognition, either together with or as an alternative to force sensing,
enhances robot decision-making and control efficiency.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:53:32 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Solak",
"Gokhan",
""
],
[
"Lahr",
"Gustavo J. G.",
""
],
[
"Ozdamar",
"Idil",
""
],
[
"Ajoudani",
"Arash",
""
]
] |
2505.10240 | Jia Tang | Jia Tang and H. Z. Shen | Frequency conversion between optical and microwave photons in
non-Markovian environments | 17 pages, 6 figures | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a scheme for frequency conversion between optical
photons and microwave photons in non-Markovian environments using both magnetic
and mechanical excitations as intermediate media. When the frequencies of
optical photons, magnons, phonons, and microwave photons resonance, the
conversion efficiency can be made close to reach 98.76$\%$ by adjusting the
defined complex cooperativities, while in the case of Markovian, the conversion
efficiency is 90.44$\%$. By controlling the environmental spectral widths, the
efficiency of frequency conversion exhibits a transition from Markovian regimes
to non-Markovian regimes. This transformation simultaneously improves frequency
conversion efficiency and conversion bandwidth, which is due to the excitation
backflow generated by the interaction between the system and the non-Markovian
environments. In the case, when the optical pump power in the non-Markovian
regimes are of a large order of magnitude, the conversion bandwidth can be
increased, but at the cost of reduced conversion efficiency. Our scheme
improves the frequency conversion efficiency and bandwidth between optical
photons and microwave photons, breaking the limitations of frequency conversion
in Markovian environments and providing a new approach for long-distance
quantum communication research of other non-Markovian quantum systems in
quantum optics.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:54:44 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Tang",
"Jia",
""
],
[
"Shen",
"H. Z.",
""
]
] |
2505.10241 | Y. Qiang Sun | Y. Qiang Sun and Pedram Hassanzadeh and Tiffany Shaw and Hamid A.
Pahlavan | Predicting Beyond Training Data via Extrapolation versus Translocation:
AI Weather Models and Dubai's Unprecedented 2024 Rainfall | null | null | null | null | physics.ao-ph | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Artificial intelligence (AI) models have transformed weather forecasting, but
their skill for gray swan extreme events is unclear. Here, we analyze GraphCast
and FuXi forecasts of the unprecedented 2024 Dubai storm, which had twice the
training set's highest rainfall in that region. Remarkably, GraphCast
accurately forecasts this event 8 days ahead. FuXi forecasts the event, but
underestimates the rainfall, especially at long lead times. GraphCast's success
stems from "translocation": learning from comparable/stronger dynamically
similar events in other regions during training via global effective receptive
fields. Evidence of "extrapolation" (learning from training set's weaker
events) is not found. Even events within the global distribution's tail are
poorly forecasted, which is not just due to data imbalance (generalization
error) but also spectral bias (optimization error). These findings demonstrate
the potential of AI models to forecast regional gray swans and opportunity to
improve them through understanding the mechanisms behind their successes and
limitations.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:56:00 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Sun",
"Y. Qiang",
""
],
[
"Hassanzadeh",
"Pedram",
""
],
[
"Shaw",
"Tiffany",
""
],
[
"Pahlavan",
"Hamid A.",
""
]
] |
2505.10242 | Hui Hu | Hui Hu and Xia-Ji Liu | Quantum Lifshitz points in an altermagnetic metal | 6 pages, 4 figures | null | null | null | cond-mat.supr-con cond-mat.quant-gas cond-mat.stat-mech cond-mat.str-el quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We predict the existence of two tri-critical quantum Lifshitz points in
recently discovered $d$-wave altermagnetic metals subjected to an external
magnetic field. These points connect a spatially modulated
Fulde--Ferrell--Larkin--Ovchinnikov (FFLO) phase, a uniform polarized
Bardeen--Cooper--Schrieffer (BCS) superconducting phase, and the normal
metallic phase in a nontrivial manner. Depending on whether the FFLO state is
primarily induced by the magnetic field or by $d$-wave altermagnetism, we
classify the corresponding Lifshitz points as field-driven or
altermagnetism-driven, respectively. Notably, the two types exhibit distinct
behaviors: the transition from the FFLO phase to the polarized BCS phase is
first-order near the field-driven Lifshitz point, as might be expected, whereas
it becomes continuous near the altermagnetism-driven Lifshitz point. We further
explore the effects of finite temperature and find that the
altermagnetism-driven Lifshitz point is significantly more sensitive to thermal
fluctuations.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:56:47 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Hu",
"Hui",
""
],
[
"Liu",
"Xia-Ji",
""
]
] |
2505.10243 | Sanna Normo | S. Normo, D. E. Morosan, P. Zhang, P. Zucca, R. Vainio | Imaging and spectropolarimetric observations of a band-split type II
solar radio burst with LOFAR | null | null | null | null | astro-ph.SR physics.space-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Type II solar radio bursts are generated by electrons accelerated by coronal
shock waves. They appear in dynamic spectra as lanes drifting from higher to
lower frequencies at the plasma frequency and its harmonic. These lanes can
often be split into two or more sub-bands that have similar drift rates. This
phenomenon is called band-splitting, and its physical origins are still under
debate. Our aim is to investigate the origin of band-splitting using novel
imaging and spectropolarimetric observations of a type II solar radio burst
from the Low Frequency Array (LOFAR). We used LOFAR imaging at multiple
frequencies and time steps to track the locations of the radio sources
corresponding to the two components of the band-split emission lane. In
addition, we estimated the degree of circular polarisation (dcp) for both
components using LOFAR's full Stokes dynamic spectra. From the imaging of the
type II burst, we found two close but clearly separated emission regions
clustered over several frequencies spanning each split band. One emission
region corresponds to the lower frequency band and the other to the higher
frequency band of the split lane. Using the full Stokes dynamic spectra, we
also found the dcp to be very similar for both bands. The two distinct emission
regions suggest that the split bands originate from two separate regions at the
shock. The similar values of dcp for both sub-bands correspond to similar
values of magnetic field strength in the two regions and indicate little to no
change in the emission region plasma. Thus, our findings are in contradiction
with previous theories, which have suggested that split bands originate in the
same region but upstream and downstream of the shock. Instead, our results
suggest that both bands originate in two separate upstream regions since we
find a clear separation in locations and no magnetic compression.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:57:46 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Normo",
"S.",
""
],
[
"Morosan",
"D. E.",
""
],
[
"Zhang",
"P.",
""
],
[
"Zucca",
"P.",
""
],
[
"Vainio",
"R.",
""
]
] |
2505.10244 | Jason Li | Jason Li | Simpler and Faster Directed Low-Diameter Decompositions | 9 pages | null | null | null | cs.DS | http://creativecommons.org/licenses/by/4.0/ | We present a simpler and faster algorithm for low-diameter decompositions on
directed graphs, matching the $O(\log m\log\log m)$ loss factor from Bringmann,
Fischer, Haeupler, and Latypov (ICALP 2025) and improving the running time to
$O((m+n\log\log n)\log^2m\log\log m)$.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:58:17 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Li",
"Jason",
""
]
] |
2505.10245 | Florian Wilsch | Ulrich Derenthal and Florian Wilsch | Iitaka fibrations and integral points: a family of arbitrarily polarized
spherical threefolds | 22 pages | null | null | null | math.NT math.AG | http://creativecommons.org/licenses/by/4.0/ | Studying Manin's program for a family of spherical log Fano threefolds, we
determine the asymptotic number of integral points whose height associated with
an arbitrary ample line bundle is bounded. This confirms a recent conjecture by
Santens and sheds new light on the logarithmic analogue of Iitaka fibrations,
which have not yet been adequately formulated.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 12:59:33 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Derenthal",
"Ulrich",
""
],
[
"Wilsch",
"Florian",
""
]
] |
2505.10246 | Kosuke Sakata | Kosuke Sakata, Tsuyoshi Takagi | An Algorithm for the Leading Monomials of a Minimal Groebner Basis of
Generic Sequences | null | null | null | null | cs.SC | http://creativecommons.org/licenses/by/4.0/ | We present an efficient algorithm for computing the leading monomials of a
minimal Groebner basis of a generic sequence of homogeneous polynomials. Our
approach bypasses costly polynomial reductions by exploiting structural
properties conjectured to hold for generic sequences-specifically, that their
leading monomial ideals are weakly reverse lexicographic and that their Hilbert
series follow a known closed-form expression. The algorithm incrementally
constructs the set of leading monomials degree by degree by comparing Hilbert
functions of monomial ideals with the expected Hilbert series of the input
ideal. To enhance computational efficiency, we introduce several optimization
techniques that progressively narrow the search space and reduce the number of
divisibility checks required at each step. We also refine the loop termination
condition using degree bounds, thereby avoiding unnecessary recomputation of
Hilbert series. Experimental results confirm that the proposed method
substantially reduces both computation time and memory usage compared to
conventional Groebner basis computations, particularly for large-scale systems.
These results indicate that our algorithm can serve as an effective
pre-processing tool for accelerating Groebner basis computations in generic
polynomial sequence solving.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:00:44 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Sakata",
"Kosuke",
""
],
[
"Takagi",
"Tsuyoshi",
""
]
] |
2505.10247 | Futoshi Minato | Futoshi Minato, Yifei Niu, Kenichi Yoshida | Correlations of $Q_{\beta}$-values with symmetry energy and effective
mass studied within Skyrme energy--density functionals | 11 pages, 7 figures | null | null | null | nucl-th nucl-ex | http://creativecommons.org/licenses/by/4.0/ | The $\beta$-decay half-lives of nuclei are sensitive to the values of
$Q_{\beta}$. For accurate theoretical predictions, it is essential to develop
an effective interaction or an energy density functional (EDF) that can
systematically reproduce experimental $Q_{\beta}$ values. The challenge lies in
identifying an appropriate EDF for an accurate $Q_{\beta}$ prediction. To
address this, we focus on the bulk properties of nuclei that have correlations
with $Q_{\beta}$. The primary objective of this study is to determine which
nuclear bulk properties are sensitive to $Q_{\beta}$, providing information on
the key nuclear characteristics that influence $\beta$-decay calculations. We
employ the Skyrme energy-density functionals to find correlations between
$Q_{\beta}$ and the nuclear bulk properties, assuming spherical symmetry. Using
$42$ different Skyrme EDFs, we analyze these correlations by evaluating Pearson
linear coefficients, focusing particularly on the relationship between
$Q_{\beta}$ and various nuclear properties. We found that the symmetry energy
at low densities shows a correlation with the $Q_{\beta}$ value. In particular,
this correlation becomes stronger for functionals with an effective mass close
to $1$. However, as the nuclear density increases, the correlation weakens.
From our analysis, we found that a symmetry energy of $32.8\pm0.7$~MeV and
effective mass of $m^{*}/m\ge0.75$ at the saturation density is the most likely
to systematically reproduce the experimental data of $Q_{\beta}$
systematically.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:02:27 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Minato",
"Futoshi",
""
],
[
"Niu",
"Yifei",
""
],
[
"Yoshida",
"Kenichi",
""
]
] |
2505.10248 | Arun Ashok | S. Y. Neyaz, A. Ashok, M. Schiek, C. Grewing, A. Zambanini, S. van
Waasen | Scalable 28nm IC implementation of coupled oscillator network featuring
tunable topology and complexity | null | null | null | null | cs.ET cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Integrated circuit implementations of coupled oscillator networks have
recently gained increased attention. The focus is usually on using these
networks for analogue computing, for example for solving computational
optimization tasks. For use within analog computing, these networks are run
close to critical dynamics. On the other hand, such networks are also used as
an analogy of transport networks such as electrical power grids to answer the
question of how exactly such critical dynamic states can be avoided. However,
simulating large network of coupled oscillators is computationally intensive,
with specifc regards to electronic ones. We have developed an integrated
circuit using integrated Phase-Locked Loop (PLL) with modifications, that
allows to flexibly vary the topology as well as a complexity parameter of the
network during operation. The proposed architecture, inspired by the brain,
employs a clustered architecture, with each cluster containing 7 PLLs featuring
programmable coupling mechanisms. Additionally, the inclusion of a RISC-V
processor enables future algorithmic implementations. Thus, we provide a
practical alternative for large-scale network simulations both in the field of
analog computing and transport network stability research.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:03:49 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Neyaz",
"S. Y.",
""
],
[
"Ashok",
"A.",
""
],
[
"Schiek",
"M.",
""
],
[
"Grewing",
"C.",
""
],
[
"Zambanini",
"A.",
""
],
[
"van Waasen",
"S.",
""
]
] |
2505.10249 | Daniela T\"auber | Maryam Ali, Robin Schneider, Anika Strecker, Nila Krishnakumar,
Sebastian Unger, Mohammad Soltaninezhad, Johanna Kirchhoff, Astrid Tannert,
Katerina A. Dragounova, Rainer Heintzmann, Anne-Dorothea M\"uller, Christoph
Krafft, Ute Neugebauer and Daniela T\"auber | Subcellular nano-chemical characterization in photothermal spectroscopic
imaging of antimicrobial interaction in model system Bacillus subtilis &
vancomycin | 5 figures in main text, 3 figures in supporting information | null | null | null | physics.bio-ph cond-mat.soft | http://creativecommons.org/licenses/by/4.0/ | The power of photothermal spectroscopic imaging to visualize antimicrobial
interaction on the surface of individual bacteria cells has been demonstrated
on the model system Bacillus subtilis (B. subtilis) and vancomycin using
mid-infrared photo-induced force microscopy (PiF-IR, also mid-IR PiFM).
High-resolution PiF contrasts obtained by merging subsequent PiF-IR scans at
two different illumination frequencies revealed chemical details of cell wall
destruction after 30 and 60 min incubation with vancomycin with a spatial
resolution of $\approx 5$ nm. This approach compensates local intensity
variations induced by near-field coupling of the illuminating electric field
with nanostructured surfaces, which appear in single-frequency contrasts in
photothermal imaging methods, as shown by [Anindo et al., J. Phys. Chem C,
2025, 129, 4517]. Spectral shifts associated with hydrogen bond formation
between vancomycin and the N-acyl-D-Ala4-D-Ala5 termini in the peptidoglycan
cell wall have been observed in chemometrics of PiF-IR spectra from treated and
untreated B. subtilis harvested after 30 min from the same experiment. The
vancomyin interaction in the piecrust of a progressing septum is located with
$\approx 10$ nm resolution exemplarily using PiF contrasts of three selected
bands of a PiF-IR hyperspectral scan of an individual B. subtilis cell
harvested after 30 min incubation. Our results provide a new perspective for
visualizing the chemical interaction of antibiotics on the surface of microbes
with few nanometer resolution.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:03:57 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Ali",
"Maryam",
""
],
[
"Schneider",
"Robin",
""
],
[
"Strecker",
"Anika",
""
],
[
"Krishnakumar",
"Nila",
""
],
[
"Unger",
"Sebastian",
""
],
[
"Soltaninezhad",
"Mohammad",
""
],
[
"Kirchhoff",
"Johanna",
""
],
[
"Tannert",
"Astrid",
""
],
[
"Dragounova",
"Katerina A.",
""
],
[
"Heintzmann",
"Rainer",
""
],
[
"Müller",
"Anne-Dorothea",
""
],
[
"Krafft",
"Christoph",
""
],
[
"Neugebauer",
"Ute",
""
],
[
"Täuber",
"Daniela",
""
]
] |
2505.10250 | Wenhao Shen | Wenhao Shen, Wanqi Yin, Xiaofeng Yang, Cheng Chen, Chaoyue Song,
Zhongang Cai, Lei Yang, Hao Wang, Guosheng Lin | ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct
Preference Optimization | Accepted by ICML 2025. Code: https://github.com/shenwenhao01/ADHMR | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human mesh recovery (HMR) from a single image is inherently ill-posed due to
depth ambiguity and occlusions. Probabilistic methods have tried to solve this
by generating numerous plausible 3D human mesh predictions, but they often
exhibit misalignment with 2D image observations and weak robustness to
in-the-wild images. To address these issues, we propose ADHMR, a framework that
Aligns a Diffusion-based HMR model in a preference optimization manner. First,
we train a human mesh prediction assessment model, HMR-Scorer, capable of
evaluating predictions even for in-the-wild images without 3D annotations. We
then use HMR-Scorer to create a preference dataset, where each input image has
a pair of winner and loser mesh predictions. This dataset is used to finetune
the base model using direct preference optimization. Moreover, HMR-Scorer also
helps improve existing HMR models by data cleaning, even with fewer training
samples. Extensive experiments show that ADHMR outperforms current
state-of-the-art methods. Code is available at:
https://github.com/shenwenhao01/ADHMR.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:04:51 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Shen",
"Wenhao",
""
],
[
"Yin",
"Wanqi",
""
],
[
"Yang",
"Xiaofeng",
""
],
[
"Chen",
"Cheng",
""
],
[
"Song",
"Chaoyue",
""
],
[
"Cai",
"Zhongang",
""
],
[
"Yang",
"Lei",
""
],
[
"Wang",
"Hao",
""
],
[
"Lin",
"Guosheng",
""
]
] |
2505.10251 | Ji Woong Kim | Ji Woong Kim, Juo-Tung Chen, Pascal Hansen, Lucy X. Shi, Antony
Goldenberg, Samuel Schmidgall, Paul Maria Scheikl, Anton Deguet, Brandon M.
White, De Ru Tsai, Richard Cha, Jeffrey Jopling, Chelsea Finn, Axel Krieger | SRT-H: A Hierarchical Framework for Autonomous Surgery via Language
Conditioned Imitation Learning | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Research on autonomous robotic surgery has largely focused on simple task
automation in controlled environments. However, real-world surgical
applications require dexterous manipulation over extended time scales while
demanding generalization across diverse variations in human tissue. These
challenges remain difficult to address using existing logic-based or
conventional end-to-end learning strategies. To bridge this gap, we propose a
hierarchical framework for dexterous, long-horizon surgical tasks. Our method
employs a high-level policy for task planning and a low-level policy for
generating task-space controls for the surgical robot. The high-level planner
plans tasks using language, producing task-specific or corrective instructions
that guide the robot at a coarse level. Leveraging language as a planning
modality offers an intuitive and generalizable interface, mirroring how
experienced surgeons instruct traineers during procedures. We validate our
framework in ex-vivo experiments on a complex minimally invasive procedure,
cholecystectomy, and conduct ablative studies to assess key design choices. Our
approach achieves a 100% success rate across n=8 different ex-vivo
gallbladders, operating fully autonomously without human intervention. The
hierarchical approach greatly improves the policy's ability to recover from
suboptimal states that are inevitable in the highly dynamic environment of
realistic surgical applications. This work represents the first demonstration
of step-level autonomy, marking a critical milestone toward autonomous surgical
systems for clinical studies. By advancing generalizable autonomy in surgical
robotics, our approach brings the field closer to real-world deployment.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:04:53 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Kim",
"Ji Woong",
""
],
[
"Chen",
"Juo-Tung",
""
],
[
"Hansen",
"Pascal",
""
],
[
"Shi",
"Lucy X.",
""
],
[
"Goldenberg",
"Antony",
""
],
[
"Schmidgall",
"Samuel",
""
],
[
"Scheikl",
"Paul Maria",
""
],
[
"Deguet",
"Anton",
""
],
[
"White",
"Brandon M.",
""
],
[
"Tsai",
"De Ru",
""
],
[
"Cha",
"Richard",
""
],
[
"Jopling",
"Jeffrey",
""
],
[
"Finn",
"Chelsea",
""
],
[
"Krieger",
"Axel",
""
]
] |
2505.10252 | Zheng Wei | Zheng Wei, Ji-Yao Fan, Kui Cao, Xin-Ran Ma, Su-Peng Kou | Generalized Non-Hermitian Skin Effect | null | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this Letter, we present a unified theory termed the generalized
non-Hermitian skin effect. This framework provides a universal characterization
of typical one-dimensional non-Hermitian skin effects within the perturbative
regime and unveils a novel type of skin effect that beyond the predictions of
the generalized Brillouin zone theory, referred to as the relative skin effect.
Previously recognized skin effects are classified as global skin effects,
thereby explicitly delineating the scope and limitations of existing non-Bloch
band theories. Additionally, we establish, for the first time, a phase
transition criterion between global skin effects and relative skin effect,
demonstrating the competition between these two distinct types of skin effects,
emphasizing the pivotal role of real-space non-Hermitian terms in understanding
skin effects and challenging the traditional reliance of non-Bloch band theory
on momentum space. Our study substantially advances the conceptual framework of
non-Hermitian physics and provides new theoretical tools for investigation.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:05:05 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Wei",
"Zheng",
""
],
[
"Fan",
"Ji-Yao",
""
],
[
"Cao",
"Kui",
""
],
[
"Ma",
"Xin-Ran",
""
],
[
"Kou",
"Su-Peng",
""
]
] |
2505.10253 | Ranjith Unnithan Dr | Alan Lee, Dechuan Sun, Gregory Tanyi, Younger Liang, Christina Lim,
Ranjith R Unnithan | Flexible-AR display for near-eye operations | null | null | null | null | physics.optics cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a new technique to fabricate flexible-near-field Argument-Reality
(AR) display using modular-molds. A near-eye flexible-AR-display is fabricated
based on parameters extracted from simulations. Our AR-display successfully
reconstructed images and videos from a light-engine. It opens a new approach to
fabricate flexible-near-field AR display with good physical stress and
collision-resilience.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:07:17 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Lee",
"Alan",
""
],
[
"Sun",
"Dechuan",
""
],
[
"Tanyi",
"Gregory",
""
],
[
"Liang",
"Younger",
""
],
[
"Lim",
"Christina",
""
],
[
"Unnithan",
"Ranjith R",
""
]
] |
2505.10254 | Nicolas Pr\"ollochs | Michelle Bobek, Nicolas Pr\"ollochs | Community Fact-Checks Do Not Break Follower Loyalty | null | null | null | null | cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Major social media platforms increasingly adopt community-based fact-checking
to address misinformation on their platforms. While previous research has
largely focused on its effect on engagement (e.g., reposts, likes), an
understanding of how fact-checking affects a user's follower base is missing.
In this study, we employ quasi-experimental methods to causally assess whether
users lose followers after their posts are corrected via community fact-checks.
Based on time-series data on follower counts for N=3516 community fact-checked
posts from X, we find that community fact-checks do not lead to meaningful
declines in the follower counts of users who post misleading content. This
suggests that followers of spreaders of misleading posts tend to remain loyal
and do not view community fact-checks as a sufficient reason to disengage. Our
findings underscore the need for complementary interventions to more
effectively disincentivize the production of misinformation on social media.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:08:11 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Bobek",
"Michelle",
""
],
[
"Pröllochs",
"Nicolas",
""
]
] |
2505.10255 | H. Z. Shen | J. X. Yang, Cheng Shang, Yan-Hui Zhou, and H. Z. Shen | Simultaneous nonreciprocal unconventional photon blockade via two
degenerate optical parametric amplifiers in spinning resonators | 12 pages, 13 figures | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a scheme to achieving simultaneous nonreciprocal unconventional
photon blockade in a system of two coupled spining resonators marked by modes a
and b, each incorporating an degenerate optical parametric amplifier (DOPA). By
rotating the resonators, input light from opposite directions induces opposite
Sagnac-Fizeau shifts. These shifts result in the emergence or absence of
quantum destructive interference in two-photon excitation processes.
Specifically, when destructive quantum interference occurs, photons from one
input direction are simultaneously blocked in both resonators, whereas the
absence of complete destructive quantum interference causes the blockade effect
to vanish for inputs from the opposite direction. We analytically give the
optimal parameter conditions to achieve simultaneous strong photon blockade
with the parametric amplification. By adjusting the Sagnac-Fizeau shifts, we
can make mode a nonreciprocal photon blockade, while mode b exhibits photon
blockade in both directions. This work lays a theoretical foundation for the
development of multimode simultaneous nonreciprocal unconventional
single-photon devices, which hold promising potential in multichannel
topological optics and chiral quantum technologies.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:08:14 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Yang",
"J. X.",
""
],
[
"Shang",
"Cheng",
""
],
[
"Zhou",
"Yan-Hui",
""
],
[
"Shen",
"H. Z.",
""
]
] |
2505.10256 | Kohei Hayashi | Patr\'icia Gon\c{c}alves, Kohei Hayashi, Jo\~ao Pedro Mangi | Stochastic oscillators out of equilibrium: scaling limits and
correlation estimates | 39 pages | null | null | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider a purely harmonic chain of oscillators which is perturbed by a
stochastic noise. Under this perturbation, the system exhibits two conserved
quantities: the volume and the energy. At the level of the hydrodynamic limit,
under diffusive scaling, we show that depending on the strength of the
Hamiltonian dynamics, energy and volume evolve according to either a system of
autonomous heat equations or a non-linear system of coupled parabolic
equations. Moreover, for general initial measures, under diffusive scaling, we
can characterize the non-equilibrium volume fluctuations. The proofs are based
on precise bounds on the two-point volume correlation function and a uniform
fourth-moment estimate.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:08:36 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Gonçalves",
"Patrícia",
""
],
[
"Hayashi",
"Kohei",
""
],
[
"Mangi",
"João Pedro",
""
]
] |
2505.10257 | Hao Lu | Hao Lu, Jiaqi Tang, Jiyao Wang, Yunfan LU, Xu Cao, Qingyong Hu, Yin
Wang, Yuting Zhang, Tianxin Xie, Yunpeng Zhang, Yong Chen, Jiayu.Gao, Bin
Huang, Dengbo He, Shuiguang Deng, Hao Chen, Ying-Cong Chen | Sage Deer: A Super-Aligned Driving Generalist Is Your Copilot | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The intelligent driving cockpit, an important part of intelligent driving,
needs to match different users' comfort, interaction, and safety needs. This
paper aims to build a Super-Aligned and GEneralist DRiving agent, SAGE DeeR.
Sage Deer achieves three highlights: (1) Super alignment: It achieves different
reactions according to different people's preferences and biases. (2)
Generalist: It can understand the multi-view and multi-mode inputs to reason
the user's physiological indicators, facial emotions, hand movements, body
movements, driving scenarios, and behavioral decisions. (3) Self-Eliciting: It
can elicit implicit thought chains in the language space to further increase
generalist and super-aligned abilities. Besides, we collected multiple data
sets and built a large-scale benchmark. This benchmark measures the deer's
perceptual decision-making ability and the super alignment's accuracy.
| [
{
"version": "v1",
"created": "Thu, 15 May 2025 13:08:44 GMT"
}
] | 2025-05-16T00:00:00 | [
[
"Lu",
"Hao",
""
],
[
"Tang",
"Jiaqi",
""
],
[
"Wang",
"Jiyao",
""
],
[
"LU",
"Yunfan",
""
],
[
"Cao",
"Xu",
""
],
[
"Hu",
"Qingyong",
""
],
[
"Wang",
"Yin",
""
],
[
"Zhang",
"Yuting",
""
],
[
"Xie",
"Tianxin",
""
],
[
"Zhang",
"Yunpeng",
""
],
[
"Chen",
"Yong",
""
],
[
"Gao",
"Jiayu.",
""
],
[
"Huang",
"Bin",
""
],
[
"He",
"Dengbo",
""
],
[
"Deng",
"Shuiguang",
""
],
[
"Chen",
"Hao",
""
],
[
"Chen",
"Ying-Cong",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.