id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.19794 | Eduardo Hafemann | Eduardo Hafemann and Eleni-Alexandra Kontou | Penrose inequality for integral energy conditions | null | null | null | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The classical Penrose inequality, a relation between the ADM mass and the
area of any cross section of the black hole event horizon, was introduced as a
test of the weak cosmic censorship: if it fails, the trapped surface is not
necessarily behind the event horizon and a naked singularity could form. Since
that original derivation, a variety of proofs have developed, mainly focused on
the initial data formulation on maximal spacelike slices of spacetime. Most of
these proofs are applicable only for classical fields, as the energy conditions
required are violated in the context of quantum field theory. In this work we
provide two generalizations of the Penrose inequality spherically symmetric
spacetimes: a proof of a classical inequality using initial data and an average
energy condition and proof of a generalized inequality for evaporating black
holes with a connection to the weak cosmic censorship. The latter case could
also be applicable to quantum fields. Finally, we provide physically motivated
examples for both.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:37:05 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Hafemann",
"Eduardo",
""
],
[
"Kontou",
"Eleni-Alexandra",
""
]
] |
2504.19795 | Fu-Jiun Jiang | Shang-Wei Li, Yuan-Heng Tseng, Ming-Che Hsieh, and Fu-Jiun Jiang | Comments on the minimal training set for CNN: a case study of the
frustrated $J_1$-$J_2$ Ising model on the square lattice | 46 figures, 16 pages | null | null | null | hep-lat cond-mat.stat-mech | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The minimal training set to train a working CNN is explored in detail. The
considered model is the frustrated $J_1$-$J_2$ Ising model on the square
lattice. Here $J_1 < 0$ and $J_2 > 0$ are the nearest and next-to-nearest
neighboring couplings, respectively. We train the CNN using the configurations
of $g \stackrel{\text{def}}{=} J_2/|J_1| = 0.7$ and employ the resulting CNN to
study the phase transition of $g = 0.8$. We find that this transfer learning is
successful. In particular, only configurations of two temperatures, one is
below and one is above the critical temperature $T_c$ of $g=0.7$, are needed to
reach accurately determination of the $T_c$ of $g=0.8$. However, it may be
subtle to use this strategy for the training. Specifically, for the considered
model, due to the inefficiency of the single spin flip algorithm used in
sampling the configurations at the low-temperature region, the two temperatures
associated with the training set should not be too far away from the $T_c$ of
$g=0.7$, otherwise, the performance of the obtained CNN is not of high quality,
hence cannot determine the $T_c$ of $g=0.8$ accurately. For the considered
model, we also uncover the condition for training a successful CNN when only
configurations of two temperatures are considered as the training set.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:38:10 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Li",
"Shang-Wei",
""
],
[
"Tseng",
"Yuan-Heng",
""
],
[
"Hsieh",
"Ming-Che",
""
],
[
"Jiang",
"Fu-Jiun",
""
]
] |
2504.19796 | Yankai Lin | Yankai Lin, Michelle S. Chong, Carlos Murguia | Modified Control Barrier Function for Quadratic Program Based Control
Design via Sum-of-Squares Programming | null | null | null | null | math.OC cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | We consider a nonlinear control affine system controlled by inputs generated
by a quadratic program (QP) induced by a control barrier functions (CBF).
Specifically, we slightly modify the condition satisfied by CBFs and study how
the modification can positively impact the closed loop behavior of the system.
We show that, QP-based controllers designed using the modified CBF condition
preserves the desired properties of QP-based controllers using standard CBF
conditions. Furthermore, using the generalized S-procedure for polynomial
functions, we formulate the design of the modified CBFs as a Sum-Of-Squares
(SOS) program, which can be solved efficiently. Via a numerical example, the
proposed CBF design is shown to have superior performance over the standard CBF
widely used in existing literature.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:38:17 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Lin",
"Yankai",
""
],
[
"Chong",
"Michelle S.",
""
],
[
"Murguia",
"Carlos",
""
]
] |
2504.19797 | Tousif Rahman | Gang Mao, Tousif Rahman, Sidharth Maheshwari, Bob Pattison, Zhuang
Shao, Rishad Shafik, Alex Yakovlev | Dynamic Tsetlin Machine Accelerators for On-Chip Training at the Edge
using FPGAs | null | null | null | null | cs.AR cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The increased demand for data privacy and security in machine learning (ML)
applications has put impetus on effective edge training on Internet-of-Things
(IoT) nodes. Edge training aims to leverage speed, energy efficiency and
adaptability within the resource constraints of the nodes. Deploying and
training Deep Neural Networks (DNNs)-based models at the edge, although
accurate, posit significant challenges from the back-propagation algorithm's
complexity, bit precision trade-offs, and heterogeneity of DNN layers. This
paper presents a Dynamic Tsetlin Machine (DTM) training accelerator as an
alternative to DNN implementations. DTM utilizes logic-based on-chip inference
with finite-state automata-driven learning within the same Field Programmable
Gate Array (FPGA) package. Underpinned on the Vanilla and Coalesced Tsetlin
Machine algorithms, the dynamic aspect of the accelerator design allows for a
run-time reconfiguration targeting different datasets, model architectures, and
model sizes without resynthesis. This makes the DTM suitable for targeting
multivariate sensor-based edge tasks. Compared to DNNs, DTM trains with fewer
multiply-accumulates, devoid of derivative computation. It is a data-centric ML
algorithm that learns by aligning Tsetlin automata with input data to form
logical propositions enabling efficient Look-up-Table (LUT) mapping and frugal
Block RAM usage in FPGA training implementations. The proposed accelerator
offers 2.54x more Giga operations per second per Watt (GOP/s per W) and uses 6x
less power than the next-best comparable design.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:38:53 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Mao",
"Gang",
""
],
[
"Rahman",
"Tousif",
""
],
[
"Maheshwari",
"Sidharth",
""
],
[
"Pattison",
"Bob",
""
],
[
"Shao",
"Zhuang",
""
],
[
"Shafik",
"Rishad",
""
],
[
"Yakovlev",
"Alex",
""
]
] |
2504.19798 | Nikita Medvedev | Nikita Medvedev | Stainless steel in an electronically excited state | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Austenitic stainless steel
(Fe$_{0.5875}$Cr$_{0.25}$Mn$_{0.09}$Ni$_{0.07}$C$_{0.0025}$) is modeled with
the XTANT-3 simulation toolkit to evaluate its properties in an electronically
excited state produced by ultrafast laser irradiation. The model relies on
density-functional tight binding to obtain the electronic Hamiltonian,
producing the electronic energy levels and wave functions. They are used to
calculate the electron heat capacity, electron heat conductivity, and
electron-phonon coupling parameter at the electronic temperatures up to 25,000
K. It is revealed that stainless steel nonthermally melts at sub-picosecond
timescales at the deposited dose of ~1.4 eV/atom (the electronic temperature
$T_e \sim 10,000$ K): the atomic lattice disorders due to modifications of the
interatomic potential induced by the excitation of electrons, even without
electron-phonon coupling (Born-Oppenheimer approximation). Including the
electron-phonon coupling (non-Born-Oppenheimer simulation) lowers the damage
threshold dose to ~0.45 eV/atom, triggering atomic disorder at picosecond
timescales via thermal melting, predicated on the atomic heating by electrons.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:40:21 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Medvedev",
"Nikita",
""
]
] |
2504.19799 | Mathias Samuelsson | Mathias Samuelsson and Thomas G. Blackburn | Collision parameters needed to measure polarisation-dependent nonlinear
pair creation | 9 pages, 9 figures | null | null | null | hep-ph physics.plasm-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pair creation by a $\gamma$ ray in a high-intensity electromagnetic field
(the nonlinear Breit-Wheeler process) is sensitive to the $\gamma$-ray
polarisation. Here we study the stability required in order to measure this
polarisation dependence in a head-on collision between a high-power linearly
polarised laser and a relativistic electron beam, where the $\gamma$-rays
produced via nonlinear Compton scattering are highly polarised. We find that
the laser strength parameter $a_0$ has to be known to within a few per cent,
and that the alignment of the laser and electron beam cannot fluctuate by more
than half the spot size, in order to attribute the reduction in positron yield
to polarisation effects. These collision parameters are difficult to achieve,
but may be realised in future precision experiments that focus on beam
stability.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:42:55 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Samuelsson",
"Mathias",
""
],
[
"Blackburn",
"Thomas G.",
""
]
] |
2504.19800 | Jiaqi Liu | Gong Chen, Jiaqi Liu, Yuanhong Tian | Long time asymptotics of a perturbed modified KdV equation | null | null | null | null | math.AP | http://creativecommons.org/licenses/by/4.0/ | We derive full asymptotics of the modified KdV equation (mKdV) with a
higher-order perturbative term. We make use of the perturbative theory of
infinite-dimensional integrable systems developed by P. Deift and X. Zhou
\cite{DZ-2}, and some new and simpler proofs of certain $L^\infty$ bounds and
$L^p$ a priori estimates developed recently in \cite{CLT}. We show that the
perturbed equation exhibits the same long-time behavior as the completely
integrable mKdV.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:44:43 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Chen",
"Gong",
""
],
[
"Liu",
"Jiaqi",
""
],
[
"Tian",
"Yuanhong",
""
]
] |
2504.19801 | Pani W. Fernando | Osanda Chinthila, Pani W. Fernando, Anuradha Mahasinghe, Kaushika De
Silva and Sarath Kumara | Stochastic quantum adiabatic algorithm with fractional Brownian motion | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adiabatic Quantum Computing relies on the quantum adiabatic theorem, which
states that a quantum system evolves along its ground state with time if the
governing Hamiltonian varies infinitely slowly. However, practical limitations
force computations to be performed within limited times, exposing the system to
transitions into excited states, and thereby reducing the success probability.
Here we investigate the counterintuitive hypothesis that incorporating
stochastic noise, specifically noise driven by fractional Brownian motion, in a
non-Markovian setup can enhance the performance of adiabatic quantum computing
by improving its success probability at limited evolution times. The study
begins by developing the mathematical framework to introduce stochastic noise
multiplicatively into the Schr\"{o}dinger equation, resulting in a stochastic
Schr\"{o}dinger equation. To preserve It\^{o} integrability within the
non-Markovian framework, a semimartingale approximation for fractional Brownian
motion is employed. We perform numerical simulations to compare the performance
of the quantum adiabatic algorithm with and without noise driven by fractional
Brownian motion using the NP-complete Exact Cover-3 problem, transformed into
the Ising model. Our results exhibit an improvement in success probability in
the presence of noise driven by fractional Brownian motion with Hurst parameter
$0<H<\frac{1}{2}$ and an increase in speedup as $H$ approaches 0. Although
simulations are limited to problems involving a modest number of qubits,
evidence suggests that the proposed approach scales favorably with the system
size.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:51:16 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Chinthila",
"Osanda",
""
],
[
"Fernando",
"Pani W.",
""
],
[
"Mahasinghe",
"Anuradha",
""
],
[
"De Silva",
"Kaushika",
""
],
[
"Kumara",
"Sarath",
""
]
] |
2504.19802 | Maurits Haverkort | Michelangelo Tagliavini, Federico Mazza, Xinlin Yan, Andrey Sidorenko,
Kevin Ackermann, Andrey Prokofiev, Kurt Kummer, Silke Paschen, and Maurits W.
Haverkort | Crystal field excitations in rattling clathrate CeBa$_7$Au$_6$Si$_{40}$ | null | null | null | null | cond-mat.str-el | http://creativecommons.org/licenses/by/4.0/ | We investigate the local crystal-field environment of cerium in the clathrate
compound CeBa$_7$Au$_6$Si$_{40}$ (Ce-BAS) using resonant inelastic x-ray
scattering (RIXS) and magnetic susceptibility measurements. Ce-BAS is a rare
example of a system where heavy-fermion physics coexists with low-energy
rattling phonon modes, making it a candidate for enhanced thermoelectric
performance. Magnetic susceptibility measurements reveal a
temperature-dependent local moment that cannot be explained within a static
crystal-field model. A fit to the susceptibility data requires strong mixing
between the j = 5/2 and j = 7/2 crystal-field states, which implies large
internal splittings inconsistent with RIXS spectra. In contrast, RIXS data are
well described by a model with negligible j = 5/2 - j = 7/2 mixing and an
energy separation of 12 meV between the ground and first excited crystal-field
states. The inability to reconcile these two datasets within a static framework
points to a dynamical modification of the crystal-field potential. We attribute
this to coupling between the Ce 4f electrons and low-energy phonons associated
with the Ce rattling motion. This interpretation is consistent with theoretical
predictions of phonon-enhanced Kondo effects and dynamical Jahn-Teller
distortions. Our results highlight the need for multi-orbital impurity models
that include phonon coupling to fully describe the low-energy physics of
Ce-BAS.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:51:32 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Tagliavini",
"Michelangelo",
""
],
[
"Mazza",
"Federico",
""
],
[
"Yan",
"Xinlin",
""
],
[
"Sidorenko",
"Andrey",
""
],
[
"Ackermann",
"Kevin",
""
],
[
"Prokofiev",
"Andrey",
""
],
[
"Kummer",
"Kurt",
""
],
[
"Paschen",
"Silke",
""
],
[
"Haverkort",
"Maurits W.",
""
]
] |
2504.19803 | Satyabrata Patnaik | P. Saha, M. Singh, P. D. Babu and S. Patnaik | Unconventional Relaxation Dynamics in Co_8Zn_7Mn_5 and Co_8Zn_8Mn_4:
Evidence of Inertial Effects | null | null | null | null | cond-mat.mtrl-sci cond-mat.other cond-mat.str-el quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Magnetization relaxation dynamics serve as an essential tool for uncovering
the intrinsic mechanisms governing the magnetic response and energy dissipation
in magnetic systems. In this work, we examine the relaxation dynamics for Beta
Mn type Co_8Zn_7Mn_5 and Co_8Zn_8Mn_4 across a frequency range of 1 kHz to 10
kHz, spanning different magnetic phases. While most magnetic systems tend to
follow the Debye-like relaxation with non-zero distribution or the Cole-Cole
formalism, our analysis reveal that these conventional models fail to capture
frequency dependence of ac susceptibility across different magnetic phases in
Co_8Zn_7Mn_5 and Co_8Zn_8Mn_4. Instead, an inertial component is needed to
successfully describe the dynamics, suggesting the presence of unconventional
relaxation behavior. The characteristic relaxation time is found to be of the
order of 10^-5 s for both the compositions. The field dependent variation of
relaxation time exhibits a non-monotonic nature, with the double peak like
structure at the skyrmion phase transitions, implying slower relaxation
dynamics at the phase boundaries. Furthermore, the presence of non-zero
difference between isothermal and adiabatic susceptibility in the pure phases
implies slower relaxation dynamics, which is consistent with the presence of
finite dissipation in pure phases. The inertial term has been previously
invoked to describe the dynamics in spin ice systems due to the propagation of
magnetic monopoles. However, its necessity in this system, points to a wider
significance in magnetization dynamics that goes beyond the conventional spin
ices and skyrmions.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:55:27 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Saha",
"P.",
""
],
[
"Singh",
"M.",
""
],
[
"Babu",
"P. D.",
""
],
[
"Patnaik",
"S.",
""
]
] |
2504.19804 | Katharina Noesges | Katharina Noesges and Thomas Mussenbrock | Nonlinear Power Absorption in CCRF Discharges: Transition from Symmetric
to Asymmetric Configurations | null | null | null | null | physics.plasm-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work builds upon previous studies of nonlinear dynamics in low-pressure
capacitively coupled radio-frequency discharges, focusing on the electron power
absorption mechanism in discharges with various geometric asymmetries. We
present a comprehensive investigation using a fully kinetic electrostatic 1d3v
Particle-in-Cell/Monte Carlo collision simulation in spherical geometry. By
systematically varying the inner electrode radius and the electrode gap
distance, we analyze the influence of geometric asymmetry on key plasma
properties, including electron density, power absorption, electron dynamics,
and current characteristics. A central focus is placed on the cumulative power
density as a diagnostic for energy deposition. In strongly asymmetric
configurations, the cumulative electron power density exhibits distinct
stepwise increases during sheath expansion, corresponding to the acceleration
of successive electron beams. These nonlinear signatures are directly linked to
the excitation of plasma series resonance and enhanced beam-driven power
absorption. In contrast, more symmetric configurations display smoother, more
symmetric cumulative power evolution, indicating balanced energy transfer at
both sheaths and reduced nonlinearities. Time- and space-resolved diagnostics
of cumulative power, current waveforms, and densities of energetic electrons
reveal the critical role of asymmetry in shaping electron confinement and
beam-driven power absorption. These findings demonstrate that the discharge
geometry is actually an important design parameter which needs to taken into
account during the design and construction phase of a reactor as it directly
influences the plasma behavior with respect to energy deposition.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:55:43 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Noesges",
"Katharina",
""
],
[
"Mussenbrock",
"Thomas",
""
]
] |
2504.19805 | Hongbin Sun | Hongbin Sun, Zhongzi Wang | On virtual chirality of 3-manifolds | 14 pages, no figures. Comments are welcome! | null | null | null | math.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We prove that if a prime 3-manifold M is not finitely covered by the 3-sphere
or a product manifold, then M is virtually chiral, i.e. it has a finite cover
that does not admit an orientation reversing self-homeomorphism. In general if
a 3-manifold contains a virtually chiral prime summand, then it is virtually
chiral.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:56:01 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sun",
"Hongbin",
""
],
[
"Wang",
"Zhongzi",
""
]
] |
2504.19806 | Zhilin Lu | Zhilin Lu, Rongpeng Li, Zhifeng Zhao, and Honggang Zhang | Reinforcement Learning-Based Heterogeneous Multi-Task Optimization in
Semantic Broadcast Communications | null | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Semantic broadcast communications (Semantic BC) for image transmission have
achieved significant performance gains for single-task scenarios. Nevertheless,
extending these methods to multi-task scenarios remains challenging, as
different tasks typically require distinct objective functions, leading to
potential conflicts within the shared encoder. In this paper, we propose a
tri-level reinforcement learning (RL)-based multi-task Semantic BC framework,
termed SemanticBC-TriRL, which effectively resolves such conflicts and enables
the simultaneous support of multiple downstream tasks at the receiver side,
including image classification and content reconstruction tasks. Specifically,
the proposed framework employs a bottom-up tri-level alternating learning
strategy, formulated as a constrained multi-objective optimization problem. At
the first level, task-specific decoders are locally optimized using supervised
learning. At the second level, the shared encoder is updated via proximal
policy optimization (PPO), guided by task-oriented rewards. At the third level,
a multi-gradient aggregation-based task weighting module adaptively adjusts
task priorities and steers the encoder optimization. Through this hierarchical
learning process, the encoder and decoders are alternately trained, and the
three levels are cohesively integrated via constrained learning objective.
Besides, the convergence of SemanticBC-TriRL is also theoretically established.
Extensive simulation results demonstrate the superior performance of the
proposed framework across diverse channel conditions, particularly in low SNR
regimes, and confirm its scalability with increasing numbers of receivers.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:59:01 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Lu",
"Zhilin",
""
],
[
"Li",
"Rongpeng",
""
],
[
"Zhao",
"Zhifeng",
""
],
[
"Zhang",
"Honggang",
""
]
] |
2504.19808 | Mauricio Garay | Mauricio Garay and Duco van Straten | Fixed point theorems for small divisors problems | null | null | null | null | math.DS math.FA | http://creativecommons.org/licenses/by/4.0/ | In the seventies', Zehnder found a Nash-Moser type implicit function theorem
in the analytic set-up. This theorem has found many applications in dynamical
systems although its applications require, as a general rule, some efforts. We
develop further the analytic theory and give fixed point theorems with direct
applications to the study of dynamical systems.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:06:12 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Garay",
"Mauricio",
""
],
[
"van Straten",
"Duco",
""
]
] |
2504.19809 | Hagop Sazdjian | Hagop Sazdjian | Meson scattering and tetraquarks in two-dimensional QCD | 48 pages, 14 figures | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | Two-quark--two-antiquark systems with four different quark flavors are
considered in the framework of two-dimensional QCD, in the light-cone gauge, at
leading orders of the $1/N_c^{}$ expansion. Introducing basis functions with
color-singlet mesonic clusters, integral equations are established for the
Green's functions and the related scattering amplitudes involved in the sectors
of the direct and quark-exchange channels. The problem of infrared divergences
is dealt with a systematic use of an infrared regulator cutoff introduced in
the gluon propagator. It is shown that the on-mass shell scattering amplitudes
are free of infrared divergences up to order $1/N_c^2$. In the limit of
vanishing of the infrared cutoff, they can be represented by effective
four-meson contact-type interaction terms with unitarity loops, calculable in
terms of the meson wave functions and propagators. The results, obtained at
order $1/N_c^2$, are summed with the constraint of unitarization. The
unitarized scattering amplitudes are continued in the total mass-squared
variable below the two-meson thresholds, leading to a tetraquark bound state
equation, which generally has one solution. Spectroscopic applications of the
above results are sketched and discussed.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:06:42 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sazdjian",
"Hagop",
""
]
] |
2504.19810 | Francesco Maio | F. Maio, V. Roccatagliata, D. Fedele, A. Garufi, A. Zurlo, C. Lazzoni,
S. Facchini, R.G. Gratton, D. Mesa, C. Toci, S. Antoniucci, S. Desidera, L. .
Pino, E. Rigliaco, C. Codella, L. Podio, V. D'Orazi, G. Lodato, F. Pedichini,
L. Testi | Protoplanetary disk insights from the first ERIS/APP survey at 4 {\mu}m | null | null | 10.1051/0004-6361/202553672 | null | astro-ph.EP | http://creativecommons.org/licenses/by/4.0/ | We present high-contrast imaging observations of seven protoplanetary disks
at 4um using the ERIS on the VLT. This study focuses on detecting scattered
light from micron-sized dust particles and assessing the potential of the vAPP
coronagraph for disk and planet characterization. Observations were performed
in pupil-stabilized mode with the vAPP coronagraph. Data were reduced using
reference differential imaging and angular differential imaging techniques,
incorporating principal component analysis for point-source detection. Contrast
curves and detection limits were computed for planetary companions and disk
features. The infrared disk signal was resolved in all systems, with first-time
4um detections around AS 209 and Elias 2-24, revealing mostly axisymmetric
structures extending up to 60au. Two gaps were detected in the radial profiles
of TW Hya (22au, 35au) and AS 209 (50au, 100au). For Elias 2-24, scattered
light emission matched ALMA observations of inner disk structures, marking
their first mid-infrared detection. In the case of HD 100546, the vAPP
uncovered flared disk structures and faint spiral arms consistent with previous
observations. HD 163296 shows a bright inner dust ring, confirming disk
asymmetries and features, but we did not detect any planet candidate within the
achieved contrast limits. The disk around PDS 70 exhibits clear features, with
faint structures detected within the cavity. The observations achieved
contrasts enabling the detection of planets down to 800 K, but no companions
were detected, implying either low-mass planets, cooler formation scenarios, or
a large dust extinction of Av>20 mag. The vAPP performed robustly for imaging
structures in protoplanetary disks at 4um, providing critical insights into
disk morphology and constraints on planet formation processes. No
planetary-mass companions with temperatures >1000K are present in our sample.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:08:38 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Maio",
"F.",
""
],
[
"Roccatagliata",
"V.",
""
],
[
"Fedele",
"D.",
""
],
[
"Garufi",
"A.",
""
],
[
"Zurlo",
"A.",
""
],
[
"Lazzoni",
"C.",
""
],
[
"Facchini",
"S.",
""
],
[
"Gratton",
"R. G.",
""
],
[
"Mesa",
"D.",
""
],
[
"Toci",
"C.",
""
],
[
"Antoniucci",
"S.",
""
],
[
"Desidera",
"S.",
""
],
[
"Pino",
"L. .",
""
],
[
"Rigliaco",
"E.",
""
],
[
"Codella",
"C.",
""
],
[
"Podio",
"L.",
""
],
[
"D'Orazi",
"V.",
""
],
[
"Lodato",
"G.",
""
],
[
"Pedichini",
"F.",
""
],
[
"Testi",
"L.",
""
]
] |
2504.19811 | Takuya Tamura | Takuya Tamura, Taro Yano, Masafumi Enomoto, Masafumi Oyamada | Can a Crow Hatch a Falcon? Lineage Matters in Predicting Large Language
Model Performance | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Accurately forecasting the performance of Large Language Models (LLMs) before
extensive fine-tuning or merging can substantially reduce both computational
expense and development time. Although prior approaches like scaling laws
account for global factors such as parameter size or training tokens, they
often overlook explicit lineage relationships - i.e., which models are derived
or merged from which parents. In this work, we propose a novel
Lineage-Regularized Matrix Factorization (LRMF) framework that encodes
ancestral ties among LLMs via a graph Laplacian regularizer. By leveraging
multi-hop parent-child connections, LRMF consistently outperforms conventional
matrix factorization and collaborative filtering methods in both instance-level
and benchmark-level performance prediction. Our large-scale study includes
2,934 publicly available Hugging Face models and 21,000+ instances across 6
major benchmarks, showing that lineage constraints yield up to 7-10 percentage
points higher correlation with actual performance compared to baselines.
Moreover, LRMF effectively addresses the cold-start problem, providing accurate
estimates for newly derived or merged models even with minimal data. This
lineage-guided strategy thus offers a resource-efficient way to inform
hyperparameter tuning, data selection, and model combination in modern LLM
development.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:08:45 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Tamura",
"Takuya",
""
],
[
"Yano",
"Taro",
""
],
[
"Enomoto",
"Masafumi",
""
],
[
"Oyamada",
"Masafumi",
""
]
] |
2504.19812 | Joseph Hart | Joseph Hart, Bart van Bloemen Waanders, Jixian Li, Timbwaoga A. J.
Ouermi, and Chris R. Johnson | Hyper-differential sensitivity analysis with respect to model
discrepancy: Prior Distributions | null | null | null | null | math.NA cs.NA | http://creativecommons.org/licenses/by/4.0/ | Hyper-differential sensitivity analysis with respect to model discrepancy was
recently developed to enable uncertainty quantification for optimization
problems. The approach consists of two primary steps: (i) Bayesian calibration
of the discrepancy between high- and low-fidelity models, and (ii) propagating
the model discrepancy uncertainty through the optimization problem. When
high-fidelity model evaluations are limited, as is common in practice, the
prior discrepancy distribution plays a crucial role in the uncertainty
analysis. However, specification of this prior is challenging due to its
mathematical complexity and many hyper-parameters. This article presents a
novel approach to specify the prior distribution. Our approach consists of two
parts: (1) an algorithmic initialization of the prior hyper-parameters that
uses existing data to initialize a hyper-parameter estimate, and (2) a
visualization framework to systematically explore properties of the prior and
guide tuning of the hyper-parameters to ensure that the prior captures the
appropriate range of uncertainty. We provide detailed mathematical analysis and
a collection of numerical examples that elucidate properties of the prior that
are crucial to ensure uncertainty quantification.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:09:37 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Hart",
"Joseph",
""
],
[
"Waanders",
"Bart van Bloemen",
""
],
[
"Li",
"Jixian",
""
],
[
"Ouermi",
"Timbwaoga A. J.",
""
],
[
"Johnson",
"Chris R.",
""
]
] |
2504.19813 | Veit Schw\"ammle | Xuan-Tung Trinh, Andr\'e Abrantes da Costa, David Bouyssi\'e, Adelina
Rogowska-Wrzesinska, Veit Schw\"ammle | OmicsQ: A User-Friendly Platform for Interactive Quantitative Omics Data
Analysis | 6 pages, 1 figure | null | null | null | q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | Motivation: High-throughput omics technologies generate complex datasets with
thousands of features that are quantified across multiple experimental
conditions, but often suffer from incomplete measurements, missing values and
individually fluctuating variances. This requires sophisticated analytical
methods for accurate, deep and insightful biological interpretations, capable
of dealing with a large variety of data properties and different amounts of
completeness. Software to handle such data complexity is rare and mostly relies
on programming-based environments, limiting accessibility for researchers
without computational expertise. Results: We present OmicsQ, an interactive,
web-based platform designed to streamline quantitative omics data analysis.
OmicsQ integrates established statistical processing tools with an intuitive,
browser-based visualization interface. It provides robust batch correction,
automated experimental design annotation, and missing-data handling without
imputation, which ensures data integrity and avoids artifacts from a priori
assumptions. OmicsQ seamlessly interacts with external applications for
statistical testing, clustering, analysis of protein complex behavior, and
pathway enrichment, offering a comprehensive and flexible workflow from data
import to biological interpretation that is broadly applicable tov data from
different domains. Availability and Implementation: OmicsQ is implemented in R
and R Shiny and is available at
https://computproteomics.bmb.sdu.dk/app_direct/OmicsQ. Source code and
installation instructions can be found at
https://github.com/computproteomics/OmicsQ
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:09:45 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Trinh",
"Xuan-Tung",
""
],
[
"da Costa",
"André Abrantes",
""
],
[
"Bouyssié",
"David",
""
],
[
"Rogowska-Wrzesinska",
"Adelina",
""
],
[
"Schwämmle",
"Veit",
""
]
] |
2504.19814 | Marie Fortin | Benedikt Bollig, Marie Fortin, Paul Gastin | High-Level Message Sequence Charts: Satisfiability and Realizability
Revisited | null | null | null | null | cs.LO cs.FL | http://creativecommons.org/licenses/by/4.0/ | Message sequence charts (MSCs) visually represent interactions in distributed
systems that communicate through FIFO channels. High-level MSCs (HMSCs) extend
MSCs with choice, concatenation, and iteration, allowing for the specification
of complex behaviors. This paper revisits two classical problems for HMSCs:
satisfiability and realizability. Satisfiability (also known as reachability or
nonemptiness) asks whether there exists a path in the HMSC that gives rise to a
valid behavior. Realizability concerns translating HMSCs into communicating
finite-state machines to ensure correct system implementations.
While most positive results assume bounded channels, we introduce a class of
HMSCs that allows for unbounded channels while maintaining effective
implementations. On the other hand, we show that the corresponding
satisfiability problem is still undecidable.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:10:33 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Bollig",
"Benedikt",
""
],
[
"Fortin",
"Marie",
""
],
[
"Gastin",
"Paul",
""
]
] |
2504.19815 | Amedeo Carbone | Amedeo Carbone, Diane-Pernille Bendixen-Fernex de Mongex, Arkady V.
Krasheninnikov, Martijn Wubs, Alexander Huck, Thomas W. Hansen, Alexander W.
Holleitner, Nicolas Stenger and Christoph Kastl | Creation and Microscopic Origins of Single-Photon Emitters in Transition
Metal Dichalcogenides and Hexagonal Boron Nitride | null | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We highlight recent advances in the controlled creation of single-photon
emitters in van der Waals materials and in the understanding of their atomistic
origin. We focus on quantum emitters created in monolayer transition-metal
dichalcogenide semiconductors, which provide spectrally sharp single-photon
emission at cryogenic temperatures, and the ones in insulating hBN, which
provide bright and stable single-photon emission up to room temperature. After
introducing the different classes of quantum emitters in terms of
band-structure properties, we review the defect creation methods based on
electron and ion irradiation as well as local strain engineering and plasma
treatments. A main focus of the review is put on discussing the microscopic
origin of the quantum emitters as revealed by various experimental platforms,
including optical and scanning probe methods.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:12:45 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Carbone",
"Amedeo",
""
],
[
"de Mongex",
"Diane-Pernille Bendixen-Fernex",
""
],
[
"Krasheninnikov",
"Arkady V.",
""
],
[
"Wubs",
"Martijn",
""
],
[
"Huck",
"Alexander",
""
],
[
"Hansen",
"Thomas W.",
""
],
[
"Holleitner",
"Alexander W.",
""
],
[
"Stenger",
"Nicolas",
""
],
[
"Kastl",
"Christoph",
""
]
] |
2504.19816 | Erblin Isaku | Erblin Isaku, Hassan Sartaj, Shaukat Ali | Digital Twin-based Out-of-Distribution Detection in Autonomous Vessels | 34 pages, 12 figures, 11 tables | null | null | null | eess.SY cs.LG cs.SE cs.SY | http://creativecommons.org/licenses/by/4.0/ | An autonomous vessel (AV) is a complex cyber-physical system (CPS) with
software enabling many key functionalities, e.g., navigation software enables
an AV to autonomously or semi-autonomously follow a path to its destination.
Digital twins of such AVs enable advanced functionalities such as running
what-if scenarios, performing predictive maintenance, and enabling fault
diagnosis. Due to technological improvements, real-time analyses using
continuous data from vessels' real-time operations have become increasingly
possible. However, the literature has little explored developing advanced
analyses in real-time data in AVs with digital twins built with machine
learning techniques. To this end, we present a novel digital twin-based
approach (ODDIT) to detect future out-of-distribution (OOD) states of an AV
before reaching them, enabling proactive intervention. Such states may indicate
anomalies requiring attention (e.g., manual correction by the ship master) and
assist testers in scenario-centered testing. The digital twin consists of two
machine-learning models predicting future vessel states and whether the
predicted state will be OOD. We evaluated ODDIT with five vessels across
waypoint and zigzag maneuvering under simulated conditions, including sensor
and actuator noise and environmental disturbances i.e., ocean current. ODDIT
achieved high accuracy in detecting OOD states, with AUROC and TNR@TPR95 scores
reaching 99\% across multiple vessels.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:12:46 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Isaku",
"Erblin",
""
],
[
"Sartaj",
"Hassan",
""
],
[
"Ali",
"Shaukat",
""
]
] |
2504.19817 | Yiqing Pan | Qihan He, Wenxuan Liu and Yiqing Pan | Positive solutions of critical Hardy-H\'{e}non equations with
logarithmic term | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the existence, non-existence and multiplicity of positive
solutions to the following critical Hardy-H\'{e}non equation with logarithmic
term \begin{equation*}\label{eq11}\left\{ \begin{array}{ll} -\Delta u
=|x|^{\alpha}|u|^{2^*_{\alpha}-2}\cdot u+\mu u\log u^2+\lambda u, &x\in
\Omega,\\ u=0, &x\in \partial \Omega,\\ \end{array} \right.\end{equation*}
where $ \Omega=B$ for $\alpha\geq 0$, $ \Omega=B\setminus\{0\}$ for
$\alpha\in(-2,0)$, $B\subset\mathbb{R}^N$ is an unit ball, $\lambda, \mu \in
\mathbb{R}$, $N\geq 3, \alpha>-2$, $2^*_{\alpha}:=\frac{2(N+\alpha)}{N-2}$ is
the critical exponent for the embedding $H_{0,r}^{1}( \Omega)\hookrightarrow
L^p( \Omega;|x|^\alpha)$, and which can be seen as a Br\'{e}zis-Nirenberg
problem. When $N \geq 4$ and $\mu>0$, we will show that the above problem has a
positive Mountain pass solution, which is also a ground state solution. At the
same time, when $\mu<0$, under some assumptions on the $N$, $\mu$, $\lambda$
and $\alpha$, we will show that the above problem has at least a positive least
energy solution and at least a positive Mountain pass solution, respectively.
What's more, when certain inequality related to $N \geq 3$, $\mu<0 $ and
$\alpha\in(-2,0]$ holds, we will demonstrate the non-existence of positive
solutions to the above-mentioned problem. The presence of logarithmic term
brings some new and interesting phenomena to this problem.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:18:48 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"He",
"Qihan",
""
],
[
"Liu",
"Wenxuan",
""
],
[
"Pan",
"Yiqing",
""
]
] |
2504.19818 | Feng Chen | Feng Chen, Ilias Stogiannidis, Andrew Wood, Danilo Bueno, Dominic
Williams, Fraser Macfarlane, Bruce Grieve, Darren Wells, Jonathan A.
Atkinson, Malcolm J. Hawkesford, Stephen A. Rolfe, Tracy Lawson, Tony
Pridmore, Mario Valerio Giuffrida, Sotirios A. Tsaftaris | PhenoAssistant: A Conversational Multi-Agent AI System for Automated
Plant Phenotyping | null | null | null | null | cs.MA cs.AI | http://creativecommons.org/licenses/by/4.0/ | Plant phenotyping increasingly relies on (semi-)automated image-based
analysis workflows to improve its accuracy and scalability. However, many
existing solutions remain overly complex, difficult to reimplement and
maintain, and pose high barriers for users without substantial computational
expertise. To address these challenges, we introduce PhenoAssistant: a
pioneering AI-driven system that streamlines plant phenotyping via intuitive
natural language interaction. PhenoAssistant leverages a large language model
to orchestrate a curated toolkit supporting tasks including automated phenotype
extraction, data visualisation and automated model training. We validate
PhenoAssistant through several representative case studies and a set of
evaluation tasks. By significantly lowering technical hurdles, PhenoAssistant
underscores the promise of AI-driven methodologies to democratising AI adoption
in plant biology.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:20:30 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Chen",
"Feng",
""
],
[
"Stogiannidis",
"Ilias",
""
],
[
"Wood",
"Andrew",
""
],
[
"Bueno",
"Danilo",
""
],
[
"Williams",
"Dominic",
""
],
[
"Macfarlane",
"Fraser",
""
],
[
"Grieve",
"Bruce",
""
],
[
"Wells",
"Darren",
""
],
[
"Atkinson",
"Jonathan A.",
""
],
[
"Hawkesford",
"Malcolm J.",
""
],
[
"Rolfe",
"Stephen A.",
""
],
[
"Lawson",
"Tracy",
""
],
[
"Pridmore",
"Tony",
""
],
[
"Giuffrida",
"Mario Valerio",
""
],
[
"Tsaftaris",
"Sotirios A.",
""
]
] |
2504.19819 | Hoang Chuong Nguyen Mr. | Hoang Chuong Nguyen, Wei Mao, Jose M. Alvarez, Miaomiao Liu | Joint Optimization of Neural Radiance Fields and Continuous Camera
Motion from a Monocular Video | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Neural Radiance Fields (NeRF) has demonstrated its superior capability to
represent 3D geometry but require accurately precomputed camera poses during
training. To mitigate this requirement, existing methods jointly optimize
camera poses and NeRF often relying on good pose initialisation or depth
priors. However, these approaches struggle in challenging scenarios, such as
large rotations, as they map each camera to a world coordinate system. We
propose a novel method that eliminates prior dependencies by modeling
continuous camera motions as time-dependent angular velocity and velocity.
Relative motions between cameras are learned first via velocity integration,
while camera poses can be obtained by aggregating such relative motions up to a
world coordinate system defined at a single time step within the video.
Specifically, accurate continuous camera movements are learned through a
time-dependent NeRF, which captures local scene geometry and motion by training
from neighboring frames for each time step. The learned motions enable
fine-tuning the NeRF to represent the full scene geometry. Experiments on Co3D
and Scannet show our approach achieves superior camera pose and depth
estimation and comparable novel-view synthesis performance compared to
state-of-the-art methods. Our code is available at
https://github.com/HoangChuongNguyen/cope-nerf.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:22:04 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Nguyen",
"Hoang Chuong",
""
],
[
"Mao",
"Wei",
""
],
[
"Alvarez",
"Jose M.",
""
],
[
"Liu",
"Miaomiao",
""
]
] |
2504.19821 | Martin Dunsche | Martin Dunsche, Patrick Bastian, Marcel Maehren, Nurullah Erinola,
Robert Merget, Nicolai Bissantz, Holger Dette, J\"org Schwenk | SILENT: A New Lens on Statistics in Software Timing Side Channels | null | null | null | null | cs.CR stat.AP stat.ME | http://creativecommons.org/licenses/by/4.0/ | Cryptographic research takes software timing side channels seriously.
Approaches to mitigate them include constant-time coding and techniques to
enforce such practices. However, recent attacks like Meltdown [42], Spectre
[37], and Hertzbleed [70] have challenged our understanding of what it means
for code to execute in constant time on modern CPUs. To ensure that assumptions
on the underlying hardware are correct and to create a complete feedback loop,
developers should also perform \emph{timing measurements} as a final validation
step to ensure the absence of exploitable side channels. Unfortunately, as
highlighted by a recent study by Jancar et al. [30], developers often avoid
measurements due to the perceived unreliability of the statistical analysis and
its guarantees.
In this work, we combat the view that statistical techniques only provide
weak guarantees by introducing a new algorithm for the analysis of timing
measurements with strong, formal statistical guarantees, giving developers a
reliable analysis tool. Specifically, our algorithm (1) is non-parametric,
making minimal assumptions about the underlying distribution and thus
overcoming limitations of classical tests like the t-test, (2) handles unknown
data dependencies in measurements, (3) can estimate in advance how many samples
are needed to detect a leak of a given size, and (4) allows the definition of a
negligible leak threshold $\Delta$, ensuring that acceptable non-exploitable
leaks do not trigger false positives, without compromising statistical
soundness. We demonstrate the necessity, effectiveness, and benefits of our
approach on both synthetic benchmarks and real-world applications.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:22:23 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Dunsche",
"Martin",
""
],
[
"Bastian",
"Patrick",
""
],
[
"Maehren",
"Marcel",
""
],
[
"Erinola",
"Nurullah",
""
],
[
"Merget",
"Robert",
""
],
[
"Bissantz",
"Nicolai",
""
],
[
"Dette",
"Holger",
""
],
[
"Schwenk",
"Jörg",
""
]
] |
2504.19822 | Minjong Cheon | Minjong Cheon | Mj\"olnir: A Deep Learning Parametrization Framework for Global
Lightning Flash Density | null | null | null | null | cs.LG cs.AI cs.CV physics.ao-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Recent advances in AI-based weather forecasting models, such as FourCastNet,
Pangu-Weather, and GraphCast, have demonstrated the remarkable ability of deep
learning to emulate complex atmospheric dynamics. Building on this momentum, we
propose Mj\"olnir, a novel deep learning-based framework for global lightning
flash density parameterization. Trained on ERA5 atmospheric predictors and
World Wide Lightning Location Network (WWLLN) observations at a daily temporal
resolution and 1 degree spatial resolution, Mj\"olnir captures the nonlinear
mapping between large-scale environmental conditions and lightning activity.
The model architecture is based on the InceptionNeXt backbone with SENet, and a
multi-task learning strategy to simultaneously predict lightning occurrence and
magnitude. Extensive evaluations yield that Mollnir accurately reproduces the
global distribution, seasonal variability, and regional characteristics of
lightning activity, achieving a global Pearson correlation coefficient of 0.96
for annual mean fields. These results suggest that Mj\"olnir serves not only as
an effective data-driven global lightning parameterization but also as a
promising AI-based scheme for next-generation Earth system models (AI-ESMs).
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:22:59 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Cheon",
"Minjong",
""
]
] |
2504.19823 | Dragos-Patru Covei D.P. | Dragos-Patru Covei | A nonlinear diffusion equation of the Gurtin-MacCamy type: existence,
uniqueness, and numerical simulations | 17 pages | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper examines a nonlinear diffusion process that arises in the modeling
of biological population dynamics and extends to applications in thermal
diffusion and material sciences. By investigating a power-law governed
diffusion mechanism coupled with a time-dependent growth rate, we establish the
existence and uniqueness of solutions to the associated boundary-value problem
within a bounded domain. Our approach departs from traditional methods by
transforming the original problem into a simplified elliptic framework, which
not only facilitates rigorous theoretical analysis but also paves the way for
efficient numerical simulations. In addition to the analytical results, we
develop a numerical scheme implemented in Python to visualize the evolution of
the population density across time and space. These simulations confirm the
theoretical predictions and demonstrate the model's sensitivity to variations
in the nonlinear exponent, growth rate, and other parameters. The
interdisciplinary nature of our work suggests that the insights gained from the
study of nonlinear diffusion may have far-reaching implications, ranging from
the control of microbial populations in biomedical contexts to the analysis of
energy transport in advanced materials. Overall, our findings contribute to a
deeper understanding of complex diffusion phenomena and offer a novel
computational framework that can be adapted to a wide range of scientific and
engineering problems.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:23:24 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Covei",
"Dragos-Patru",
""
]
] |
2504.19824 | Mohamed Hassan | Mohamed Hassan, Mohammad Wasil and Sebastian Houben | Taming the Randomness: Towards Label-Preserving Cropping in Contrastive
Learning | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Contrastive learning (CL) approaches have gained great recognition as a very
successful subset of self-supervised learning (SSL) methods. SSL enables
learning from unlabeled data, a crucial step in the advancement of deep
learning, particularly in computer vision (CV), given the plethora of unlabeled
image data. CL works by comparing different random augmentations (e.g.,
different crops) of the same image, thus achieving self-labeling. Nevertheless,
randomly augmenting images and especially random cropping can result in an
image that is semantically very distant from the original and therefore leads
to false labeling, hence undermining the efficacy of the methods. In this
research, two novel parameterized cropping methods are introduced that increase
the robustness of self-labeling and consequently increase the efficacy. The
results show that the use of these methods significantly improves the accuracy
of the model by between 2.7\% and 12.4\% on the downstream task of classifying
CIFAR-10, depending on the crop size compared to that of the non-parameterized
random cropping method.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:24:25 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Hassan",
"Mohamed",
""
],
[
"Wasil",
"Mohammad",
""
],
[
"Houben",
"Sebastian",
""
]
] |
2504.19825 | Samuel Smith | Ilan Ben-Zvi, Alick Macpherson, Samuel Smith | Detailed Design and Optimization of Ferro-Electric Tuners | 11 pages, 12 figures | null | null | null | physics.acc-ph | http://creativecommons.org/licenses/by/4.0/ | A detailed, step-by-step design methodology of a Ferroelectric Fast Reactive
Tuner (FE-FRT) capable of modulating Mega VAR reactive powers on a
sub-microsecond time scale is given. Closed expressions of values for all the
components of the tuner are detailed, and tuner performance optimization is
addressed, resulting in a Figure of Merit measure of FE-FRT tuner performance
and use case applicability over a wide range of RF frequencies and reactive
power levels. This enables addressing feasibility and rapid assessment of
design parameters given an FE-FRT tuning scenario defined by required tuning
range, cavity operating frequency a cavity stored energy.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:25:23 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Ben-Zvi",
"Ilan",
""
],
[
"Macpherson",
"Alick",
""
],
[
"Smith",
"Samuel",
""
]
] |
2504.19826 | Justin Lynd | Ellen Henke and Justin Lynd | Components and realizability of fusion systems | null | null | null | null | math.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For $p=2$ it is known that a saturated fusion system is realizable if and
only if each of its components is realizable by a finite quasisimple group. For
odd primes $p$ this is false. Building on work of Broto, M{\o}ller, Oliver, and
Ruiz, we show however that a fusion system $\mathcal{F}$ is realizable if and
only if for each of its components $\mathcal{C}$ there exists a realizable
subnormal subsystem $\mathcal{E}$ of $\mathcal{F}$ with
$O^{p^\prime}(\mathcal{E})=\mathcal{C}$.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:26:21 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Henke",
"Ellen",
""
],
[
"Lynd",
"Justin",
""
]
] |
2504.19827 | Sven Danz | Sven Danz, Tobias Stollenwerk and Alessandro Ciani | Quantum oracles for the finite element method | 14+7 pages, 17 figures | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | In order to assess potential advantages of quantum algorithms that require
quantum oracles as subroutines, the careful evaluation of the overall
complexity of the oracles themselves is crucial. This study examines the
quantum routines required for the implementation of oracles used in the
block-encoding of the $N \times N$ stiffness and mass matrices, which typically
emerge in the finite element analysis of elastic structures. Starting from
basic quantum adders, we show how to construct the necessary oracles, which
require the calculation of polynomials, square root and the implementation of
conditional operations. We propose quantum subroutines based on fixed-point
arithmetic that, given an $r$-qubit register, construct the oracle using
$\mathcal{O}((K + L + N_{\mathrm{geo}} + N_{\mathrm{D}}) r)$ ancilla qubits and
have a $\mathcal{O}((K + L)r^2 + \log_2(N_{\mathrm{geo}} + N_{\mathrm{D}}))$
runtime, with $K$ the order at which we truncate the polynomials, $L$ the
number of iterations in the Newton-Raphson subroutine for the square root,
while $N_{\mathrm{geo}}$ and $N_{\mathrm{D}}$ are the number of hypercuboids
used to approximate the geometry and the boundary, respectively. Since in
practice $r$ scales as $r = \mathcal{O}(\log_2 N)$, and assuming that the other
parameters are fixed independently of $N$, this shows that the oracles, while
still costly in practice, do not endanger potential polynomial or exponential
advantages in $N$.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:28:31 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Danz",
"Sven",
""
],
[
"Stollenwerk",
"Tobias",
""
],
[
"Ciani",
"Alessandro",
""
]
] |
2504.19828 | Zhiming Hu | Zhiming Hu, Daniel Haeufle, Syn Schmitt, Andreas Bulling | HOIGaze: Gaze Estimation During Hand-Object Interactions in Extended
Reality Exploiting Eye-Hand-Head Coordination | Accepted at SIGGRAPH 2025, link:
https://zhiminghu.net/hu25_hoigaze.html | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present HOIGaze - a novel learning-based approach for gaze estimation
during hand-object interactions (HOI) in extended reality (XR). HOIGaze
addresses the challenging HOI setting by building on one key insight: The eye,
hand, and head movements are closely coordinated during HOIs and this
coordination can be exploited to identify samples that are most useful for gaze
estimator training - as such, effectively denoising the training data. This
denoising approach is in stark contrast to previous gaze estimation methods
that treated all training samples as equal. Specifically, we propose: 1) a
novel hierarchical framework that first recognises the hand currently visually
attended to and then estimates gaze direction based on the attended hand; 2) a
new gaze estimator that uses cross-modal Transformers to fuse head and
hand-object features extracted using a convolutional neural network and a
spatio-temporal graph convolutional network; and 3) a novel eye-head
coordination loss that upgrades training samples belonging to the coordinated
eye-head movements. We evaluate HOIGaze on the HOT3D and Aria digital twin
(ADT) datasets and show that it significantly outperforms state-of-the-art
methods, achieving an average improvement of 15.6% on HOT3D and 6.0% on ADT in
mean angular error. To demonstrate the potential of our method, we further
report significant performance improvements for the sample downstream task of
eye-based activity recognition on ADT. Taken together, our results underline
the significant information content available in eye-hand-head coordination
and, as such, open up an exciting new direction for learning-based gaze
estimation.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:31:43 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Hu",
"Zhiming",
""
],
[
"Haeufle",
"Daniel",
""
],
[
"Schmitt",
"Syn",
""
],
[
"Bulling",
"Andreas",
""
]
] |
2504.19829 | Lukas K\"uhne | Siddarth Kannan and Lukas K\"uhne | Chow groups of braid matroids and relative maps to $\mathbb{P}^1$ | 21 pages. Comments are welcome | null | null | null | math.AG math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We provide a modular interpretation for the rational Chow groups of the braid
matroid $B_n$, with respect to the maximal building set, in terms of certain
moduli spaces of genus zero relative stable maps to $\mathbb{P}^1$. We use this
perspective to compute the generating function for the Chow polynomials.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:32:32 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Kannan",
"Siddarth",
""
],
[
"Kühne",
"Lukas",
""
]
] |
2504.19830 | Levon Pogosian | Ali Nezhadsafavi, Levon Pogosian | Cosmic strings in the complex symmetron model | 7 pages, 4 figures | null | null | null | hep-ph astro-ph.CO gr-qc hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study cosmic strings in the complex symmetron model, a scalar-tensor
theory with a spontaneously broken local $U(1)$ symmetry in low matter density
regions. Using numerical simulations, we show that these strings preferentially
attach to matter halos, leading to the stabilization of string loops. While the
requirement for screening of fifth-force interactions in the solar system
limits observable signatures in theories with universal coupling to matter,
analogous topological defects in the dark sector may still influence cosmic
structure formation, offering a novel avenue to constrain dark-sector
interactions.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:33:14 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Nezhadsafavi",
"Ali",
""
],
[
"Pogosian",
"Levon",
""
]
] |
2504.19831 | Haiyan Zhu | Haiyan Zhu and Yingchun Zhou | Optimal real-time dynamic treatment regimes with application to oxytocin
use in preventing postpartum hemorrhage | null | null | null | null | stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Real-time dynamic treatment regimes (real-time DTRs) refers to decision rules
that personalize patient treatment in real-time based on treatment and
covariate histories. These rules are crucial for real-time clinical decision
support systems and automated drug delivery systems for chronic diseases.
Although considerable statistical and machine learning DTR methods have been
developed, they are designed for a small number of fixed decision points, and
thus can not adapt to real-time cases. This paper proposes a new semiparametric
Bayesian method for estimating an optimal treatment regime in real-time, which
allows for the existence of latent individual level variables. Specifically,
random real-time DTRs are defined through interventional parameters, the
optimal values of which are estimated by maximizing the posterior predictive
utility. The proposed approach is compared with alternative methods using
simulated datasets, and applied to estimate the optimal real-time oxytocin
administration regime for preventing postpartum hemorrhage.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:33:51 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Zhu",
"Haiyan",
""
],
[
"Zhou",
"Yingchun",
""
]
] |
2504.19832 | Dan Ben-Moshe | Dan Ben-Moshe, David Genesove | Identifying the Frontier Structural Function and Bounding Mean
Deviations | null | null | null | null | econ.EM econ.GN q-fin.EC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper analyzes a model in which an outcome variable equals the
difference between a frontier function of inputs and a nonnegative unobserved
deviation. If zero is in the support of the deviation at a given input value,
then the frontier function is identified by the maximum outcome there. This
obviates the need for instrumental variables. Implementation requires allowing
for the distribution of deviations to depend on inputs, thus not ruling out
endogenous inputs and ensuring the estimated frontier is not merely a constant
shift of a biased conditional expectation. Including random errors results in a
stochastic frontier analysis model generalized to allow the joint distribution
of deviations and errors to depend on inputs. If the minimum deviation is a
function of inputs, then we derive a lower bound for the mean deviation using
variance and skewness, without making parametric distributional assumptions. We
apply our results to a frontier production function, with deviations
representing inefficiencies.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:34:54 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Ben-Moshe",
"Dan",
""
],
[
"Genesove",
"David",
""
]
] |
2504.19833 | Mohd Shah Nurisya | Valentine Nyirahafashimana, Nurisya Mohd Shah, Umair Abdul Halim,
Mohamed Othman, Sharifah Kartini Said Husain | Quantum Error Correction in Quaternionic Hilbert Spaces | 10 pages, 1 figure, 2 tables. New submission on quaternionic
extensions of quantum error correction. A quaternionic five-qubit code with
improved error threshold is constructed and analyzed | null | null | null | quant-ph math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | We propose quaternion-based strategies for quantum error correction by
extending quantum mechanics into quaternionic Hilbert spaces. Building on the
properties of quaternionic quantum states, we define quaternionic analogues of
Pauli operators and quantum gates, ensuring inner product preservation and
Hilbert space conditions. A simple encoding scheme maps logical qubits into
quaternionic systems, introducing natural redundancy and enhanced resilience
against noise. We construct a quaternionic extension of the five-qubit code,
introducing a framework of 15 syndrome measurements to detect quaternionic
errors, including quaternionically rotated error components. Numerical
estimates show that the quaternionic five-qubit code achieves a logical error
threshold of approximately $p_{th} \approx 0.015$, demonstrating improved
performance compared to the standard complex-valued code. These results suggest
a new pathway for quantum error correction in high-noise environments,
leveraging the richer structure of quaternionic quantum mechanics to improve
fault tolerance.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:34:57 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Nyirahafashimana",
"Valentine",
""
],
[
"Shah",
"Nurisya Mohd",
""
],
[
"Halim",
"Umair Abdul",
""
],
[
"Othman",
"Mohamed",
""
],
[
"Husain",
"Sharifah Kartini Said",
""
]
] |
2504.19834 | Xiaoyu Liu | Xiaoyu Liu, Mingshuai Yao, Yabo Zhang, Xianhui Lin, Peiran Ren,
Xiaoming Li, Ming Liu, Wangmeng Zuo | AnimateAnywhere: Rouse the Background in Human Image Animation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human image animation aims to generate human videos of given characters and
backgrounds that adhere to the desired pose sequence. However, existing methods
focus more on human actions while neglecting the generation of background,
which typically leads to static results or inharmonious movements. The
community has explored camera pose-guided animation tasks, yet preparing the
camera trajectory is impractical for most entertainment applications and
ordinary users. As a remedy, we present an AnimateAnywhere framework, rousing
the background in human image animation without requirements on camera
trajectories. In particular, based on our key insight that the movement of the
human body often reflects the motion of the background, we introduce a
background motion learner (BML) to learn background motions from human pose
sequences. To encourage the model to learn more accurate cross-frame
correspondences, we further deploy an epipolar constraint on the 3D attention
map. Specifically, the mask used to suppress geometrically unreasonable
attention is carefully constructed by combining an epipolar mask and the
current 3D attention map. Extensive experiments demonstrate that our
AnimateAnywhere effectively learns the background motion from human pose
sequences, achieving state-of-the-art performance in generating human animation
results with vivid and realistic backgrounds. The source code and model will be
available at https://github.com/liuxiaoyu1104/AnimateAnywhere.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:35:01 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Liu",
"Xiaoyu",
""
],
[
"Yao",
"Mingshuai",
""
],
[
"Zhang",
"Yabo",
""
],
[
"Lin",
"Xianhui",
""
],
[
"Ren",
"Peiran",
""
],
[
"Li",
"Xiaoming",
""
],
[
"Liu",
"Ming",
""
],
[
"Zuo",
"Wangmeng",
""
]
] |
2504.19837 | Madhumita Kabiraj | Madhumita Kabiraj, Raka Dasgupta | Finite temperature phase diagram of the extended Bose-Hubbard model in
the presence of disorder | null | null | null | null | cond-mat.quant-gas cond-mat.stat-mech quant-ph | http://creativecommons.org/licenses/by/4.0/ | We study the finite- and non-zero temperature phase diagram of the Extended
Bose-Hubbard Model for both pure and disordered systems. Such a system can be
experimentally realized by trapping ultracold Rydberg atoms in optical
lattices. By regulating the Rydberg excitation level and the lattice spacing,
the system can be engineered to effectively have (i) only the nearest-neighbor
interaction and (ii) both nearest-neighbor and next-nearest-neighbor
interactions. For both of these situations, we construct the mean-field phase
diagrams. It is found that the presence of a non-zero temperature significantly
changes the phase diagram because now there is a competition between quantum
and thermal fluctuations. We observe that conventional Mott insulator (MI) or
charge-density-wave (CDW) lobes vanish at higher temperatures. In a pure
system, they melt into a normal fluid. In contrast, the only insulating phase
that survives at high temperatures in the presence of disorder is a Bose glass.
It is evident that the CDW lobes melt at a lower temperature and the Mott lobes
melt at higher temperatures. These transition temperatures depend on the
on-site and nearest-neighbor interaction strengths, respectively. It is also
found that, with the addition of disorder, the insulating lobes are destroyed
at a relatively lower temperature. The mathematical framework that we present
here is capable of treating long-range interactions, disorder, and finite
temperature simultaneously, and versatile enough so that it can be extended to
study different forms of disorder or longer-range interactions.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:38:51 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Kabiraj",
"Madhumita",
""
],
[
"Dasgupta",
"Raka",
""
]
] |
2504.19838 | Liu Guangyi | Guangyi Liu, Pengxiang Zhao, Liang Liu, Yaxuan Guo, Han Xiao, Weifeng
Lin, Yuxiang Chai, Yue Han, Shuai Ren, Hao Wang, Xiaoyu Liang, Wenhao Wang,
Tianze Wu, Linghao Li, Hao Wang, Guanjing Xiong, Yong Liu, Hongsheng Li | LLM-Powered GUI Agents in Phone Automation: Surveying Progress and
Prospects | 37 pages, 10 figures, 7 tables, Project Homepage:
https://github.com/PhoneLLM/Awesome-LLM-Powered-Phone-GUI-Agents | null | null | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | With the rapid rise of large language models (LLMs), phone automation has
undergone transformative changes. This paper systematically reviews LLM-driven
phone GUI agents, highlighting their evolution from script-based automation to
intelligent, adaptive systems. We first contextualize key challenges, (i)
limited generality, (ii) high maintenance overhead, and (iii) weak intent
comprehension, and show how LLMs address these issues through advanced language
understanding, multimodal perception, and robust decision-making. We then
propose a taxonomy covering fundamental agent frameworks (single-agent,
multi-agent, plan-then-act), modeling approaches (prompt engineering,
training-based), and essential datasets and benchmarks. Furthermore, we detail
task-specific architectures, supervised fine-tuning, and reinforcement learning
strategies that bridge user intent and GUI operations. Finally, we discuss open
challenges such as dataset diversity, on-device deployment efficiency,
user-centric adaptation, and security concerns, offering forward-looking
insights into this rapidly evolving field. By providing a structured overview
and identifying pressing research gaps, this paper serves as a definitive
reference for researchers and practitioners seeking to harness LLMs in
designing scalable, user-friendly phone GUI agents.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:39:25 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Liu",
"Guangyi",
""
],
[
"Zhao",
"Pengxiang",
""
],
[
"Liu",
"Liang",
""
],
[
"Guo",
"Yaxuan",
""
],
[
"Xiao",
"Han",
""
],
[
"Lin",
"Weifeng",
""
],
[
"Chai",
"Yuxiang",
""
],
[
"Han",
"Yue",
""
],
[
"Ren",
"Shuai",
""
],
[
"Wang",
"Hao",
""
],
[
"Liang",
"Xiaoyu",
""
],
[
"Wang",
"Wenhao",
""
],
[
"Wu",
"Tianze",
""
],
[
"Li",
"Linghao",
""
],
[
"Wang",
"Hao",
""
],
[
"Xiong",
"Guanjing",
""
],
[
"Liu",
"Yong",
""
],
[
"Li",
"Hongsheng",
""
]
] |
2504.19839 | Yulong Guo | Yulong Guo, Zilun Zhang, Yongheng Shang, Tiancheng Zhao, Shuiguang
Deng, Yingchun Yang, Jianwei Yin | SRMF: A Data Augmentation and Multimodal Fusion Approach for Long-Tail
UHR Satellite Image Segmentation | None | null | 10.1109/TGRS.2025.3565600 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The long-tail problem presents a significant challenge to the advancement of
semantic segmentation in ultra-high-resolution (UHR) satellite imagery. While
previous efforts in UHR semantic segmentation have largely focused on
multi-branch network architectures that emphasize multi-scale feature
extraction and fusion, they have often overlooked the importance of addressing
the long-tail issue. In contrast to prior UHR methods that focused on
independent feature extraction, we emphasize data augmentation and multimodal
feature fusion to alleviate the long-tail problem. In this paper, we introduce
SRMF, a novel framework for semantic segmentation in UHR satellite imagery. Our
approach addresses the long-tail class distribution by incorporating a
multi-scale cropping technique alongside a data augmentation strategy based on
semantic reordering and resampling. To further enhance model performance, we
propose a multimodal fusion-based general representation knowledge injection
method, which, for the first time, fuses text and visual features without the
need for individual region text descriptions, extracting more robust features.
Extensive experiments on the URUR, GID, and FBP datasets demonstrate that our
method improves mIoU by 3.33\%, 0.66\%, and 0.98\%, respectively, achieving
state-of-the-art performance. Code is available at:
https://github.com/BinSpa/SRMF.git.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:39:59 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Guo",
"Yulong",
""
],
[
"Zhang",
"Zilun",
""
],
[
"Shang",
"Yongheng",
""
],
[
"Zhao",
"Tiancheng",
""
],
[
"Deng",
"Shuiguang",
""
],
[
"Yang",
"Yingchun",
""
],
[
"Yin",
"Jianwei",
""
]
] |
2504.19840 | Behrouz Arash | Shadab Zakavati, Shahriar Salimi, Behrouz Arash | Optimizing the Charging of Open Quantum Batteries using Long Short-Term
Memory-Driven Reinforcement Learning | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Controlling the charging process of a quantum battery involves strategies to
efficiently transfer, store, and retain energy, while mitigating decoherence,
energy dissipation, and inefficiencies caused by surrounding interactions. We
develop a model to study the charging process of a quantum battery in an open
quantum setting, where the battery interacts with a charger and a structured
reservoir. To overcome the limitations of static charging protocols, a
reinforcement learning (RL) charging strategy is proposed, which utilizes the
deep deterministic policy gradient algorithm alongside long short-term memory
(LSTM) networks. The LSTM networks enable the RL model to capture temporal
correlations driven by non-Markovian dynamics, facilitating a continuous,
adaptive charging strategy. The RL protocols consistently outperform
conventional fixed heuristic strategies by real-time controlling the driving
field amplitude and coupling parameters. By penalizing battery-to-charger
backflow in the reward function, the RL-optimized charging strategy promotes
efficient unidirectional energy transfer from charger to battery, achieving
higher and more stable extractable work. The proposed RL controller would
provide a framework for designing efficient charging schemes in broader
configurations and multi-cell quantum batteries.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:40:11 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Zakavati",
"Shadab",
""
],
[
"Salimi",
"Shahriar",
""
],
[
"Arash",
"Behrouz",
""
]
] |
2504.19841 | Luis Alvarez | Luis Alvarez and Bruno Ferman and Kaspar W\"uthrich | Inference with few treated units | null | null | null | null | econ.EM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In many causal inference applications, only one or a few units (or clusters
of units) are treated. An important challenge in such settings is that standard
inference methods that rely on asymptotic theory may be unreliable, even when
the total number of units is large. This survey reviews and categorizes
inference methods that are designed to accommodate few treated units,
considering both cross-sectional and panel data methods. We discuss trade-offs
and connections between different approaches. In doing so, we propose slight
modifications to improve the finite-sample validity of some methods, and we
also provide theoretical justifications for existing heuristic approaches that
have been proposed in the literature.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:41:13 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Alvarez",
"Luis",
""
],
[
"Ferman",
"Bruno",
""
],
[
"Wüthrich",
"Kaspar",
""
]
] |
2504.19843 | Enea Parini | Azahara DelaTorre, Enea Parini | On Hopf's Lemma for sign-changing supersolutions to fractional Laplacian
equations | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we investigate the validity of Hopf's Lemma for a (possibly
sign-changing) function $u \in H^s_0(\Omega)$ satisfying \[ (-\Delta)^s u(x)
\geq c(x)u(x) \quad \text{in }\Omega,\] where $\Omega \subset \mathbb{R}^N$ is
an open, bounded domain, $c \in L^\infty(\Omega)$, and $(-\Delta)^s u$ is the
fractional Laplacian of $u$. We show that, under suitable assumptions, the
validity of Hopf's Lemma for $u$ at a point $x_0 \in \partial \Omega$ is
essentially equivalent to the validity of Hopf's Lemma for the
Caffarelli-Silvestre extension of $u$ at the point $(x_0,0) \in \mathbb{R}^N
\times \mathbb{R}^+$. We also provide a slightly more precise characterization
of a dichotomy result stated in a recent paper by Dipierro, Soave and
Valdinoci.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:43:00 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"DelaTorre",
"Azahara",
""
],
[
"Parini",
"Enea",
""
]
] |
2504.19844 | Yu-Xuan Li | Yu-Xuan Li, Yicheng Liu, and Cheng-Cheng Liu | Tunable Topological Superconductivity by Fully Compensated Ferrimagnets | 7 pages, 4 figures | null | null | null | cond-mat.supr-con cond-mat.mes-hall cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | We propose a platform based on a fully compensated ferrimagnet (fFIM) for
realizing and controlling topological superconductivity with Majorana bound
states across multiple dimensions. Through symmetry analysis and microscopic
modeling, we demonstrate that fFIM-based heterostructures host (i) Majorana
zero modes localized at the ends of one-dimensional nanowires, (ii) chiral
Majorana edge states along two-dimensional boundaries, and (iii) tunable
Majorana corner modes in higher-order topological phases. The unique properties
of fFIMs enable an electric field to drive topological superconductivity phase
transitions and N\'eel vector orientation to control the spatial distribution
of Majorana modes, without external magnetic fields. Crucially, the absence of
net magnetization in fFIM-based heterostructures preserves superconductivity,
circumventing the usual trade-off between tunability and superconducting
coherence in magnetized systems. Our results establish fFIM-based
heterostructures as a versatile platform for tunable topological
superconductivity.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:43:05 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Li",
"Yu-Xuan",
""
],
[
"Liu",
"Yicheng",
""
],
[
"Liu",
"Cheng-Cheng",
""
]
] |
2504.19845 | Kangfa Cheng | Cheng Kangfa, Mao Jirong, Zhao Xiaohong, Liu Hongbang, Chen Zhegeng,
Yu Gaojin, and Chen Zhifu | Synchrotron Circular Polarization in Gamma-Ray Burst Prompt Optical
Emission: Relativistic Thermal Electron Contribution | 12 pages, 5 figures, accepted for publication in ApJ | null | null | null | astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Synchrotron circular polarization of a non-thermal power-law electron
distribution in gamma-ray bursts (GRBs) has been studied. However, some
numerical simulations have shown that the resulting distribution of electrons
is a combination of a thermal component and a non-thermal power-law component.
In this paper, we investigate synchrotron circular polarization using such a
hybrid energy distribution of relativistic thermal and nonthermal electrons
within a globally toroidal magnetic field in GRB prompt optical emission. Our
results show that compared to the solely nonthermal electron model, the
synchrotron circular polarization degree (PD) in the hybrid electron model can
vary widely in the optical band, depending on different parameters. The lower
the electron temperature, the higher the circular PD. The time-averaged
circular PD in the hybrid electron model can be higher than $\sim 1\%$ when the
electron temperature is as low as $\sim 10^{10}$ K, while in the solely
nonthermal electron model is usually lower than $\sim 1\%$. We further
calculate the radiative transfer of the circular and linear polarization in the
optical band. Our results show that both of the circular and linear PDs
decrease with the increase of optical depth, but the linear PDs decline faster
than the circular PDs. To further examine the physical mechanisms of both
radiation and particle acceleration, we expect that instruments will be capable
of measuring the circular polarization of GRB prompt optical emission in the
future.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:44:29 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Kangfa",
"Cheng",
""
],
[
"Jirong",
"Mao",
""
],
[
"Xiaohong",
"Zhao",
""
],
[
"Hongbang",
"Liu",
""
],
[
"Zhegeng",
"Chen",
""
],
[
"Gaojin",
"Yu",
""
],
[
"Zhifu",
"Chen",
""
]
] |
2504.19846 | Kazumune Hashimoto | Kazunobu Serizawa and Kazumune Hashimoto and Wataru Hashimoto and
Masako Kishida and Shigemasa Takai | Clustering-based Recurrent Neural Network Controller synthesis under
Signal Temporal Logic Specifications | submitted for publication | null | null | null | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autonomous robotic systems require advanced control frameworks to achieve
complex temporal objectives that extend beyond conventional stability and
trajectory tracking. Signal Temporal Logic (STL) provides a formal framework
for specifying such objectives, with robustness metrics widely employed for
control synthesis. Existing optimization-based approaches using neural network
(NN)-based controllers often rely on a single NN for both learning and control.
However, variations in initial states and obstacle configurations can lead to
discontinuous changes in the optimization solution, thereby degrading
generalization and control performance. To address this issue, this study
proposes a method to enhance recurrent neural network (RNN)-based control by
clustering solution trajectories that satisfy STL specifications under diverse
initial conditions. The proposed approach utilizes trajectory similarity
metrics to generate clustering labels, which are subsequently used to train a
classification network. This network assigns new initial states and obstacle
configurations to the appropriate cluster, enabling the selection of a
specialized controller. By explicitly accounting for variations in solution
trajectories, the proposed method improves both estimation accuracy and control
performance. Numerical experiments on a dynamic vehicle path planning problem
demonstrate the effectiveness of the approach.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:44:58 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Serizawa",
"Kazunobu",
""
],
[
"Hashimoto",
"Kazumune",
""
],
[
"Hashimoto",
"Wataru",
""
],
[
"Kishida",
"Masako",
""
],
[
"Takai",
"Shigemasa",
""
]
] |
2504.19847 | Kyungjae Lee | Juhan Park, Kyungjae Lee, Hyung Jin Chang, and Jungchan Cho | Foundation Model-Driven Framework for Human-Object Interaction
Prediction with Segmentation Mask Integration | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | In this work, we introduce Segmentation to Human-Object Interaction
(\textit{\textbf{Seg2HOI}}) approach, a novel framework that integrates
segmentation-based vision foundation models with the human-object interaction
task, distinguished from traditional detection-based Human-Object Interaction
(HOI) methods. Our approach enhances HOI detection by not only predicting the
standard triplets but also introducing quadruplets, which extend HOI triplets
by including segmentation masks for human-object pairs. More specifically,
Seg2HOI inherits the properties of the vision foundation model (e.g.,
promptable and interactive mechanisms) and incorporates a decoder that applies
these attributes to HOI task. Despite training only for HOI, without additional
training mechanisms for these properties, the framework demonstrates that such
features still operate efficiently. Extensive experiments on two public
benchmark datasets demonstrate that Seg2HOI achieves performance comparable to
state-of-the-art methods, even in zero-shot scenarios. Lastly, we propose that
Seg2HOI can generate HOI quadruplets and interactive HOI segmentation from
novel text and visual prompts that were not used during training, making it
versatile for a wide range of applications by leveraging this flexibility.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:45:26 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Park",
"Juhan",
""
],
[
"Lee",
"Kyungjae",
""
],
[
"Chang",
"Hyung Jin",
""
],
[
"Cho",
"Jungchan",
""
]
] |
2504.19848 | Simona Casini | Simona Casini, Pietro Ducange, Francesco Marcelloni, Lorenzo Pollini | Human-Centered AI and Autonomy in Robotics: Insights from a Bibliometric
Study | International Joint Conference on Neural Network 2025 - Accepted | null | null | null | cs.RO cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The development of autonomous robotic systems offers significant potential
for performing complex tasks with precision and consistency. Recent advances in
Artificial Intelligence (AI) have enabled more capable intelligent automation
systems, addressing increasingly complex challenges. However, this progress
raises questions about human roles in such systems. Human-Centered AI (HCAI)
aims to balance human control and automation, ensuring performance enhancement
while maintaining creativity, mastery, and responsibility. For real-world
applications, autonomous robots must balance task performance with reliability,
safety, and trustworthiness. Integrating HCAI principles enhances human-robot
collaboration and ensures responsible operation.
This paper presents a bibliometric analysis of intelligent autonomous robotic
systems, utilizing SciMAT and VOSViewer to examine data from the Scopus
database. The findings highlight academic trends, emerging topics, and AI's
role in self-adaptive robotic behaviour, with an emphasis on HCAI architecture.
These insights are then projected onto the IBM MAPE-K architecture, with the
goal of identifying how these research results map into actual robotic
autonomous systems development efforts for real-world scenarios.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:45:48 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Casini",
"Simona",
""
],
[
"Ducange",
"Pietro",
""
],
[
"Marcelloni",
"Francesco",
""
],
[
"Pollini",
"Lorenzo",
""
]
] |
2504.19849 | Chick Woodward | Charles E. Woodward, Dominique Bockelee-Morvan, David E. Harker,
Michael S.P. Kelley, Nathan X. Roth, Diane H. Wooden, Stefanie N. Milam | A JWST Study of the Remarkable Oort Cloud Comet C/2017 K2 (PanSTARRS) | 35 page, 26 figures, 4 tables, 1 appendix, PSJ accepted | null | null | null | astro-ph.EP astro-ph.IM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Comets, during their journeys into the inner solar system, deliver volatile
gases, organics, and particulates into their comae that provide crucial
information for assessing the physico-chemical conditions in the outer disk
from which they formed. Here we present observational and modeling results of a
JWST NIRSpec and MIRI MRS integral-field-unit (IFU) spatial-spectral study of
the inner coma of the Oort Cloud comet C/2017 K2 (PanSTARRS) at a heliocentric
distance of 2.35 au. We find the comet is hyperactive (water ice active
fraction greater than or equal to 86%), with a nucleus radius of $<$4.2 km,
exhibiting strong emission from H$_{2}$O, $^{12}$CO, $^{13}$CO, and CO$_{2}$ as
well as CN, H$_2$CO, CH$_3$OH, CH$_4$, C$_2$H$_6$, HCN, NH$_2$, and OH prompt
emission. The water ortho-to-para ratio is greater than or equal to 2.75. The
modeled dust composition (relative mass fraction of the sub-micron grains) in
the coma is dominated by amorphous carbon ($\simeq 25$%), amorphous Mg:Fe
olivine ($\simeq 19$%), amorphous Mg:Fe pyroxene ($\simeq 16$%), and Mg-rich
crystalline olivine ($\simeq 39$%) and the crystalline mass fraction of the
sub-micron grains in the coma is, $f_{cryst} \simeq 0.384 \pm 0.065$. Analysis
of residuals in 3 to 8 $\mu$m region of the spectral energy distribution
strongly suggests the presence of polycyclic aromatic hydrocarbon (PAHs)
species in the coma.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:45:55 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Woodward",
"Charles E.",
""
],
[
"Bockelee-Morvan",
"Dominique",
""
],
[
"Harker",
"David E.",
""
],
[
"Kelley",
"Michael S. P.",
""
],
[
"Roth",
"Nathan X.",
""
],
[
"Wooden",
"Diane H.",
""
],
[
"Milam",
"Stefanie N.",
""
]
] |
2504.19850 | Kyo Gerrits | Kyo Gerrits and Ana Guerberof-Arenas | To MT or not to MT: An eye-tracking study on the reception by Dutch
readers of different translation and creativity levels | This paper has been accepted to the MT Summit 2025 to be held in
Geneva on June 23-27 2025 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This article presents the results of a pilot study involving the reception of
a fictional short story translated from English into Dutch under four
conditions: machine translation (MT), post-editing (PE), human translation (HT)
and original source text (ST). The aim is to understand how creativity and
errors in different translation modalities affect readers, specifically
regarding cognitive load. Eight participants filled in a questionnaire, read a
story using an eye-tracker, and conducted a retrospective think-aloud (RTA)
interview. The results show that units of creative potential (UCP) increase
cognitive load and that this effect is highest for HT and lowest for MT; no
effect of error was observed. Triangulating the data with RTAs leads us to
hypothesize that the higher cognitive load in UCPs is linked to increases in
reader enjoyment and immersion. The effect of translation creativity on
cognitive load in different translation modalities at word-level is novel and
opens up new avenues for further research. All the code and data are available
at https://github.com/INCREC/Pilot_to_MT_or_not_to_MT
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:45:56 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Gerrits",
"Kyo",
""
],
[
"Guerberof-Arenas",
"Ana",
""
]
] |
2504.19851 | Andre Schreiber | Andre Schreiber and Katherine Driggs-Campbell | Do You Know the Way? Human-in-the-Loop Understanding for Fast
Traversability Estimation in Mobile Robotics | Accepted by RA-L. Code is available at
https://github.com/andreschreiber/CHUNGUS | null | 10.1109/LRA.2025.3563819 | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | The increasing use of robots in unstructured environments necessitates the
development of effective perception and navigation strategies to enable field
robots to successfully perform their tasks. In particular, it is key for such
robots to understand where in their environment they can and cannot travel -- a
task known as traversability estimation. However, existing geometric approaches
to traversability estimation may fail to capture nuanced representations of
traversability, whereas vision-based approaches typically either involve
manually annotating a large number of images or require robot experience. In
addition, existing methods can struggle to address domain shifts as they
typically do not learn during deployment. To this end, we propose a
human-in-the-loop (HiL) method for traversability estimation that prompts a
human for annotations as-needed. Our method uses a foundation model to enable
rapid learning on new annotations and to provide accurate predictions even when
trained on a small number of quickly-provided HiL annotations. We extensively
validate our method in simulation and on real-world data, and demonstrate that
it can provide state-of-the-art traversability prediction performance.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:46:51 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Schreiber",
"Andre",
""
],
[
"Driggs-Campbell",
"Katherine",
""
]
] |
2504.19853 | Marcelo Barraza-Alfaro | Marcelo Barraza-Alfaro, Mario Flock, William B\'ethune, Richard
Teague, Jaehan Bae, Myriam Benisty, Gianni Cataldi, Pietro Curone, Ian
Czekala, Stefano Facchini, Daniele Fasano, Misato Fukagawa, Maria
Galloway-Sprietsma, Himanshi Garg, Cassandra Hall, Jane Huang, John D. Ilee,
Andr\'es F. Izquierdo, Kazuhiro Kanagawa, Eric W. Koch, Geoffroy Lesur,
Cristiano Longarini, Ryan A. Loomis, Ryuta Orihara, Christophe Pinte, Daniel
J. Price, Giovanni Rosotti, Jochen Stadler, Gaylor Wafflard-Fernandez, Andrew
J. Winter, Lisa W\"olfer, Hsi-Wei Yen, Tomohiro C. Yoshida, Brianna Zawadzki | exoALMA. XVI. Predicting Signatures of Large-scale Turbulence in
Protoplanetary Disks | This paper is part of the exoALMA Focus Issue of The Astrophysical
Journal Letters. 29 pages, 11 figures and 1 table | null | null | null | astro-ph.EP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Turbulent gas motions drive planet formation and protoplanetary disk
evolution. However, empirical constraints on turbulence are scarce, halting our
understanding of its nature. Resolving signatures of the large-scale
perturbations driven by disk instabilities may reveal clues on the origin of
turbulence in the outer regions of planet-forming disks. We aim to predict the
observational signatures of such large-scale flows, as they would appear in
high-resolution Atacama Large Millimeter/submillimeter Array observations of CO
rotational lines, such as those conducted by the exoALMA Large Program.
Post-processing 3D numerical simulations, we explored the observational
signatures produced by three candidate (magneto-)hydrodynamical instabilities
to operate in the outer regions of protoplanetary disks: the vertical shear
instability (VSI), the magneto-rotational instability (MRI), and the
gravitational instability (GI). We found that exoALMA-quality observations
should capture signatures of the large-scale motions induced by these
instabilities. Mainly, flows with ring, arc, and spiral morphologies are
apparent in the residuals of synthetic velocity centroid maps. A qualitative
comparison between our predictions and the perturbations recovered from exoALMA
data suggests the presence of two laminar disks and a scarcity of ring- and
arc-like VSI signatures within the sample. Spiral features produced by the MRI
or the GI are still plausible in explaining observed disk perturbations.
Supporting these scenarios requires further methodically comparing the
predicted perturbations and the observed disks' complex dynamic structure.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:47:30 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Barraza-Alfaro",
"Marcelo",
""
],
[
"Flock",
"Mario",
""
],
[
"Béthune",
"William",
""
],
[
"Teague",
"Richard",
""
],
[
"Bae",
"Jaehan",
""
],
[
"Benisty",
"Myriam",
""
],
[
"Cataldi",
"Gianni",
""
],
[
"Curone",
"Pietro",
""
],
[
"Czekala",
"Ian",
""
],
[
"Facchini",
"Stefano",
""
],
[
"Fasano",
"Daniele",
""
],
[
"Fukagawa",
"Misato",
""
],
[
"Galloway-Sprietsma",
"Maria",
""
],
[
"Garg",
"Himanshi",
""
],
[
"Hall",
"Cassandra",
""
],
[
"Huang",
"Jane",
""
],
[
"Ilee",
"John D.",
""
],
[
"Izquierdo",
"Andrés F.",
""
],
[
"Kanagawa",
"Kazuhiro",
""
],
[
"Koch",
"Eric W.",
""
],
[
"Lesur",
"Geoffroy",
""
],
[
"Longarini",
"Cristiano",
""
],
[
"Loomis",
"Ryan A.",
""
],
[
"Orihara",
"Ryuta",
""
],
[
"Pinte",
"Christophe",
""
],
[
"Price",
"Daniel J.",
""
],
[
"Rosotti",
"Giovanni",
""
],
[
"Stadler",
"Jochen",
""
],
[
"Wafflard-Fernandez",
"Gaylor",
""
],
[
"Winter",
"Andrew J.",
""
],
[
"Wölfer",
"Lisa",
""
],
[
"Yen",
"Hsi-Wei",
""
],
[
"Yoshida",
"Tomohiro C.",
""
],
[
"Zawadzki",
"Brianna",
""
]
] |
2504.19854 | Soujanya Poria | Chia-Yu Hung, Qi Sun, Pengfei Hong, Amir Zadeh, Chuan Li, U-Xuan Tan,
Navonil Majumder, Soujanya Poria | NORA: A Small Open-Sourced Generalist Vision Language Action Model for
Embodied Tasks | null | null | null | null | cs.RO cs.AI cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Existing Visual-Language-Action (VLA) models have shown promising performance
in zero-shot scenarios, demonstrating impressive task execution and reasoning
capabilities. However, a significant challenge arises from the limitations of
visual encoding, which can result in failures during tasks such as object
grasping. Moreover, these models typically suffer from high computational
overhead due to their large sizes, often exceeding 7B parameters. While these
models excel in reasoning and task planning, the substantial computational
overhead they incur makes them impractical for real-time robotic environments,
where speed and efficiency are paramount. To address the limitations of
existing VLA models, we propose NORA, a 3B-parameter model designed to reduce
computational overhead while maintaining strong task performance. NORA adopts
the Qwen-2.5-VL-3B multimodal model as its backbone, leveraging its superior
visual-semantic understanding to enhance visual reasoning and action grounding.
Additionally, our \model{} is trained on 970k real-world robot demonstrations
and equipped with the FAST+ tokenizer for efficient action sequence generation.
Experimental results demonstrate that NORA outperforms existing large-scale VLA
models, achieving better task performance with significantly reduced
computational overhead, making it a more practical solution for real-time
robotic autonomy.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:47:34 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Hung",
"Chia-Yu",
""
],
[
"Sun",
"Qi",
""
],
[
"Hong",
"Pengfei",
""
],
[
"Zadeh",
"Amir",
""
],
[
"Li",
"Chuan",
""
],
[
"Tan",
"U-Xuan",
""
],
[
"Majumder",
"Navonil",
""
],
[
"Poria",
"Soujanya",
""
]
] |
2504.19857 | Kai Zehmisch | Florian Buck and Kai Zehmisch | Connected sums of Brieskorn contact $5$-spheres | null | null | null | null | math.SG | http://creativecommons.org/licenses/by/4.0/ | In dimension $5$, the contact connected sum of Brieskorn contact spheres is,
in general, not a Brieskorn contact sphere.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:49:07 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Buck",
"Florian",
""
],
[
"Zehmisch",
"Kai",
""
]
] |
2504.19858 | Pablo Romero | Pablo Romero | Existence of most reliable two-terminal graphs with distance constraints | null | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A two-terminal graph is a simple graph equipped with two distinguished
vertices, called terminals. Let $T_{n,m}$ be the class consisting of all
nonisomorphic two-terminal graphs on $n$ vertices and $m$ edges. Let $G$ be any
two-terminal graph in $T_{n,m}$, and let $d$ be any positive integer. For each
$\rho\in [0,1]$, the \emph{$d$-constrained two-terminal reliability of $G$ at
$\rho$}, denoted $R_G^d(\rho)$, is the probability that $G$ has some path of
length at most $d$ joining its terminals after each of its edges is
independently deleted with probability $\rho$. We say $G$ is a
\emph{$d$-uniformly most reliable two-terminal graph} ($d$-UMRTTG) if for each
$H$ in $T_{n,m}$ and every $\rho \in [0,1]$ it holds that $R_{G}^d(\rho)\geq
R_H^d(\rho)$. Previous works studied the existence of $d$-UMRTTG in $T_{n,m}$
when $d$ is greater than or equal to $n-1$, or equivalently, when the distance
constraint is dropped. In this work, a characterization of all $1$-UMRTTGs and
$2$-UMRTTGs is given. Then, it is proved that there exists a unique $3$-UMRTTG
in $T_{n,m}$ when $n\geq 6$ and $5 \leq m \leq 2n-3$. Finally, for each $d\geq
4$ and each $n\geq 11$ it is proved that there is no $d$-UMRTTG in $T_{n,m}$
when $20 \leq m \leq 3n-9$ or when $3n-5 \leq m \leq \binom{n}{2}-2$.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:49:34 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Romero",
"Pablo",
""
]
] |
2504.19859 | Edoardo Lombardo | Edoardo Lombardo | Some PDE results in Heston model with applications | null | null | null | null | math.AP cs.NA math.NA q-fin.CP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present here some results for the PDE related to the logHeston model. We
present different regularity results and prove a verification theorem that
shows that the solution produced via the Feynman-Kac theorem is the unique
viscosity solution for a wide choice of initial data (even discontinuous) and
source data. In addition, our techniques do not use Feller's condition at any
time. In the end, we prove a convergence theorem to approximate this solution
by means of a hybrid (finite differences/tree scheme) approach.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:50:06 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Lombardo",
"Edoardo",
""
]
] |
2504.19860 | Chenhan Jiang | Chenhan Jiang, Yihan Zeng, Hang Xu, Dit-Yan Yeung | CoherenDream: Boosting Holistic Text Coherence in 3D Generation via
Multimodal Large Language Models Feedback | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Score Distillation Sampling (SDS) has achieved remarkable success in
text-to-3D content generation. However, SDS-based methods struggle to maintain
semantic fidelity for user prompts, particularly when involving multiple
objects with intricate interactions. While existing approaches often address 3D
consistency through multiview diffusion model fine-tuning on 3D datasets, this
strategy inadvertently exacerbates text-3D alignment degradation. The
limitation stems from SDS's inherent accumulation of view-independent biases
during optimization, which progressively diverges from the ideal text alignment
direction. To alleviate this limitation, we propose a novel SDS objective,
dubbed as Textual Coherent Score Distillation (TCSD), which integrates
alignment feedback from multimodal large language models (MLLMs). Our TCSD
leverages cross-modal understanding capabilities of MLLMs to assess and guide
the text-3D correspondence during the optimization. We further develop
3DLLaVA-CRITIC - a fine-tuned MLLM specialized for evaluating multiview text
alignment in 3D generations. Additionally, we introduce an LLM-layout
initialization that significantly accelerates optimization convergence through
semantic-aware spatial configuration. Comprehensive evaluations demonstrate
that our framework, CoherenDream, establishes state-of-the-art performance in
text-aligned 3D generation across multiple benchmarks, including T$^3$Bench and
TIFA subset. Qualitative results showcase the superior performance of
CoherenDream in preserving textual consistency and semantic interactions. As
the first study to incorporate MLLMs into SDS optimization, we also conduct
extensive ablation studies to explore optimal MLLM adaptations for 3D
generation tasks.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:50:45 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Jiang",
"Chenhan",
""
],
[
"Zeng",
"Yihan",
""
],
[
"Xu",
"Hang",
""
],
[
"Yeung",
"Dit-Yan",
""
]
] |
2504.19861 | Jason Francis Parisi | J.F. Parisi and J. G. Clark and J. W. Berkery and C. Bowman and C. J.
Fitzpatrick and S. M. Kaye and M. Lampert | HIPED: Machine Learning Framework for Spherical Tokamak Pedestal
Prediction and Optimization | 21 pages, 18 figures | null | null | null | physics.plasm-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a Machine Learning framework, HIPED (HeIght and width Predictor
for Edge Dynamics), for predicting and optimizing pedestal and core performance
in spherical tokamak plasmas. Trained on pedestal and core datasets from the
third MAST-U campaign, HIPED provides accurate estimates of pedestal height and
width. The results reveal notable differences compared with conventional
aspect-ratio studies; for instance, a simple power-law relation between
pedestal width and height has very low accuracy. Instead, additional parameters
such as normalized plasma pressure, elongation, and Greenwald fraction
significantly improve accuracy. HIPED can also be trained only on `control room
parameters' to inform experimentalists of which controllable parameters to
adjust for improving core-integrated performance. The framework further
includes a multi-objective optimization scheme that helps guide experimental
planning and optimization. We find Pareto-optimal discharges with respect to
various features, including distance from edge-localized modes and normalized
plasma pressure, track their parameter trajectories over time, and identify the
control room parameters required for these Pareto-optimal discharges. This
provides a framework for systematically optimizing core and edge performance
according to different experimental priorities.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:51:55 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Parisi",
"J. F.",
""
],
[
"Clark",
"J. G.",
""
],
[
"Berkery",
"J. W.",
""
],
[
"Bowman",
"C.",
""
],
[
"Fitzpatrick",
"C. J.",
""
],
[
"Kaye",
"S. M.",
""
],
[
"Lampert",
"M.",
""
]
] |
2504.19862 | Li Mingjin | Mingjin Li, Jianren Long, Pengcheng Wu | Two weights inequality for Hankel operators on weighted Bergman spaces
induced by radial weights | null | null | null | null | math.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The two weights inequality for Hankel operators $$\|H_f^\omega
(\cdot)\|_{L_\eta^q}\leq C \|\cdot\|_{A_v^p},$$ induced by some radial weights
under the regular assumptions is considered, the boundedness and compactness of
Hankel operators $H^\omega_f$ is characterized for $1<p,q<\infty$ and
$\omega\neq v\neq \eta\in \mathcal{R}$.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:54:17 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Li",
"Mingjin",
""
],
[
"Long",
"Jianren",
""
],
[
"Wu",
"Pengcheng",
""
]
] |
2504.19863 | Daniel Kienzle | Daniel Kienzle, Robin Sch\"on, Rainer Lienhart, Shin'Ichi Satoh | Towards Ball Spin and Trajectory Analysis in Table Tennis Broadcast
Videos via Physically Grounded Synthetic-to-Real Transfer | To be published in 2025 IEEE/CVF International Conference on Computer
Vision and Pattern Recognition Workshops (CVPRW) | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Analyzing a player's technique in table tennis requires knowledge of the
ball's 3D trajectory and spin. While, the spin is not directly observable in
standard broadcasting videos, we show that it can be inferred from the ball's
trajectory in the video. We present a novel method to infer the initial spin
and 3D trajectory from the corresponding 2D trajectory in a video. Without
ground truth labels for broadcast videos, we train a neural network solely on
synthetic data. Due to the choice of our input data representation, physically
correct synthetic training data, and using targeted augmentations, the network
naturally generalizes to real data. Notably, these simple techniques are
sufficient to achieve generalization. No real data at all is required for
training. To the best of our knowledge, we are the first to present a method
for spin and trajectory prediction in simple monocular broadcast videos,
achieving an accuracy of 92.0% in spin classification and a 2D reprojection
error of 0.19% of the image diagonal.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:55:12 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Kienzle",
"Daniel",
""
],
[
"Schön",
"Robin",
""
],
[
"Lienhart",
"Rainer",
""
],
[
"Satoh",
"Shin'Ichi",
""
]
] |
2504.19864 | Prasun Roychowdhury | Avas Banerjee, Riju Basak, and Prasun Roychowdhury | Extremizer Stability of Higher-order Hardy--Rellich inequalities for
Baouendi--Grushin vector fields | 35 pages | null | null | null | math.AP math.CA math.FA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In this paper, we improve the $L^p$-Rellich and Hardy-Rellich inequalities in
the setting of radial Baouendi--Grushin vector fields. We establish an identity
relating the subcritical and critical Hardy inequalities, thereby demonstrating
their equivalence. Moreover, we obtain improved versions of these inequalities
via an analysis of extremizer stability. In the higher-order setting, we derive
Hardy-Rellich type inequalities involving all radial operators in the Grushin
framework and prove that all resulting constants are sharp. Finally, for the
$L^2$-higher-order cases, we compute exact remainder terms by establishing
identities rather than inequalities.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:56:12 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Banerjee",
"Avas",
""
],
[
"Basak",
"Riju",
""
],
[
"Roychowdhury",
"Prasun",
""
]
] |
2504.19865 | Ion Nechita | Tim Achenbach, Andreas Bluhm, Leevi Lepp\"aj\"arvi, Ion Nechita,
Martin Pl\'avala | Factorization of multimeters: a unified view on nonclassical quantum
phenomena | null | null | null | null | quant-ph math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum theory exhibits various nonclassical features, such as measurement
incompatibility, contextuality, steering, and Bell nonlocality, which
distinguish it from classical physics. These phenomena are often studied
separately, but they possess deep interconnections. This work introduces a
unified mathematical framework based on commuting diagrams that unifies them.
By representing collections of measurements (multimeters) as maps to the set of
column-stochastic matrices, we show that measurement compatibility and
simulability correspond to specific factorizations of these maps through
intermediate systems. We apply this framework to put forward connections
between different nonclassical notions and provide factorization-based
characterizations for steering assemblages and Bell correlations, including a
perspective on the CHSH inequality witnessing measurement incompatibility.
Furthermore, we investigate robustness to noise of nonlocal features by
examining factorization conditions for maps involving noisy state spaces,
providing geometric criteria for when noisy multimeters can be simulated by
simpler measurement settings.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:57:46 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Achenbach",
"Tim",
""
],
[
"Bluhm",
"Andreas",
""
],
[
"Leppäjärvi",
"Leevi",
""
],
[
"Nechita",
"Ion",
""
],
[
"Plávala",
"Martin",
""
]
] |
2504.19866 | Alessandro Mazzi | A. Mazzi, J. S. Thomsen, A. Miglio, K. Brogaard, L. Girardi, D.
Bossini, M. Matteuzzi, W. E. van Rossem | DUETS: Setting expectations for asteroseismic binaries and binary
products with synthetic populations | 15 pages, 16 figures, 3 appendix pages. Accepted for publication in
A&A | null | null | null | astro-ph.SR astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Asteroseismology gives us the opportunity to better characterize binaries and
their products, and shed light on their role in Galactic populations. We
estimate occurrence rates, mass distributions, and evolutionary states of
asteroseismic binaries, exhibiting solar-like oscillations from both
components, and of products of binary interactions with detectable solar-like
oscillations. Additionally, we explore the effects of mass accretion or loss on
apparent age-metallicity relations. Using the TRILEGAL code, we simulate
Kepler's field of view adopting the Eggleton and Moe & Di Stefano distributions
of initial binary parameters, and run an additional simulation with
non-interacting binaries for comparison. We find that asteroseismic binaries
require an initial mass ratio close to one, and even small mass transfer events
can prevent the detection of oscillations from both components. The fraction of
asteroseismic binaries for red giant stars with detectable oscillations ranges
from 0.5\% for non-interacting binaries to a minimum of 0.06% when taking
interactions into account. Moreover, asteroseismic binaries composed of two red
clump stars are not expected at separations $\leq 500~\text{R}_\odot$ due to
the interplay of stellar evolution and binary interactions. Finally, we
estimate that at least ~1% of the Kepler red giants with detectable
oscillations have undergone significant mass accretion or loss, potentially
affecting Galactic age-metallicity relations, although there is a strong
dependence on the assumed initial binary parameters. Comparing predicted and
observed asteroseismic binaries, as well as over- and under-massive stars,
offers a way to constrain key binary evolution assumptions, and to reduce
uncertainties in mass-transfer modeling.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:59:43 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Mazzi",
"A.",
""
],
[
"Thomsen",
"J. S.",
""
],
[
"Miglio",
"A.",
""
],
[
"Brogaard",
"K.",
""
],
[
"Girardi",
"L.",
""
],
[
"Bossini",
"D.",
""
],
[
"Matteuzzi",
"M.",
""
],
[
"van Rossem",
"W. E.",
""
]
] |
2504.19867 | Ke Hong | Ke Hong, Lufang Chen, Zhong Wang, Xiuhong Li, Qiuli Mao, Jianping Ma,
Chao Xiong, Guanyu Wu, Buhe Han, Guohao Dai, Yun Liang, Yu Wang | semi-PD: Towards Efficient LLM Serving via Phase-Wise Disaggregated
Computation and Unified Storage | 18 pages, 16 figures | null | null | null | cs.CL cs.DC cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Existing large language model (LLM) serving systems fall into two categories:
1) a unified system where prefill phase and decode phase are co-located on the
same GPU, sharing the unified computational resource and storage, and 2) a
disaggregated system where the two phases are disaggregated to different GPUs.
The design of the disaggregated system addresses the latency interference and
sophisticated scheduling issues in the unified system but leads to storage
challenges including 1) replicated weights for both phases that prevent
flexible deployment, 2) KV cache transfer overhead between the two phases, 3)
storage imbalance that causes substantial wasted space of the GPU capacity, and
4) suboptimal resource adjustment arising from the difficulties in migrating KV
cache. Such storage inefficiency delivers poor serving performance under high
request rates.
In this paper, we identify that the advantage of the disaggregated system
lies in the disaggregated computation, i.e., partitioning the computational
resource to enable the asynchronous computation of two phases. Thus, we propose
a novel LLM serving system, semi-PD, characterized by disaggregated computation
and unified storage. In semi-PD, we introduce a computation resource controller
to achieve disaggregated computation at the streaming multi-processor (SM)
level, and a unified memory manager to manage the asynchronous memory access
from both phases. semi-PD has a low-overhead resource adjustment mechanism
between the two phases, and a service-level objective (SLO) aware dynamic
partitioning algorithm to optimize the SLO attainment. Compared to
state-of-the-art systems, semi-PD maintains lower latency at higher request
rates, reducing the average end-to-end latency per request by 1.27-2.58x on
DeepSeek series models, and serves 1.55-1.72x more requests adhering to latency
constraints on Llama series models.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:00:03 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Hong",
"Ke",
""
],
[
"Chen",
"Lufang",
""
],
[
"Wang",
"Zhong",
""
],
[
"Li",
"Xiuhong",
""
],
[
"Mao",
"Qiuli",
""
],
[
"Ma",
"Jianping",
""
],
[
"Xiong",
"Chao",
""
],
[
"Wu",
"Guanyu",
""
],
[
"Han",
"Buhe",
""
],
[
"Dai",
"Guohao",
""
],
[
"Liang",
"Yun",
""
],
[
"Wang",
"Yu",
""
]
] |
2504.19868 | Charles Gardner | Charles H. Gardner, Andrea Isella, Hui Li, Shengtai Li, Jaehan Bae,
Marcelo Barraza-Alfaro, Myriam Benisty, Gianni Cataldi, Pietro Curone, Josh
A. Eisner, Stefano Facchini, Daniele Fasano, Mario Flock, Katherine B.
Follette, Misato Fukagawa, Maria Galloway-Sprietsma, Himanshi Garg, Cassandra
Hall, Jane Huang, John D. Ilee, Michael J. Ireland, Andr\'es F. Izquierdo,
Christopher M. Johns-Krull, Kazuhiro Kanagawa, Adam L. Kraus, Geoffroy Lesur,
Shangfei Liu, Cristiano Longarini, Ryan A. Loomis, Francois Menard, Ryuta
Orihara, Christophe Pinte, Daniel Price, Luca Ricci, Giovanni Rosotti, Steph
Sallum, Jochen Stadler, Richard Teague, Gaylor Wafflard-Fernandez, David J.
Wilner, Andrew J. Winter, Lisa W\"olfer, Hsi-Wei Yen, Tomohiro C. Yoshida,
Brianna Zawadzki, Zhaohuan Zhu | exoALMA XI: ALMA Observations and Hydrodynamic Models of LkCa 15:
Implications for Planetary Mass Companions in the Dust Continuum Cavity | 23 pages, 12 Figures. This letter was accepted for publication in The
Astrophysical Journal Letters. This paper is part of the exoALMA Focus Issue
of The Astrophysical Journal Letters | null | null | null | astro-ph.EP astro-ph.SR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the past decade, the Atacama Large Millimeter/submillimeter Array (ALMA)
has revealed a plethora of substructures in the disks surrounding young stars.
These substructures have several proposed formation mechanisms, with one
leading theory being the interaction between the disk and newly formed planets.
In this Letter, we present high angular resolution ALMA observations of
LkCa~15's disk that reveal a striking difference in dust and CO emission
morphology. The dust continuum emission shows a ring-like structure
characterized by a dust-depleted inner region of $\sim$40 au in radius.
Conversely, the CO emission is radially smoother and shows no sign of gas
depletion within the dust cavity. We compare the observations with models for
the disk-planet interaction, including radiative transfer calculation in the
dust and CO emission. This source is particularly interesting as the presence
of massive planets within the dust cavity has been suggested based on previous
NIR observations. We find that the level of CO emission observed within the
dust cavity is inconsistent with the presence of planets more massive than
Jupiter orbiting between 10-40 au. Instead, we argue that the LkCa~15 innermost
dust cavity might be created either by a chain of low-mass planets, or by other
processes that do not require the presence of planets.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:01:13 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Gardner",
"Charles H.",
""
],
[
"Isella",
"Andrea",
""
],
[
"Li",
"Hui",
""
],
[
"Li",
"Shengtai",
""
],
[
"Bae",
"Jaehan",
""
],
[
"Barraza-Alfaro",
"Marcelo",
""
],
[
"Benisty",
"Myriam",
""
],
[
"Cataldi",
"Gianni",
""
],
[
"Curone",
"Pietro",
""
],
[
"Eisner",
"Josh A.",
""
],
[
"Facchini",
"Stefano",
""
],
[
"Fasano",
"Daniele",
""
],
[
"Flock",
"Mario",
""
],
[
"Follette",
"Katherine B.",
""
],
[
"Fukagawa",
"Misato",
""
],
[
"Galloway-Sprietsma",
"Maria",
""
],
[
"Garg",
"Himanshi",
""
],
[
"Hall",
"Cassandra",
""
],
[
"Huang",
"Jane",
""
],
[
"Ilee",
"John D.",
""
],
[
"Ireland",
"Michael J.",
""
],
[
"Izquierdo",
"Andrés F.",
""
],
[
"Johns-Krull",
"Christopher M.",
""
],
[
"Kanagawa",
"Kazuhiro",
""
],
[
"Kraus",
"Adam L.",
""
],
[
"Lesur",
"Geoffroy",
""
],
[
"Liu",
"Shangfei",
""
],
[
"Longarini",
"Cristiano",
""
],
[
"Loomis",
"Ryan A.",
""
],
[
"Menard",
"Francois",
""
],
[
"Orihara",
"Ryuta",
""
],
[
"Pinte",
"Christophe",
""
],
[
"Price",
"Daniel",
""
],
[
"Ricci",
"Luca",
""
],
[
"Rosotti",
"Giovanni",
""
],
[
"Sallum",
"Steph",
""
],
[
"Stadler",
"Jochen",
""
],
[
"Teague",
"Richard",
""
],
[
"Wafflard-Fernandez",
"Gaylor",
""
],
[
"Wilner",
"David J.",
""
],
[
"Winter",
"Andrew J.",
""
],
[
"Wölfer",
"Lisa",
""
],
[
"Yen",
"Hsi-Wei",
""
],
[
"Yoshida",
"Tomohiro C.",
""
],
[
"Zawadzki",
"Brianna",
""
],
[
"Zhu",
"Zhaohuan",
""
]
] |
2504.19869 | Philipp Schmidt | Philipp Schmidt, Katarina Stanojevi\'c, Kenji Watanabe, Takashi
Taniguchi, Bernd Beschoten, Vincent Mourik, Christoph Stampfer | Anisotropic supercurrent suppression and revivals in a graphene-based
Josephson junction under in-plane magnetic fields | null | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We report on a tunable Josephson junction formed by a bilayer graphene ribbon
encapsulated in WSe$_2$ with superconducting niobium contacts. We characterize
the junction by measurements of the magnetic field induced interference
pattern, and the AC Josephson effect manifested as "Shapiro steps", examining
current dependent hysteresis and junction dynamics. The latter can be tuned by
temperature, gate voltage, and magnetic field. Finally, we examine the
evolution of the supercurrent when subjected to in-plane magnetic fields.
Notably, we observe a strong anisotropy in the supercurrent with respect to the
orientation of the in-plane magnetic field. When the field is parallel to the
current direction, the supercurrent is suppressed, and shows revivals with
increasing magnetic field, whereas it remains almost unaffected when the field
is oriented in a perpendicular direction. We suggest that this anisotropy is
caused by the dependence of supercurrent interference on the junction geometry.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:01:38 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Schmidt",
"Philipp",
""
],
[
"Stanojević",
"Katarina",
""
],
[
"Watanabe",
"Kenji",
""
],
[
"Taniguchi",
"Takashi",
""
],
[
"Beschoten",
"Bernd",
""
],
[
"Mourik",
"Vincent",
""
],
[
"Stampfer",
"Christoph",
""
]
] |
2504.19870 | Ryan Loomis | Ryan A. Loomis, Stefano Facchini, Myriam Benisty, Pietro Curone, John
D. Ilee, Gianni Cataldi, Hsi-Wei Yen, Richard Teague, Christophe Pinte, Jane
Huang, Himanshi Garg, Ryuta Orihara, Ian Czekala, Brianna Zawadzki, Sean M.
Andrews, David J. Wilner, Jaehan Bae, Marcelo Barraza-Alfaro, Daniele Fasano,
Mario Flock, Misato Fukagawa, Maria Galloway-Sprietsma, Andres F. Izquierdo,
Kazuhiro Kanagawa, Geoffroy Lesur, Cristiano Longarini, Francois Menard,
Daniel J. Price, Giovanni Rosotti, Jochen Stadler, Gaylor Wafflard-Fernandez,
Lisa Wolfer, Tomohiro C. Yoshida | exoALMA II: Data Calibration and Imaging Pipeline | 24 pages, 10 figures. This paper is accepted as part of the exoALMA
Focus Issue of The Astrophysical Journal Letters | null | null | null | astro-ph.EP astro-ph.IM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The exoALMA Large Program was designed to search for subtle kinematic
deviations from Keplerian motion, indicative of embedded planets, in high
angular and spectral resolution Band 7 observations of $^{12}$CO, $^{13}$CO and
CS emission from protoplanetary disks. This paper summarizes the calibration
and imaging pipelines used by the exoALMA collaboration. With sources ranging
in diameter from 2.4" to 13.8" when probed by $^{12}$CO, multiple antennae
configurations were required to maximally recover all spatial information
(including the ACA for 7 sources). Combining these datasets warranted
particular care in their alignment during calibration and prior to imaging, so
as not to introduce spurious features that might resemble the kinematic
deviations being investigated. Phase decoherence was found in several datasets,
which was corrected by an iterative self-calibration procedure, and we explored
the effects of the order of operations of spatial alignment, flux scaling, and
self-calibration. A number of different imaging sets were produced for the
continuum and line emission, employing an iterative masking procedure that
minimizes bias due to non-Keplerian motions in the disk.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:01:57 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Loomis",
"Ryan A.",
""
],
[
"Facchini",
"Stefano",
""
],
[
"Benisty",
"Myriam",
""
],
[
"Curone",
"Pietro",
""
],
[
"Ilee",
"John D.",
""
],
[
"Cataldi",
"Gianni",
""
],
[
"Yen",
"Hsi-Wei",
""
],
[
"Teague",
"Richard",
""
],
[
"Pinte",
"Christophe",
""
],
[
"Huang",
"Jane",
""
],
[
"Garg",
"Himanshi",
""
],
[
"Orihara",
"Ryuta",
""
],
[
"Czekala",
"Ian",
""
],
[
"Zawadzki",
"Brianna",
""
],
[
"Andrews",
"Sean M.",
""
],
[
"Wilner",
"David J.",
""
],
[
"Bae",
"Jaehan",
""
],
[
"Barraza-Alfaro",
"Marcelo",
""
],
[
"Fasano",
"Daniele",
""
],
[
"Flock",
"Mario",
""
],
[
"Fukagawa",
"Misato",
""
],
[
"Galloway-Sprietsma",
"Maria",
""
],
[
"Izquierdo",
"Andres F.",
""
],
[
"Kanagawa",
"Kazuhiro",
""
],
[
"Lesur",
"Geoffroy",
""
],
[
"Longarini",
"Cristiano",
""
],
[
"Menard",
"Francois",
""
],
[
"Price",
"Daniel J.",
""
],
[
"Rosotti",
"Giovanni",
""
],
[
"Stadler",
"Jochen",
""
],
[
"Wafflard-Fernandez",
"Gaylor",
""
],
[
"Wolfer",
"Lisa",
""
],
[
"Yoshida",
"Tomohiro C.",
""
]
] |
2504.19871 | Fraser Binns | Fraser Binns, Sudipta Ghosh | Degree-1 maps and rank inequalities in Heegaard Floer homology | 10 pages, 7 figures | null | null | null | math.GT | http://creativecommons.org/licenses/by/4.0/ | Ghosh-Sivek-Zentner constructed degree-1 maps from certain rational homology
solid tori to the twisted $I$-bundle over the Klein bottle. We show that these
maps yield rank inequalities for Heegaard Floer homology. To do so, we use
Hanselman-Rasmussen-Watson's immersed curve interpretation of bordered Floer
homology, extending their proof of a similar rank inequality corresponding to
degree-1 maps to the solid torus. Our result provides further evidence for
Kronheimer-Mowka's conjectured relationship between Heegaard Floer homology and
instanton Floer homology.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:03:39 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Binns",
"Fraser",
""
],
[
"Ghosh",
"Sudipta",
""
]
] |
2504.19872 | Xuan Ji | Xuan Ji, Richard D. Chatterjee, Brandon Park Coy, Edwin S. Kite | The Cosmic Shoreline Revisited: A Metric for Atmospheric Retention
Informed by Hydrodynamic Escape | submitted to ApJ | null | null | null | astro-ph.EP | http://creativecommons.org/licenses/by/4.0/ | The "cosmic shoreline", a semi-empirical relation that separates airless
worlds from worlds with atmospheres as proposed by Zahnle & Catling (2017), is
now guiding large-scale JWST surveys aimed at detecting rocky exoplanet
atmospheres. We expand upon this framework by revisiting the shorelines using
existing hydrodynamic escape models applied to Earth-like, Venus-like, and
steam atmospheres for rocky exoplanets, and we estimate energy-limited escape
rates for CH4 atmospheres. We determine the critical instellation required for
atmospheric retention by calculating time-integrated atmospheric mass loss. Our
analysis introduces a new metric for target selection in the Rocky Worlds DDT
and refines expectations for rocky planet atmosphere searches in Cycle 4.
Exploring initial volatile inventory ranging from 0.01% to 1% of planetary
mass, we find that its variation prevents the definition of a unique clear-cut
shoreline, though non-linear escape physics can reduce this sensitivity to
initial conditions. Additionally, uncertain distributions of high-energy
stellar evolution and planet age further blur the critical instellations for
atmospheric retention, yielding broad shorelines. Hydrodynamic escape models
find atmospheric retention is markedly more favorable for higher-mass planets
orbiting higher-mass stars, with carbon-rich atmospheres remaining plausible
for 55 Cancri e despite its extreme instellation. Dedicated modeling efforts
are needed to better constrain the escape dynamics of secondary atmospheres,
such as the role of atomic line cooling, especially for Earth-sized planets.
Finally, we illustrate how density measurements can be used to statistically
test the existence of the cosmic shorelines, emphasizing the need for more
precise mass and radius measurements.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:03:56 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Ji",
"Xuan",
""
],
[
"Chatterjee",
"Richard D.",
""
],
[
"Coy",
"Brandon Park",
""
],
[
"Kite",
"Edwin S.",
""
]
] |
2504.19873 | Zhi-Hua Li | Zhi-Hua Li | Infinite temperature transport in the strong coupling regime of a
nonintegrable quantum spin chain | 10 pages, 7 figures | null | null | null | cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study spin transport of the XXZ model with next-nearest neighbor
$\Delta_2$ terms. We compute numerically dependence of spin conductivity
$\sigma(\omega)$ on the anisotropy $\Delta$ and the ratio $r= \Delta_2/\Delta$,
in the large $\Delta$ regime. We find that, when $0<r<1$, the low-frequency
conductivity assumes an anomalous form $\sigma(\omega)\approx a \omega^2 + b
\Delta^{-2} $. In particular, when $\Delta\to\infty$ the model becomes
dynamically constrained and most states are localized. We show,
microscopically, existence of magnon bound states in the strong coupling
regime, which behave as self-generated disorders for single magnons. Based on
this quasiparticle picture, we obtain analytical scalings, which match well
with the numerical results.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:05:30 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Li",
"Zhi-Hua",
""
]
] |
2504.19874 | Majid Daliri | Amir Zandieh, Majid Daliri, Majid Hadian, Vahab Mirrokni | TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate | 25 pages | null | null | null | cs.LG cs.AI cs.DB cs.DS | http://creativecommons.org/licenses/by/4.0/ | Vector quantization, a problem rooted in Shannon's source coding theory, aims
to quantize high-dimensional Euclidean vectors while minimizing distortion in
their geometric structure. We propose TurboQuant to address both mean-squared
error (MSE) and inner product distortion, overcoming limitations of existing
methods that fail to achieve optimal distortion rates. Our data-oblivious
algorithms, suitable for online applications, achieve near-optimal distortion
rates (within a small constant factor) across all bit-widths and dimensions.
TurboQuant achieves this by randomly rotating input vectors, inducing a
concentrated Beta distribution on coordinates, and leveraging the
near-independence property of distinct coordinates in high dimensions to simply
apply optimal scalar quantizers per each coordinate. Recognizing that
MSE-optimal quantizers introduce bias in inner product estimation, we propose a
two-stage approach: applying an MSE quantizer followed by a 1-bit Quantized JL
(QJL) transform on the residual, resulting in an unbiased inner product
quantizer. We also provide a formal proof of the information-theoretic lower
bounds on best achievable distortion rate by any vector quantizer,
demonstrating that TurboQuant closely matches these bounds, differing only by a
small constant ($\approx 2.7$) factor. Experimental results validate our
theoretical findings, showing that for KV cache quantization, we achieve
absolute quality neutrality with 3.5 bits per channel and marginal quality
degradation with 2.5 bits per channel. Furthermore, in nearest neighbor search
tasks, our method outperforms existing product quantization techniques in
recall while reducing indexing time to virtually zero.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:05:35 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Zandieh",
"Amir",
""
],
[
"Daliri",
"Majid",
""
],
[
"Hadian",
"Majid",
""
],
[
"Mirrokni",
"Vahab",
""
]
] |
2504.19875 | Yifan Zhu | Xiaowei Ge, Yifan Zhu, Dingcheng Sun, Hongli Ni, Yueming Li, Chinmayee
V. Prabhu Dessai, Ji-Xin Cheng | Fiber laser based stimulated Raman photothermal microscopy with long
working distance optics | 27 pages main text, 7 figures. 15 pages supplementary information, 11
supplementary figures | null | null | null | physics.optics physics.ins-det | http://creativecommons.org/licenses/by/4.0/ | Stimulated Raman scattering (SRS) microscopy is a highly sensitive chemical
imaging technique. However, the broader application of SRS has been limited by
two key challenges: the reliance on low-noise but bulky solid-state laser
sources and stringent sample requirements necessitated by high numerical
aperture (NA) optics. Here, we present a fiber laser based stimulated Raman
photothermal (SRP) microscope that addresses these limitations. While
appreciating the portability and compactness of a noisy source, fiber laser SRP
enables a two-order-of-magnitude improvement in signal to noise ratio over
fiber laser SRS without balance detection. Furthermore, with the use of low NA,
long working distance optics for signal collection, SRP expands the allowed
sample space from millimeters to centimeters, which diversifies the sample
formats to multi-well plates and thick tissues. The sensitivity and imaging
depth are further amplified by using urea for both thermal enhancement and
tissue clearance. Together, fiber laser SRP microscopy provides a robust,
user-friendly platform for diverse applications.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:05:58 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Ge",
"Xiaowei",
""
],
[
"Zhu",
"Yifan",
""
],
[
"Sun",
"Dingcheng",
""
],
[
"Ni",
"Hongli",
""
],
[
"Li",
"Yueming",
""
],
[
"Dessai",
"Chinmayee V. Prabhu",
""
],
[
"Cheng",
"Ji-Xin",
""
]
] |
2504.19876 | Mamadou Keita | Mamadou Keita and Wassim Hamidouche and Hessen Bougueffa Eutamene and
Abdelmalik Taleb-Ahmed and Abdenour Hadid | DeeCLIP: A Robust and Generalizable Transformer-Based Framework for
Detecting AI-Generated Images | null | null | null | null | cs.CV cs.CR | http://creativecommons.org/licenses/by/4.0/ | This paper introduces DeeCLIP, a novel framework for detecting AI-generated
images using CLIP-ViT and fusion learning. Despite significant advancements in
generative models capable of creating highly photorealistic images, existing
detection methods often struggle to generalize across different models and are
highly sensitive to minor perturbations. To address these challenges, DeeCLIP
incorporates DeeFuser, a fusion module that combines high-level and low-level
features, improving robustness against degradations such as compression and
blurring. Additionally, we apply triplet loss to refine the embedding space,
enhancing the model's ability to distinguish between real and synthetic
content. To further enable lightweight adaptation while preserving pre-trained
knowledge, we adopt parameter-efficient fine-tuning using low-rank adaptation
(LoRA) within the CLIP-ViT backbone. This approach supports effective zero-shot
learning without sacrificing generalization. Trained exclusively on 4-class
ProGAN data, DeeCLIP achieves an average accuracy of 89.00% on 19 test subsets
composed of generative adversarial network (GAN) and diffusion models. Despite
having fewer trainable parameters, DeeCLIP outperforms existing methods,
demonstrating superior robustness against various generative models and
real-world distortions. The code is publicly available at
https://github.com/Mamadou-Keita/DeeCLIP for research purposes.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:06:28 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Keita",
"Mamadou",
""
],
[
"Hamidouche",
"Wassim",
""
],
[
"Eutamene",
"Hessen Bougueffa",
""
],
[
"Taleb-Ahmed",
"Abdelmalik",
""
],
[
"Hadid",
"Abdenour",
""
]
] |
2504.19877 | David Forsman | David Forsman | Semi-abelian by Design: Johnstone Algebras Unifying Implication and
Division | 19 pages, Z3 verification code available at
https://doi.org/10.5281/zenodo.15292851 | null | null | null | math.LO math.CT math.RA | http://creativecommons.org/licenses/by/4.0/ | Johnstone demonstrated that Heyting semilattices form a semi-abelian category
via a specific triple of terms. Inspired by this work, we introduce
\emph{Johnstone algebras} or J-algebras. The algebraic $(*,\to,e)$-theory $J$
of arities $(2,2,0)$ consists of three axioms carefully chosen to ensure
protomodularity in alignment with Johnstone's terms. Johnstone algebras
generalize well-known structures such as groups (division) and Heyting
semilattices (implication) providing a unified framework within the
well-behaved setting of semi-abelian categories.
We present two primary contributions. First, we identify the M-axiom, \[
(t(x,y)\to x)\to (t(x,y)\to z) \approx x\to z, \text{ where }t(x,y) = (x\to
y)\to y. \] The M-axiom is satisfied by residuated Johnstone algebras, and it
can be considered a weakening of the H-axiom to comparable elements. We show
that $t(x,y)$ defines a \emph{relative closure term} in MBC-algebras, and it
implies that MBC-algebras form a variety of algebras, thereby generalizing the
corresponding theorem related to HBCK-algebras. Second, we prove several no-go
results, demonstrating that balanced theories or theories admitting
non-discrete monotone or inflationary algebras cannot possess Malcev terms.
Together, these results establish Johnstone algebras as significant
structures that achieve desirable categorical properties by carefully
integrating both logical and symmetric features, while closely avoiding the
constraints imposed by our no-go results.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:06:42 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Forsman",
"David",
""
]
] |
2504.19878 | Patrick Iff | Patrick Iff, Maciej Besta, Torsten Hoefler | FoldedHexaTorus: An Inter-Chiplet Interconnect Topology for
Chiplet-based Systems using Organic and Glass Substrates | null | null | null | null | cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Chiplet-based systems are rapidly gaining traction in the market. Two
packaging options for such systems are the established organic substrates and
the emerging glass substrates. These substrates are used to implement the
inter-chiplet interconnect (ICI), which is crucial for overall system
performance. To guide the development of ICIs, we introduce three design
principles for ICI network topologies on organic and glass substrates. Based on
our design principles, we propose the novel FoldedHexaTorus network topology.
Our evaluation shows that the FoldedHexaTorus achieves significantly higher
throughput than state-of-the-art topologies while maintaining low latency.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:07:36 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Iff",
"Patrick",
""
],
[
"Besta",
"Maciej",
""
],
[
"Hoefler",
"Torsten",
""
]
] |
2504.19879 | Adrija Datta | Adrija Datta (Department of Earth Sciences, Indian Institute of
Technology, Gandhinagar, Gujarat, India), Sarth Dubey (Department of Computer
Science and Engineering, Indian Institute of Technology, Gandhinagar,
Gujarat, India), Tarik C. Gouhier (Department of Marine and Environmental
Sciences, Marine Science Center, Northeastern University, Nahant, MA, USA),
Auroop R. Ganguly (Sustainability and Data Sciences Lab, Department of Civil
and Environmental Engineering, Northeastern University, Boston, MA, USA, The
Institute for Experiential AI and Roux Institute, Northeastern University,
Boston, MA, USA, Pacific Northwest National Laboratory, Richland, WA, USA),
Udit Bhatia (Department of Computer Science and Engineering, Indian Institute
of Technology, Gandhinagar, Gujarat, India, Department of Civil Engineering,
Indian Institute of Technology, Gandhinagar, Gujarat, India) | Warming demands extensive tropical but minimal temperate management in
plant-pollinator networks | null | null | null | null | q-bio.PE | http://creativecommons.org/licenses/by/4.0/ | Anthropogenic warming impacts ecological communities and disturbs species
interactions, particularly in temperature sensitive plant pollinator networks.
While previous assessments indicate that rising mean temperatures and shifting
temporal variability universally elevate pollinator extinction risk, many
studies often overlook how plant-pollinator networks of different ecoregions
require distinct management approaches. Here, we integrate monthly near-surface
temperature projections from various Shared Socioeconomic Pathways of CMIP6
Earth System Models with region-specific thermal performance parameters to
simulate population dynamics in 11 plant pollinator networks across tropical,
temperate, and Mediterranean ecosystems. Our results show that tropical
networks, already near their thermal limits, face pronounced (50 percent)
pollinator declines under high-emissions scenarios (SSP5-8.5). Multi-species
management targeting keystone plants emerges as a critical strategy for
stabilizing these high risk tropical systems, boosting both pollinator
abundance and evenness. In contrast, temperate networks remain well below
critical temperature thresholds, with minimal (5 percent) pollinator declines
and negligible gains from any intensive management strategy. These findings
challenge single-species models and uniform-parameter frameworks, which
consistently underestimate tropical vulnerability while overestimating
temperate risk. We demonstrate that explicitly incorporating complex network
interactions, region-specific thermal tolerances, and targeted multi species
interventions is vital for maintaining pollination services. By revealing when
and where limited interventions suffice versus extensive management becomes
indispensable, our study provides a clear blueprint for adaptive, ecosystem
specific management under accelerating climate change.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:09:30 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Datta",
"Adrija",
"",
"Department of Earth Sciences, Indian Institute of\n Technology, Gandhinagar, Gujarat, India"
],
[
"Dubey",
"Sarth",
"",
"Department of Computer\n Science and Engineering, Indian Institute of Technology, Gandhinagar,\n Gujarat, India"
],
[
"Gouhier",
"Tarik C.",
"",
"Department of Marine and Environmental\n Sciences, Marine Science Center, Northeastern University, Nahant, MA, USA"
],
[
"Ganguly",
"Auroop R.",
"",
"Sustainability and Data Sciences Lab, Department of Civil\n and Environmental Engineering, Northeastern University, Boston, MA, USA, The\n Institute for Experiential AI and Roux Institute, Northeastern University,\n Boston, MA, USA, Pacific Northwest National Laboratory, Richland, WA, USA"
],
[
"Bhatia",
"Udit",
"",
"Department of Computer Science and Engineering, Indian Institute\n of Technology, Gandhinagar, Gujarat, India, Department of Civil Engineering,\n Indian Institute of Technology, Gandhinagar, Gujarat, India"
]
] |
2504.19880 | Edson Ribeiro Alvares | Edson Ribeiro Alvares, Clezio Aparecido Braga, Sonia Trepode, Heily
Wagner | On the global dimension three endomorphism algebras of the minimal
generator-cogenerator | 25 pages | null | null | null | math.RT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The main goal of this paper is to study the class of algebras for which the
global dimension of the endomorphism ring of the generator-cogenerator, given
by the sum of the projective and injective modules, is equal to three. We will
refer to these algebras as representation-hereditary algebras. We show that
these algebras are torsionless-finite, as defined by Ringel. These algebras do
not necessarily have finite global dimension; however, when there is no
non-zero morphism from an injective to a projective module, they have global
dimension less than or equal to two, with some additional homological
properties. By utilizing the general framework provided by the study of the
representation dimension of an algebra, we present further homological
consequences. In the case where these algebras are tame quasi-tilted algebras,
we prove that they belong to certain classes of tilted algebras. Although not
all tilted algebras are representation-hereditary, we provide sufficient
conditions for them to be representation-hereditary.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:09:58 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Alvares",
"Edson Ribeiro",
""
],
[
"Braga",
"Clezio Aparecido",
""
],
[
"Trepode",
"Sonia",
""
],
[
"Wagner",
"Heily",
""
]
] |
2504.19881 | Hubert P. H. Shum | Claire Warwick, Andrew Beresford, Soazig Casteau, Hubert P. H. Shum,
Dan Smith, Francis Xiatian Zhang | Using Fixed and Mobile Eye Tracking to Understand How Visitors View Art
in a Museum: A Study at the Bowes Museum, County Durham, UK | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The following paper describes a collaborative project involving researchers
at Durham University, and professionals at the Bowes Museum, Barnard Castle,
County Durham, UK, during which we used fixed and mobile eye tracking to
understand how visitors view art. Our study took place during summer 2024 and
builds on work presented at DH2017 (Bailey-Ross et al., 2017). Our
interdisciplinary team included researchers from digital humanities,
psychology, art history and computer science, working in collaboration with
professionals from the museum. We used fixed and mobile eye tracking to
understand how museum visitors view art in a physical gallery setting. This
research will enable us to make recommendations about how the Museum's
collections could be more effectively displayed, encouraging visitors to engage
with them more fully.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:12:30 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Warwick",
"Claire",
""
],
[
"Beresford",
"Andrew",
""
],
[
"Casteau",
"Soazig",
""
],
[
"Shum",
"Hubert P. H.",
""
],
[
"Smith",
"Dan",
""
],
[
"Zhang",
"Francis Xiatian",
""
]
] |
2504.19882 | Zhuang Qi | Runhui Zhang, Sijin Zhou, Zhuang Qi | Federated Out-of-Distribution Generalization: A Causal Augmentation View | IJCNN 2025 Accepted | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning aims to collaboratively model by integrating multi-source
information to obtain a model that can generalize across all client data.
Existing methods often leverage knowledge distillation or data augmentation to
mitigate the negative impact of data bias across clients. However, the limited
performance of teacher models on out-of-distribution samples and the inherent
quality gap between augmented and original data hinder their effectiveness and
they typically fail to leverage the advantages of incorporating rich contextual
information. To address these limitations, this paper proposes a Federated
Causal Augmentation method, termed FedCAug, which employs causality-inspired
data augmentation to break the spurious correlation between attributes and
categories. Specifically, it designs a causal region localization module to
accurately identify and decouple the background and objects in the image,
providing rich contextual information for causal data augmentation.
Additionally, it designs a causality-inspired data augmentation module that
integrates causal features and within-client context to generate counterfactual
samples. This significantly enhances data diversity, and the entire process
does not require any information sharing between clients, thereby contributing
to the protection of data privacy. Extensive experiments conducted on three
datasets reveal that FedCAug markedly reduces the model's reliance on
background to predict sample labels, achieving superior performance compared to
state-of-the-art methods.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:13:48 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Zhang",
"Runhui",
""
],
[
"Zhou",
"Sijin",
""
],
[
"Qi",
"Zhuang",
""
]
] |
2504.19883 | Denis Sergeev | Denis E. Sergeev, James W. McDermott, Lottie Woods, Marrick Braam,
Jake K. Eager-Nash, Ian A. Boutle | Lightning activity on a tidally locked terrestrial exoplanet in
storm-resolving simulations for a range of surface pressures | 12 pages, 6 figures; Submitted for publication in the Monthly Notices
of the Royal Astronomical Society | null | null | null | astro-ph.EP physics.ao-ph | http://creativecommons.org/licenses/by/4.0/ | Cloudy atmospheres produce electric discharges, including lightning.
Lightning, in turn, provides sufficient energy to break down air molecules into
reactive species and thereby affects the atmospheric composition. The climate
of tidally locked rocky exoplanets orbiting M-dwarf stars may have intense and
highly localised thunderstorm activity associated with moist convection on
their day side. The distribution and structure of lightning-producing
convective clouds is shaped by various climate parameters, of which a key one
is atmospheric mass, i.e. surface air pressure. In this study, we use a global
storm-resolving climate model to predict thunderstorm occurrence for a tidally
locked exoplanet over a range of surface pressures. We compare two lightning
parameterisations: one based on ice cloud microphysics and one based on the
vertical extent of convective clouds. We find that both parameterisations
predict that the amount of lightning monotonically decreases with surface
pressure due to weaker convection and fewer ice clouds. The spatial
distribution of lightning on the planet changes with respect to the surface
pressure, responding to the changes in the large-scale circulation and the
vertical stratification of the atmosphere. Our study provides revised,
high-resolution estimates for lightning activity on a tidally locked Earth-like
exoplanet, with implications for global atmospheric chemistry.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:13:49 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sergeev",
"Denis E.",
""
],
[
"McDermott",
"James W.",
""
],
[
"Woods",
"Lottie",
""
],
[
"Braam",
"Marrick",
""
],
[
"Eager-Nash",
"Jake K.",
""
],
[
"Boutle",
"Ian A.",
""
]
] |
2504.19884 | Amit Kashi | Bhawna Mukhija, Amit Kashi | Accretion and Recovery in Giant Eruptions of Massive Stars | Accepted for publication in ApJ | null | null | null | astro-ph.SR astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | Giant Eruptions (GEs) are episodic high-rate mass loss events that massive
stars experience in the late stage of evolutions before exploding as a
core-collapse supernova. If it occurs in a binary system, the companion star
can accrete part of the mass. We use numerical simulations to analyze how the
companion responds to accretion and how its structure and evolution are
altered. We run a grid of massive stars with masses from $20~\rm M_{\odot}$ to
$60~\rm M_{\odot}$, and accretion rates from $\rm 10^{-4}$ to $\rm
0.1~M_{\odot}~\rm yr^{-1}$, over a duration of $20$ yrs. For accretion rates
$\rm \lesssim 0.01~M_{\odot}~\rm yr^{-1}$ the star remains on the hotter side
of the HR diagram with a minor increase in luminosity without expanding, as the
accretion timescale exceeds the thermal time scale by a larger factor. Mass
loss through stellar winds leads to a minor drop in luminosity shortly after
the accretion phase as the star enters the recovery phase. For $\rm \gtrsim
0.01~M_{\odot}~\rm yr^{-1}$ the companion star experiences a sudden increase in
luminosity by about one order of magnitude, inflates, and cools. Under the
accreted gas layer the star retains its structure and continues to eject
radiation-driven wind during the recovery phase, namely the time it takes to
regain equilibrium. Eventually, the accreted material mixes with the inner
layers of the star, and the star continues to evolve as a more massive star.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:14:36 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Mukhija",
"Bhawna",
""
],
[
"Kashi",
"Amit",
""
]
] |
2504.19885 | Elie Attal | Eduardo Abi Jaber and Elie Attal | Simulating integrated Volterra square-root processes and Volterra Heston
models via Inverse Gaussian | null | null | null | null | q-fin.MF math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a novel simulation scheme, iVi (integrated Volterra implicit),
for integrated Volterra square-root processes and Volterra Heston models based
on the Inverse Gaussian distribution. The scheme is designed to handle $L^1$
kernels with singularities by relying solely on integrated kernel quantities,
and it preserves the non-decreasing property of the integrated process. We
establish weak convergence of the iVi scheme by reformulating it as a
stochastic Volterra equation with a measure kernel and proving a stability
result for this class of equations. Numerical results demonstrate that
convergence is achieved with very few time steps. Remarkably, for the rough
fractional kernel, unlike existing schemes, convergence seems to improve as the
Hurst index $H$ decreases and approaches $-1/2$.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:17:37 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Jaber",
"Eduardo Abi",
""
],
[
"Attal",
"Elie",
""
]
] |
2504.19886 | Riasat Sheikh | Riasat Sheikh, Takashi Toma, and Koji Tsumura | Pseudo-Nambu-Goldstone-boson Dark Matter from Three Complex Scalars | 15 pages, 5 figures | null | null | KYUSHU-HET-319, KANAZAWA-25-01 | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study explores a dark matter model in which a pseudo-Nambu-Goldstone
boson arises as a viable dark matter candidate from the spontaneous and soft
breaking of global $U(1)$ symmetries and stabilized by a residual
$\mathbb{Z}_3$ discrete symmetry. The model introduces three complex scalar
fields, singlets under the Standard Model gauge group, and charged under a dark
$U(1)_V$ gauge symmetry together with a permutative exchange symmetry among
three scalars. These features naturally suppress the dark matter--nucleon
scattering cross section by its Nambu-Goldstone boson nature. In addition to
conventional annihilation channels, the $\mathbb{Z}_3$ structure allows
semi-annihilation processes, potentially leading to new phenomenological
signatures. We analyze theoretical and experimental constraints, including
relic abundance, Higgs invisible decays, and perturbative unitarity, and
evaluate the elastic scattering cross section for boosted dark matter.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:18:29 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sheikh",
"Riasat",
""
],
[
"Toma",
"Takashi",
""
],
[
"Tsumura",
"Koji",
""
]
] |
2504.19887 | Fredrik Johansson Viklund | Klara Courteaut, Kurt Johansson, Fredrik Viklund | Planar Coulomb gas on a Jordan arc at any temperature | null | null | null | null | math.CV math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study a planar Coulomb gas confined to a sufficiently smooth Jordan arc
$\gamma$ in the complex plane, at inverse temperature $\beta > 0$. Let
\[\bar{Z}_{n}^\beta(\gamma) = Z_{n}^\beta(\gamma)/\left(2
\textrm{cap}(\gamma)\right)^{\beta n^2/2+(1-\beta/2)n}\] be the normalized
partition function. We compute the free energy as the number of particles tends
to infinity, including the constant term: \[ \lim_{n\to \infty}\log
\frac{\bar{Z}_{n}^\beta(\gamma)}{\bar{Z}_{n}^\beta([-1,1])} =
\frac{1}{24}J^A(\gamma) + \frac{1}{8} \left( \sqrt{ \frac{\beta}{2} } - \sqrt{
\frac{2}{\beta}} \right)^2 J^{F}(\gamma). \] Here $\bar{Z}_{n}^\beta([-1,1])$
is an explicit Selberg integral, $J^A(\gamma)$ is half the Loewner energy of a
certain Jordan curve associated to $\gamma$ plus a covariance term, and
$J^F(\gamma)$ is the Fekete energy, related to the zero temperature limit of
the model.
We also prove an asymptotic formula for the Laplace transform of linear
statistics for sufficiently regular test functions. As a consequence, the
centered empirical measure converges to a Gaussian field with explicit
asymptotic mean, and asymptotic variance given by the Dirichlet energy of the
bounded harmonic extension of the test function outside of the arc.
A key tool in our analysis is the arc-Grunsky operator $B$ associated to
$\gamma$, reminiscent to but different from the classical Grunsky operator. We
derive several basic properties the arc-Grunsky operator, including an estimate
analogous to the strengthened Grunsky inequality and the relation to the
Dirichlet integral. In our proofs $J^A(\gamma)$ arises from the Fredholm
determinant $\det(I+B)$ and a substantial part of the paper is devoted to
relating this to the Loewner energy.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:19:09 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Courteaut",
"Klara",
""
],
[
"Johansson",
"Kurt",
""
],
[
"Viklund",
"Fredrik",
""
]
] |
2504.19888 | Han Chen | Han Chen and Anne L. Martel | Enhancing breast cancer detection on screening mammogram using
self-supervised learning and a hybrid deep model of Swin Transformer and
Convolutional Neural Network | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Purpose: The scarcity of high-quality curated labeled medical training data
remains one of the major limitations in applying artificial intelligence (AI)
systems to breast cancer diagnosis. Deep models for mammogram analysis and mass
(or micro-calcification) detection require training with a large volume of
labeled images, which are often expensive and time-consuming to collect. To
reduce this challenge, we proposed a novel method that leverages
self-supervised learning (SSL) and a deep hybrid model, named \textbf{HybMNet},
which combines local self-attention and fine-grained feature extraction to
enhance breast cancer detection on screening mammograms.
Approach: Our method employs a two-stage learning process: (1) SSL
Pretraining: We utilize EsViT, a SSL technique, to pretrain a Swin Transformer
(Swin-T) using a limited set of mammograms. The pretrained Swin-T then serves
as the backbone for the downstream task. (2) Downstream Training: The proposed
HybMNet combines the Swin-T backbone with a CNN-based network and a novel
fusion strategy. The Swin-T employs local self-attention to identify
informative patch regions from the high-resolution mammogram, while the
CNN-based network extracts fine-grained local features from the selected
patches. A fusion module then integrates global and local information from both
networks to generate robust predictions. The HybMNet is trained end-to-end,
with the loss function combining the outputs of the Swin-T and CNN modules to
optimize feature extraction and classification performance.
Results: The proposed method was evaluated for its ability to detect breast
cancer by distinguishing between benign (normal) and malignant mammograms.
Leveraging SSL pretraining and the HybMNet model, it achieved AUC of 0.864 (95%
CI: 0.852, 0.875) on the CMMD dataset and 0.889 (95% CI: 0.875, 0.903) on the
INbreast dataset, highlighting its effectiveness.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:23:28 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Chen",
"Han",
""
],
[
"Martel",
"Anne L.",
""
]
] |
2504.19889 | Riya Sultana | Riya Sultana, Khushboo Agarwal and Veeraruna Kavitha | Queue or lounge: strategic design for strategic customer | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | Considering an M/M/1 queue with an additional lounge facility (LF), the quest
of this paper is to understand the instances when LF is an attractive option,
from customer perspective as well as from system perspective: will the
customers choose to join the queue or prefer to detour briefly to lounge? In
reality, customers do not perform complex computations for such tasks, but
instead choose based on some heuristics. We further assume that the customers
pessimistically anticipate the future congestion while making the choice. Our
analysis reveals that the customers use the LF only when the queue is too
crowded, and the lounge is relatively empty; however, strikingly, the customer
choice is more inclined towards rejection for the LF in systems with higher
traffic (load).
We also explore an optimization problem where the system determines whether
to implement an LF and what capacity it should have, while accounting for
customers' behavioral responses. Under low load conditions, the system benefits
from designing a high-capacity lounge, and the customers also prefer to use the
LF actively. Surprisingly, neither the system prefers big LF, nor the customers
prefer to use the LF profusely at high load conditions; optimal for either is
to use the LF sparingly. Thus, importantly, the strategic system and the
bounded-rational customers are not in a tug-of-war situation.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:23:30 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sultana",
"Riya",
""
],
[
"Agarwal",
"Khushboo",
""
],
[
"Kavitha",
"Veeraruna",
""
]
] |
2504.19890 | Matej Praprotnik | Nikolaos Ntarakas, Ma\v{s}a Lah, Daniel Sven\v{s}ek, Tilen Potisk,
Matej Praprotnik | Dissipative particle dynamics models of encapsulated microbubbles and
gas vesicles for biomedical ultrasound simulations | 55 pages, 18 figures, 4 tables | null | null | null | cond-mat.soft physics.bio-ph physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Ultrasound-guided drug and gene delivery (usdg) enables controlled and
spatially precise delivery of drugs and macromolecules, encapsulated in
microbubbles (embs) and submicron gas vesicles (gvs), to target areas such as
cancer tumors. It is a non-invasive, high precision, low toxicity process with
drastically reduced drug dosage. Rheological and acoustic properties of gvs and
embs critically affect the outcome of usdg and imaging. Detailed understanding
and modeling of their physical properties is thus essential for
ultrasound-mediated therapeutic applications. State-of-the-art continuuum
models of shelled bodies cannot incorporate critical details such as varying
thickness of the encapsulating shell or specific interactions between its
constituents and interior or exterior solvents. Such modeling approaches also
do not allow for detailed modeling of chemical surface functionalizations,
which are crucial for tuning the gv-blood interactions. We develop a general
particle-based modeling framework for encapsulated bodies that accurately
captures elastic and rheological properties of gvs and embs. We use dissipative
particle dynamics to model the solvent, the gaseous phase in the capsid, and
the triangulated surfaces of immersed objects. Their elastic behavior is
studied and validated through stretching and buckling simulations, eigenmode
analysis, shear flow simulations, and comparison of predicted gv buckling
pressure with experimental data from the literature. The presented modeling
approach paves the way for large-scale simulations of encapsulated bodies,
capturing their dynamics, interactions, and collective behavior.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:23:45 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Ntarakas",
"Nikolaos",
""
],
[
"Lah",
"Maša",
""
],
[
"Svenšek",
"Daniel",
""
],
[
"Potisk",
"Tilen",
""
],
[
"Praprotnik",
"Matej",
""
]
] |
2504.19891 | Bruno Dular | Bruno Dular | Bending parameterization of one-sided degenerated Kleinian surface
groups | 14 pages, 1 figure, comments are welcome! | null | null | null | math.GT | http://creativecommons.org/licenses/by/4.0/ | It was recently proved that quasi-Fuchsian manifolds are uniquely determined
by their bending laminations. This paper concerns a similar result for certain
non-quasi-Fuchsian manifolds~: those obtained by degenerating one end but not
the other, i.e. those appearing in boundaries of Bers slices. More precisely,
we show that such hyperbolic manifolds are uniquely determined by the end
structure of the degenerated end and the bending lamination of the other. The
end structure consists of the parabolic locus, which is a multicurve, together
with ending laminations or conformal structures on each components of its
complement.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:24:55 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Dular",
"Bruno",
""
]
] |
2504.19892 | Milad Sabzehparvar | Milad Sabzehparvar, Fatemeh Kiani, Germ\'an Garc\'ia Mart\'inez, Omer
Can Karaman, Victor Boureau, Lucie Navratilova, Giulia Tagliabue | Thin-wall Single-crystal Gold Nanoelectrodes towards Advanced Chemical
Probing and Imaging | null | null | null | null | physics.chem-ph cond-mat.mes-hall cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | Thin-wall metal ultramicro- and nanoelectrodes (UMEs/NEs), especially gold
NEs, are indispensable for high-resolution electrochemical microscopy,
biosensing, and fundamental research. However, their damage susceptibility and
the lack of scalable fabrication methods hinder broader adoption. We present a
versatile wet-chemical approach for high-throughput fabrication of thin-wall Au
NEs/UMEs and multifunctional NEs with ~80% reproducibility. This method is
based on a unique template-assisted 1D growth of single-crystalline Au in
borosilicate nanopipettes followed by electrochemical contacting with tungsten
microwires, and focused ion beam milling, ensuring precise control over NEs
dimensions. Adaptable to various metals and integrable in multifunctional
probes, the method facilitates batch production of high-quality NEs with
standardized electrical connections. Structural and electrochemical
characterization reveals a twinned single-crystalline Au core, a seamless
Au/glass interface, and highly stable electrochemical performance. Notably,
smaller electrodes exhibit higher current densities, enhancing chemical
detection sensitivity. Specifically, we demonstrate outstanding spatial (< 200
nm) and current (< 1 pA) resolutions, low limit of detection (~11.0 {\mu}M) and
high stability (7 h) in scanning photoelectrochemical microscopy (photo-SECM),
by detecting photo-oxidation reaction on atomically smooth Au micro-flakes. We
also demonstrate growth in double-barrel pipettes for SECM/SICM probes as well
as Pt NEs. Overall, this scalable method addresses longstanding challenges in
NEs, paving the way for advanced electrochemical and spectro-electrochemical
microscopy, including SERS/TERS integration. With single-crystalline surfaces,
these electrodes open new frontiers in catalysis, interfacial electrochemistry,
biosensing, and molecular-scale investigations.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:27:06 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Sabzehparvar",
"Milad",
""
],
[
"Kiani",
"Fatemeh",
""
],
[
"Martínez",
"Germán García",
""
],
[
"Karaman",
"Omer Can",
""
],
[
"Boureau",
"Victor",
""
],
[
"Navratilova",
"Lucie",
""
],
[
"Tagliabue",
"Giulia",
""
]
] |
2504.19893 | Leonie M\"uhlherr | Leonie M\"uhlherr | Separator-based derivations of graphic arrangements | 18 pages, 7 figures | null | null | null | math.CO math.AC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The class of subarrangements of the well-studied braid arrangement, so-called
graphic hyperplane arrangements, is important for analysing new concepts and
properties in hyperplane arrangement theory. While there is a nice
characterization of free graphic arrangements, many interesting questions
beyond the free case remain open. This paper introduces an explicit set of
generators for the module of logarithmic derivations of a general graphic
arrangement based on graph separators. We obtain new insights in the derivation
degree sequence in the non-free case and give bounds on the highest degree in
the sequence. Moreover, this can broadly be used for future research in this
area.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:27:07 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Mühlherr",
"Leonie",
""
]
] |
2504.19894 | Quynh Phung | Quynh Phung, Long Mai, Fabian David Caba Heilbron, Feng Liu, Jia-Bin
Huang, Cusuh Ham | CineVerse: Consistent Keyframe Synthesis for Cinematic Scene Composition | link website: https://cinevers.github.io/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present CineVerse, a novel framework for the task of cinematic scene
composition. Similar to traditional multi-shot generation, our task emphasizes
the need for consistency and continuity across frames. However, our task also
focuses on addressing challenges inherent to filmmaking, such as multiple
characters, complex interactions, and visual cinematic effects. In order to
learn to generate such content, we first create the CineVerse dataset. We use
this dataset to train our proposed two-stage approach. First, we prompt a large
language model (LLM) with task-specific instructions to take in a high-level
scene description and generate a detailed plan for the overall setting and
characters, as well as the individual shots. Then, we fine-tune a text-to-image
generation model to synthesize high-quality visual keyframes. Experimental
results demonstrate that CineVerse yields promising improvements in generating
visually coherent and contextually rich movie scenes, paving the way for
further exploration in cinematic video synthesis.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:28:14 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Phung",
"Quynh",
""
],
[
"Mai",
"Long",
""
],
[
"Heilbron",
"Fabian David Caba",
""
],
[
"Liu",
"Feng",
""
],
[
"Huang",
"Jia-Bin",
""
],
[
"Ham",
"Cusuh",
""
]
] |
2504.19895 | Islam Ahmed | Islam Ahmed, Stefan De Gendt, Clement Merckling | Epitaxial growth of BaBiO3 thin films on SrTiO3(001) and MgO(001)
substrates using molecular beam epitaxy and controlling their crystal
orientations competition | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | BaBiO3 has lately gained high research attention as a parent material for an
interesting family of alloyed compositions with multiple technological
applications. In order to grow a variety of structures, a versatile deposition
tool such as molecular beam epitaxy has to be employed. In this work, molecular
beam epitaxy growth of BaBiO3 on SrTiO3(001) and MgO(001) substrates is
studied. When grown by molecular beam epitaxy on SrTiO3(001) or MgO(001)
substrates, BaBiO3 is known to have two competing orientations namely, (001)
and (011). Characterization of the thin film is carried out by X-ray
diffraction, X-ray reflectivity, atomic force microscopy, Rutherford
backscattering, and transmission electron microscopy. Pathways to block the
growth of BaBiO3(011) and only grow the technologically relevant BaBiO3(001)
are described for both substrates. Understanding of the enabler mechanism of
the co-growth is established from epitaxy point of view. This can be
beneficially utilized for growth of the different compositions of BaBiO3
material family in a more controlled manner.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:28:24 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Ahmed",
"Islam",
""
],
[
"De Gendt",
"Stefan",
""
],
[
"Merckling",
"Clement",
""
]
] |
2504.19896 | Salvatore Orlando | S. Orlando, M. Miceli, M. Ono, S. Nagataki, M.-A. Aloy, F. Bocchino,
M. Gabler, B. Giudici, R. Giuffrida, E. Greco, G. La Malfa, S.-H. Lee, M.
Obergaulinger, O. Petruk, V. Sapienza, S. Ustamujic, J. Weng | Tracing the ejecta structure of SN 1987A: Insights and diagnostics from
3D MHD simulations | The manuscript consists of 24 pages and 16 figures | null | null | null | astro-ph.HE astro-ph.SR | http://creativecommons.org/licenses/by/4.0/ | Supernova (SN) 1987A provides a unique window into the aftermath of a massive
stellar explosion, offering key insights into the ejecta's morphology,
composition, explosion mechanism, progenitor system, and circumstellar medium
(CSM) interaction. We investigate large-scale ejecta asymmetries in SN 1987A.
By comparing the simulations with JWST observations and making predictions for
XRISM, we aim to refine our understanding of the explosion mechanism and the
remnant's evolution. We performed 3D MHD simulations that trace the evolution
of SN 1987A from the SN to the SNR, extending our predictions up to 5000 years
into the future and considering the Ni-bubble effects. The simulation results
are compared with JWST observations and used to predict XRISM spectra, to
evaluate the accuracy of the modeled ejecta structure. Our simulations
reproduce the large-scale Fe-rich ejecta morphology seen by JWST, revealing two
clumps suggestive of a bipolar explosion. Ni-bubble effects in the first year
boost Fe-rich ejecta expansion and their interaction with the reverse shock.
However, discrepancies with JWST observations in clump velocities and spatial
distribution suggest stronger explosion asymmetries than modeled. Since 2021,
our models predict that shocked ejecta have contributed increasingly to X-ray
emission, now rivaling shocked CSM and soon dominating as the latter fades.
Future XRISM observations will trace the evolution of these ejecta structures,
refining constraints on explosion geometry. Early remnant asymmetries from CSM
interaction may persist for at least 100 years. Our results underscore the role
of asymmetric core-collapse mechanisms in shaping SN 1987A's ejecta and
constraining its explosion geometry. Future studies should explore more extreme
asymmetries, in neutrino-driven core collapse or magneto-rotational SN models,
to identify the origin of its bipolar Fe-rich structure.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:28:40 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Orlando",
"S.",
""
],
[
"Miceli",
"M.",
""
],
[
"Ono",
"M.",
""
],
[
"Nagataki",
"S.",
""
],
[
"Aloy",
"M. -A.",
""
],
[
"Bocchino",
"F.",
""
],
[
"Gabler",
"M.",
""
],
[
"Giudici",
"B.",
""
],
[
"Giuffrida",
"R.",
""
],
[
"Greco",
"E.",
""
],
[
"La Malfa",
"G.",
""
],
[
"Lee",
"S. -H.",
""
],
[
"Obergaulinger",
"M.",
""
],
[
"Petruk",
"O.",
""
],
[
"Sapienza",
"V.",
""
],
[
"Ustamujic",
"S.",
""
],
[
"Weng",
"J.",
""
]
] |
2504.19897 | Yun-Wei Yu | Jing-Yao Li, Yun-Wei Yu, Liang-Duan Liu, Ming-Yan Xiao | The Dust Echo Emission of Fast Blue Optical Transients and Application
to the Near-Infrared Excess of AT 2018cow | null | null | null | null | astro-ph.HE astro-ph.SR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A near-infrared (NIR) excess has been discovered in the emission of the
representative fast blue optical transient (FBOT): AT 2018cow. It was suggested
that this NIR excess could be emitted by the dust surrounding the source and,
thus, could provide a probe into the nature of its progenitor. We develop a
model to describe the influence of the FBOT emission on the environmental dust
and, as a result, a dust-free evaporation cavity can be formed on a timescale
of one day. Outside this cavity, the surviving dust grains can have different
size distributions at different distances to the source. With such a special
dust environment, we fit the multi-wavelength light curves of AT 2018cow by
taking into account the evolutionary dust echo of the FBOT emission. It is
found that the dust temperature can vary with time along with the evolution of
the irradiating FBOT emission. Even at a fixed time, the dust temperature can
be distributed in a wide range rather than having only a unique value.
Furthermore, both the mass of the dust shell and its distance to the FBOT are
found to be much larger than those derived with a direct empirical fitting of
the NIR spectra but without considering the evolutionary relationship between
the spectra.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:28:44 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Li",
"Jing-Yao",
""
],
[
"Yu",
"Yun-Wei",
""
],
[
"Liu",
"Liang-Duan",
""
],
[
"Xiao",
"Ming-Yan",
""
]
] |
2504.19898 | Mingqian He | Mingqian He, Fei Zhao, Chonggang Lu, Ziyan Liu, Yue Wang, Haofu Qian | GenCLS++: Pushing the Boundaries of Generative Classification in LLMs
Through Comprehensive SFT and RL Studies Across Diverse Datasets | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a fundamental task in machine learning, text classification plays a
crucial role in many areas. With the rapid scaling of Large Language Models
(LLMs), particularly through reinforcement learning (RL), there is a growing
need for more capable discriminators. Consequently, advances in classification
are becoming increasingly vital for enhancing the overall capabilities of LLMs.
Traditional discriminative methods map text to labels but overlook LLMs'
intrinsic generative strengths. Generative classification addresses this by
prompting the model to directly output labels. However, existing studies still
rely on simple SFT alone, seldom probing the interplay between training and
inference prompts, and no work has systematically leveraged RL for generative
text classifiers and unified SFT, RL, and inference-time prompting in one
framework. We bridge this gap with GenCLS++, a framework that jointly optimizes
SFT and RL while systematically exploring five high-level strategy
dimensions-in-context learning variants, category definitions, explicit
uncertainty labels, semantically irrelevant numeric labels, and
perplexity-based decoding-during both training and inference. After an SFT
"policy warm-up," we apply RL with a simple rule-based reward, yielding sizable
extra gains. Across seven datasets, GenCLS++ achieves an average accuracy
improvement of 3.46% relative to the naive SFT baseline; on public datasets,
this improvement rises to 4.00%. Notably, unlike reasoning-intensive tasks that
benefit from explicit thinking processes, we find that classification tasks
perform better without such reasoning steps. These insights into the role of
explicit reasoning provide valuable guidance for future LLM applications.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:30:58 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"He",
"Mingqian",
""
],
[
"Zhao",
"Fei",
""
],
[
"Lu",
"Chonggang",
""
],
[
"Liu",
"Ziyan",
""
],
[
"Wang",
"Yue",
""
],
[
"Qian",
"Haofu",
""
]
] |
2504.19899 | Felipe Hern\'andez Castro | Felipe Hern\'andez | Multiple Polynomial Recurrence in Weyl Systems | null | null | null | null | math.DS | http://creativecommons.org/licenses/by/4.0/ | In this work we give a full characterization of sets of multiple polynomial
recurrence in Weyl systems, which are ergodic unipotent affine transformations
on products of tori and finite abelian groups. In particular, we show that
measurable and topological recurrence in Weyl systems coincide. Our analysis
also yields a structure theorem for polynomial multicorrelation sequences in
Weyl systems. These results stem from an in-depth study of the Weyl complexity
of a set of polynomials and the introduction of a new concept: the \textit{Weyl
polynomials} generated by a set of polynomials.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:31:01 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Hernández",
"Felipe",
""
]
] |
2504.19900 | Han Chen | Han Chen and Anne L. Martel | Breast Cancer Detection from Multi-View Screening Mammograms with Visual
Prompt Tuning | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Accurate detection of breast cancer from high-resolution mammograms is
crucial for early diagnosis and effective treatment planning. Previous studies
have shown the potential of using single-view mammograms for breast cancer
detection. However, incorporating multi-view data can provide more
comprehensive insights. Multi-view classification, especially in medical
imaging, presents unique challenges, particularly when dealing with
large-scale, high-resolution data. In this work, we propose a novel Multi-view
Visual Prompt Tuning Network (MVPT-NET) for analyzing multiple screening
mammograms. We first pretrain a robust single-view classification model on
high-resolution mammograms and then innovatively adapt multi-view feature
learning into a task-specific prompt tuning process. This technique selectively
tunes a minimal set of trainable parameters (7\%) while retaining the
robustness of the pre-trained single-view model, enabling efficient integration
of multi-view data without the need for aggressive downsampling. Our approach
offers an efficient alternative to traditional feature fusion methods,
providing a more robust, scalable, and efficient solution for high-resolution
mammogram analysis. Experimental results on a large multi-institution dataset
demonstrate that our method outperforms conventional approaches while
maintaining detection efficiency, achieving an AUROC of 0.852 for
distinguishing between Benign, DCIS, and Invasive classes. This work highlights
the potential of MVPT-NET for medical imaging tasks and provides a scalable
solution for integrating multi-view data in breast cancer detection.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:31:08 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Chen",
"Han",
""
],
[
"Martel",
"Anne L.",
""
]
] |
2504.19901 | Jerry Yao-Chieh Hu | Hude Liu, Jerry Yao-Chieh Hu, Zhao Song, Han Liu | Attention Mechanism, Max-Affine Partition, and Universal Approximation | null | null | null | null | cs.LG cs.AI stat.ML | http://creativecommons.org/licenses/by-sa/4.0/ | We establish the universal approximation capability of single-layer,
single-head self- and cross-attention mechanisms with minimal attached
structures. Our key insight is to interpret single-head attention as an input
domain-partition mechanism that assigns distinct values to subregions. This
allows us to engineer the attention weights such that this assignment imitates
the target function. Building on this, we prove that a single self-attention
layer, preceded by sum-of-linear transformations, is capable of approximating
any continuous function on a compact domain under the $L_\infty$-norm.
Furthermore, we extend this construction to approximate any Lebesgue integrable
function under $L_p$-norm for $1\leq p <\infty$. Lastly, we also extend our
techniques and show that, for the first time, single-head cross-attention
achieves the same universal approximation guarantees.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:31:45 GMT"
}
] | 2025-04-29T00:00:00 | [
[
"Liu",
"Hude",
""
],
[
"Hu",
"Jerry Yao-Chieh",
""
],
[
"Song",
"Zhao",
""
],
[
"Liu",
"Han",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.