Peer Revenue Now accepting positions

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Friday 6 November

Astrophysics of Galaxies

SN 2026ahvu in NGC 1433 May Be a Prompt Type Ia Supernova with a Delay Time Under 100 Myr

SN 2026ahvu is a Type Ia supernova (SN) that occurred in the nearby galaxy NGC 1433. Using archival Hubble and JWST imaging from the PHANGS survey, we show that the SN lies within 0.19$''$ and 0.26$''$ of two UV-bright star clusters. This $0.19/0.26''$ separation corresponds to $8/11$ pc ($17/24$ pc) at the reported distances of $9.1$ and $18.9$ Mpc, respectively. The probability of finding an unrelated Type Ia SN this close to a cataloged star cluster by chance is only $\sim$0.4 %. Thus, SN 2026ahvu is likely the first Type Ia SN that can be associated with relatively young star clusters in a nearby galaxy. From spectral energy distribution fitting using HST+JWST (or HST-only) data, we find that the clusters have age estimates of $37_{-5}^{+3}$ ($57_{-9}^{+14}$) Myr and $95_{-13}^{+15}$ ($88_{-12}^{+14}$) Myr. Association with either cluster would mean that SN 2026ahvu should be considered a prompt SN Ia and provide one of the most concrete measurements to date of a prompt Ia delay time. The median age of the younger star cluster is close to the minimum possible delay time for SN Ia progenitors. If the SN is indeed associated with the younger cluster, it would imply that Type Ia SNe can occur almost immediately after white dwarf formation. If the SN is associated with the older cluster, this discovery still presents the first example of a prompt Type Ia SN with delay time under 100 Myr.
Read the full abstract

All authors Sumit K. Sarbadhicary, Kiana Henny, Adam Leroy, Karin Sandstrom, Kirsten L. Larson, David A. Thilker, Gagandeep S. Anand, Ashley T. Barnes, Laura A. Lopez, Daniel Maschmann, Marco Troncoso, Hwihyun Kim, Janice C. Lee, Frank Bigiel, Médéric Boquien, Rupali Chandar, Ryan Chown, Daniel A. Dale, Sinan Deger, Oleg V. Egorov, Simon C. O. Glover, Oleg Y. Gnedin, Jay González Lobos, Hao He, Rebecca J. Houghton, Debosmita Pathak, M. Jimena Rodriguez, Eva Schinnerer, Michael Tucker, Tony D. Weinbeck, Thomas G. Williams, Bradley C. Whitmore, Leonardo Ubeda

Geometric Topology

Pure mapping class group fixed points in $\mathrm{PU}(2,1)$

We classify all the fixed points for the action of the pure mapping class group on $\mathrm{PU}(2,1)$ character varieties of surface groups. Up to trivial peripheral monodromies, we prove that the irreducible fixed points are either pair of pants representations, or restrictions of reduced Burau representations to a once-punctured torus and restrictions of reduced Gassner representations to a 4-punctured sphere. On the other hand, we prove that the only non-trivial reducible fixed points appear in genus 0 where they map all but at most three peripheral loops in the centralizer of the image, and in genus 1 where they are valued in the quaternion group $H_8$. As consequences, we deduce (i) that an irreducible representation in genus 0 with at least four non-trivial peripheral monodromies is rigid if and only if it is a fixed point, and (ii) a partial classification of irreducible representations of pure braid groups into $\mathrm{PU}(2,1)$.
Read the full abstract

Mesoscale and Nanoscale Physics

Trajectory-Dependent Decoherence and Leakage During Shuttling of Multi-Spin Qubits

Logical qubits encoded in collective states of multiple spins are promising building blocks for large-scale semiconductor quantum processors. As novel fault-tolerant architectures increasingly rely on qubit shuttling, understanding decoherence of the shuttled multi-spin qubits becomes crucial. We analyze fidelity and leakage of the logical qubits dephased by the spatio-temporally correlated $1/f$ noise, considering two-, three-, and four-spin encodings. We study different shuttling protocols and establish that for all encodings generally, parallel shuttling provides the best protection against decoherence at short distances, whereas sequential shuttling outperforms others for long-range transport (although the exact boundaries depend on the specific device parameters). We explain this common behavior in terms of the dephasing factors of pairs of the shuttled spins; we provide explicit expressions for fidelities and leakage, thus enabling experimentalists to predict performance of the multi-spin qubits from only several independently measured pairwise dephasing factors. Our results establish a connection between shuttling trajectories, noise correlations, and qubit decoherence, and provide guidance for achieving noise-resilient shuttling in scalable semiconductor quantum processors.
Read the full abstract

All authors Yu-Ning Zhang, Aleksandr S. Mokeev, Maximilian Rimbach-Russ, David P. DiVincenzo, Viatcheslav V. Dobrovitski

Artificial Intelligence

On the Reliability of Majority-Vote Pseudo-Labels in Test-Time Reinforcement Learning

Test-time reinforcement learning (TTRL) adapts pretrained language models to unlabeled target-domain data using self-generated pseudo-labels, whose reliability remains poorly understood. We analyze majority-vote supervision for TTRL with online Direct Preference Optimization (DPO) and decompose its error into population-majority bias and prompt-dependent finite-rollout variation. Because population-majority bias is generally unidentifiable without external supervision, we focus on mitigating the downstream effects of finite-rollout variation. We propose Prompt-Conditional Distributionally Robust DPO (PC-Dr.DPO), which conditions within-prompt robust reweighting on observed vote reliability to make updates from sampling-sensitive prompts more conservative. We further introduce a cache-based mechanism to reduce self-supervised forgetting during online adaptation. Across six mathematical and knowledge-intensive reasoning benchmarks and five model backbones, PC-Dr.DPO matches or outperforms the strongest TTRL baseline in 24 of 30 settings.
Read the full abstract

All authors Haoyu Wang, Ru Wang, Yuanyuan Wang, Jiaxian Guo, Mingming Gong

Astrophysics of Galaxies

Multi-phase Carbon Structure and Dynamics in NGC 6334: Insights from CO, [C I], and [C II]

We present high-angular and velocity resolution observations of the emission lines of [CII] at 158 micron, [CI] at 609 micron, and the 12CO and 13CO (J=3-2) transitions toward the central region of NGC 6334, providing a detailed multi-phase view of carbon in this massive star-forming complex. By combining morphology and spatial scales information, kinematics, line profiles, and line ratio analyses, we quantify the distribution and physical state of ionized and atomic carbon, and carbon in its molecular form in CO. [CII] emission is the most extended, permeates the mapped region and is dynamically complex, while 13CO is compact and quiescent, and 12CO and [CI] occupy intermediate regimes connecting dense regions to transitional diffuse layers. We interpret these emission distributions such that the brightest [CII] emission traces photodissociation regions (PDRs) close to the H II region bubbles located along the central filament, while the extended [CII] emission arises from the more diffuse warm gas, indicating either cloud formation or dispersal processes. Remarkably, the [CI] is more extended than the 13CO, but less than [CII] and may probe the gas velocity structure resulting from colliding flows or cloud-cloud collisions. Line ratios involving [CII], such as [CII]/[CI] or [CII]/CO are systematically high, revealing widespread diffuse emission and localized UV-illuminated surfaces. Line ratio analysis and K-means clustering of [CI]/13CO versus dust continuum (cold 870 micron and warm 70 micron) emission reveal stratified chemical structure that UV photons penetrate deeply into the clumpy ridge (Av~18), suggesting a porous PDR. These results together depict that NGC 6334 is a clumpy, feedback-dominated PDR ridge with coexisting carbon phases, where [CI] and [CII] also trace dynamical processing through multiple velocity components linked to cloud-cloud collisions.
Read the full abstract

All authors S. Neupane, F. Wyrowski, N. Schneider, A.G.G.M. Tielens, R. Simon, J. S. Urquhart, R. Güsten, M-Y. Lee, D. Russeil, G. Garay, R. Higgins, K. M. Menten

Computers and Society

AI Exposure and the Changing Responsibilities in Job Postings

The growing adoption of artificial intelligence (AI) in the workplace has raised questions about how employer demand for different work activities is changing. We examine how the prevalence and content of work activities in online job postings change in association with activity-level measures of AI exposure, derived from Copilot and Claude usage. To enable systematic comparisons, we develop a pipeline that extracts the responsibilities from job advertisements and maps them onto O*NET's standardized taxonomy of Intermediate Work Activities (IWAs). We then apply this approach to 50,000 job advertisements from 33 firms between 2023 and 2025. From this analysis, we first find that explicit mentions of AI increased by 66%, although much of this growth occurs outside descriptions of the actual work activities associated with a job. Looking specifically at the advertised job responsibilities, activities with greater AI exposure tend to decline in advertised prevalence, but only mildly as a share of the total number of responsibilities. At the same time, more AI-exposed work activities increasingly are associated with explicitly AI-related responsibilities, suggesting that employers are looking to hire people who can perform these responsibilities with AI.
Read the full abstract

All authors Dora Zhao, Lin Ai, Pranav Khadpe, Siddharth Suri, Scott Counts

Software Engineering

An Empirical Study on Clustering-based Quantum Approximate Optimization Algorithm for Regression Test Case Selection

Regression testing assesses whether software components that have already been developed and tested continue to work correctly after modifications. Ideally, regression testing re-executes the entire test suite after each modification, but this is infeasible in real-world scenarios because of the computational burden; therefore, optimized approaches, such as Test Case Selection, have been developed. The state of the art in regression testing optimization includes both classical approaches, such as Additional Greedy and DIV-GA, and quantum approaches, such as SelectQA, BootQA, and IGDec-QAOA. However, while the former incur high computational costs, the latter are constrained by the physical limitations of currently available quantum machines. This paper extends the evaluation of QAOA-TCS, our quantum approach for the test case selection problem. We thoroughly evaluate its performance, compare it with state-of-the-art classical and quantum approaches, and assess its robustness to quantum noise. The results show that QAOA-TCS outperforms state-of-the-art multi-objective approaches in most cases and maintains stable performance even under high noise levels. In the single-objective formulation, QAOA-based methods perform comparably. In terms of efficiency, our previous method based on quantum annealing, namely SelectQA, is the most efficient for large problems in the multi-objective formulation, whereas QAOA-TCS offers the best trade-off between effectiveness and efficiency in the single-objective formulation when run on realistic noisy simulators. However, such analysis is constrained by the different computational environments used to run the algorithms.
Read the full abstract

Computation and Language

CodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs

Large reasoning models (LRMs) improve task performance by generating longer intermediate reasoning traces, but this substantially increases inference costs and makes low-precision quantization increasingly important. However, aggressively quantized models often perform poorly on reasoning tasks. We observe larger activation fluctuations during the first decoding steps of reasoning. Calibration that omits these early states may miss their activation fluctuations. To address these reasoning-specific quantization challenges, we propose CodeQuant2, a clustering-based LUT quantization framework for low-precision large reasoning models. CodeQuant2 mitigates activation outliers with rotation and scaling, uses clustering to represent non-uniform weight distributions, and adapts codebook sizes to the quantization difficulty of each layer. Guided by our observation of reasoning-time activation behavior, CodeQuant2 further improves calibration sample selection to better match actual decoding statistics. Combined with our customized LUT CUDA kernel, CodeQuant2 delivers substantial accuracy improvements on reasoning benchmarks while achieving up to $2.85\times$ speedup over BF16.
Read the full abstract

All authors Xingyu Liu, Xiangyang Yin, Tianhua Xia, Haiyu Wang, Sai Qian Zhang

High Energy Physics (Theory)

Existence of extremal solutions in the 1d conformal bootstrap

Extremal solutions and functionals play an important role in the conformal bootstrap. These are the solutions that saturate unitarity bounds, and the dual functionals that encode the optimal bounds. While they have fundamental theoretical implications, their existence has not been rigorously established. This is particularly subtle in the setting of the full infinite-dimensional bootstrap, i.e. the limit of zero numerical truncation. In this article, we prove that extremal solutions exist for OPE maximisation problems in 1d CFT. These are a large class of generically interacting, unitary, exact solutions of the crossing equation that maximise an OPE coefficient. We also prove that these problems have 'zero duality gap', meaning there exist functionals arbitrarily close to extremality. Our approach is to formulate the bootstrap as a convex optimisation problem in a Hilbert space and apply the Fenchel-Rockafellar theorem.
Read the full abstract

Computer Vision

GRAC: Graph-Based Retrieval-Augmented Captioning for Surgical Scene Description

Purpose: Surgical operative reports support procedural documentation, surgical training, performance assessment, and workflow review. However, extracting meaningful information often requires extensive manual review and clinical expertise. Automated surgical scene descriptions could reduce this burden by leveraging recurring surgical activities and prior knowledge. Methods: We propose GRAC (Graph-Based Retrieval-Augmented Captioning), a framework that represents surgical scenes as graphs capturing relationships among instruments, actions, and anatomical structures. Given an input scene graph, GRAC learns to retrieve relevant graph-caption pairs from training data. It compares instrument-tissue interactions between retrieved examples and the input scene, using comparison features to condition retrieved-caption representations during generation. Results: On the EndoVis-18 benchmark, GRAC achieved a CIDEr-D of 5.88, compared with 4.23 for SGT and 3.36 for the strongest conventional Transformer baseline. Exact instrument-action-target triplet-set accuracy reached 52.6%, compared with 32.7% for SGT and 26.2% for the strongest conventional baseline. These results demonstrate improvements in language quality and factual agreement. On TORS, GRAC achieved a CIDEr-D of 5.06, compared with 3.80 for the strongest baseline. Conclusion: GRAC demonstrates that combining structured surgical scene representations with retrieved examples improves surgical scene captioning and produces more contextually meaningful descriptions.
Read the full abstract

All authors Rochak Dhakal, Chen Zhao, Soup Cobb, Xin Qian, Xiaofeng Yang, Weihua Zhou

Lattice Field Theory

Emergence and polymer physics of the center-vortex network: Confronting SU(4) lattice Yang-Mills theory

The geometric organization and statistical properties of center-vortex networks provide a window into the infrared dynamics underlying confinement in Yang-Mills theory. Motivated by this connection, we investigate new consequences of a previously proposed confinement mechanism based on physically equivalent elementary Cartan fluxes. We show that, in the percolating phase, these fluxes dynamically reorganize into center-charge sectors whose dominant vortex tensions satisfy a Casimir scaling law, similar to that observed for confining string tensions. We then confront this result with the first nontrivial Yang-Mills case: the doubly charged vortex chains recently observed in SU(4) lattice simulations. To extract the parameters needed to connect the lattice ensemble with the effective description, we characterize how center-vortex chains grow and bend in SU(3) and SU(4) using polymer-inspired techniques not previously applied to center vortices. We find good quantitative agreement for the SU(4) doubly charged sector, providing further evidence for the proposed mechanism.
Read the full abstract

Robotics

Grasp-TAGS: Task-Aware Grasp Selection via Critic Distillation for Prehensile Loco-Manipulation

Loco-manipulation, which requires mobile manipulators to coordinate locomotion and manipulation over long horizons, is a key capability for large-object rearrangement in applications such as service robotics and warehouse automation. Despite recent progress, reliably executing these tasks remains challenging because performance depends on complex robot-object and object-environment interaction dynamics. Existing approaches address this challenge either through model-based methods, which often require accurate dynamics models, or through learning-based methods, which reduce this requirement but typically assume a fixed or predefined initial grasp. However, different geometrically stable grasps can lead to substantially different downstream manipulation performance. Therefore, we propose Grasp-TAGS, a task-aware grasp selection framework for goal-conditioned large-object rearrangement. We train a unified PPO-based reinforcement learning policy across multiple initial grasp configurations and distill the learned critic into a lightweight grasp utility network that ranks candidate grasps using only the candidate grasp configuration and goal information. At inference time, we select the grasp with the highest predicted downstream utility for the downstream policy to complete the task, before even making a contact. Experiments demonstrate that Grasp-TAGS improves rearrangement success over both fixed single-grasp policies and random grasp selection, achieving an average success rate of 89.84% compared with 79.87% for the best-performing single-grasp policies, an improvement of 9.97 percentage points. Grasp-TAGS also reduces final orientation errors across all evaluated objects. For more qualitative results, refer to the supplementary webpage: https://rpm-lab-umn.github.io/Grasp-TAGS-web-page/
Read the full abstract

Artificial Intelligence

Flow-Guided Target-Space Alignment via Path Consistency

Many multimodal alignment problems require mapping a source modality into a fixed space defined by a target modality, where the target space is not merely an intermediate embedding but is used directly for downstream analysis. In this setting, successful alignment should preserve both the global structure of the target distribution and the instance-level correspondence observed in paired source–target training data. Existing objectives typically address only part of this goal: contrastive losses encourage relative similarity, distribution-matching losses improve marginal agreement, and pointwise regression fits each source sample to its paired target without explicitly modeling target-space path geometry. We propose FlowAlign, a flow-guided target-space alignment method that repurposes conditional flow matching as a training-time alignment signal rather than an inference-time transport model. FlowAlign uses the learned flow field to guide the alignment of source-side representations with the fixed target space, encouraging them to preserve paired correspondence while remaining compatible with the global target-space structure. We further provide theoretical analysis showing how FlowAlign helps improve paired correspondence by controlling endpoint accuracy and capturing local target-space geometry. Experiments on synthetic data, paired single-cell multiome benchmarks, and image–text alignment demonstrate that FlowAlign improves instance level correspondence while preserving target-space structure across different modality settings.
Read the full abstract

All authors Ruizhi Yuan, Zeqiu Yu, Wei Gao, Wei Chen, Xiwei Tang, Lu Tang

Group Theory

Finitely Presented Covers of Topological Full Groups

Commutator subgroups of infinite minimal subshifts provide the first examples of infinite, finitely generated, simple amenable groups. We prove that if $G$ is a finitely presented group which maps onto the commutator subgroup of the topological full group of an infinite minimal subshift, then $G$ contains a non-abelian free subgroup. This in particular implies that these groups are not finitely presented. We also show that the topological full group of an irreducible subshift of finite type contains non-abelian free subgroups.
Read the full abstract

Fluid Dynamics

A Quantum Mechanical Description of Active Matter

The natural environments habitable to life are inseparable from dissipative and entropic forces, which mandate active matter to operate on constant fluxes of energy and momentum to survive and propagate through spacetime. In contrast to fundamental particles which are characterized by energy and momentum, we postulate that energy expenditure rate and momentum generation rate are the fundamental dynamical quantities driving this evolution of swimming active matter and accordingly construct the corresponding Lagrangian and Hamiltonian of motion of the active matter. Furthermore, living systems broadly employ periodic, non-repeating motion for both locomotion and various biological functions, with meaningful information stored in the periods of these functions. Consequently, analyzing these dynamics in the Fourier space lands us at the familiar tools of the operator based, quantum mechanics. This is accomplished by introducing a complex field to describe the evolving probability distribution of a single unit, as well as the population density of mutually non-interacting active particles under the constraint of scarce resources. The resulting single-unit wavefunction evolves analogously to the Dirac equation with minimal coupling, wherein gauge symmetry reflects the generality of the governing equations across arbitrary resource landscapes. Additionally, we present a Schrödinger-like equation where the effective potential term represents the environmentally mandated, non-motile base metabolic rate. Finally, we formalize the bidirectional coupling between active swimmers and the surrounding fluid via a multipole expansion of the active force with an emphasis on force monopoles and dipoles, where each pole corresponds to a distinct mode of active motility.
Read the full abstract

Optimization and Control

Geometric Pairwise Classification of Affine Constraints over Ellipsoidal Regions

This paper presents a deterministic geometric procedure for analysing affine constraints over fixed ellipsoidal regions, motivated by chance-constrained set approximation. Analytical tests classify inequalities as always violated, never violated, or cutting. Pairwise tests determine whether two inequalities can be strictly violated at a common point, yielding a compatibility matrix. The tests follow from quadratic optimisation and the Karush–Kuhn–Tucker conditions, with explicit treatment of tangencies, parallel normals, and degenerate inequalities. The formulation applies in arbitrary finite dimensions and provides geometric information for subsequent bounds on simultaneous sampled violations and safe-region approximation.
Read the full abstract

All authors Igor M. L. Pataro, Martina Mammarella, Fabrizio Dabbene, Teodoro Alamo

General Relativity and Quantum Cosmology

Particle production from noncommutative-inspired Reissner-Nordström black hole

By considering a massless Klein-Gordon field and the moving mirror model, we calculate particle production for a noncommutative-inspired Reissner-Nordström black hole sourced by smeared distributions of mass and charge. We show that noncommutative truncated trajectories diminish the emission spectrum at large frequencies, due to noncommutative effects on the external surface gravity of the black hole, in sharp contrast with the spectrum of a classical Reissner-Nordström black hole.
Read the full abstract

Analysis of PDEs

Large-Time Behavior and Asymptotic Profiles for Time-Fractional Nonlinear Diffusion Equations via Entropy Methods

We investigate the large-time behavior of nonnegative solutions to the time-fractional nonlinear diffusion equation $\partial_t^{\alpha}(u-u_0) = \Delta u^m$, $0<\alpha<1$, $m>0$, in a bounded domain with the homogeneous Dirichlet boundary condition. We prove that the rescaled solution $t^{\frac{\alpha}{m}}u$ converges as $t\to\infty$ to an asymptotic profile $S$ characterized by $-\Delta S^m = u_0/\Gamma(1-\alpha)$. Moreover, we show that $t^\alpha u^m$ converges to $S^m$, which admits an explicit representation in terms of the Green function and the initial datum. In contrast to the classical case $\alpha=1$, the asymptotic profile depends explicitly on the initial datum.
Read the full abstract

Superconductivity

Energy- and momentum-resolved Cooper-pair emission from a cuprate superconductor

Superconductivity arises from Cooper pairing, but conventional photoemission probes electrons individually rather than as pairs. Here we report the observation of Cooper-pair emission from an optimally doped cuprate superconductor using two-electron angle-resolved photoemission (2e-ARPES). Below the superconducting transition, we observe a distinct two-electron emission peak at zero pair binding energy and zero pair momentum that disappears in the normal state, revealing the characteristic fingerprint of a Cooper pair. Unexpectedly, the joint energy distribution broadens markedly along the relative-energy axis as temperature decreases, while the total energy width remains nearly unchanged. This anomalous behavior cannot be explained by instrumental resolution, static disorder or thermal broadening, and instead points to scattering processes involving collective excitations of the superconductor. Our results establish 2e-ARPES as a probe of both Cooper-pair emission and the many-body dynamics of the superconducting state.
Read the full abstract

All authors Henry Amir, Jack Zwettler, Ka Ho Wong, Genda Gu, Rafael M. Fernandes, Jorge Noronha, Dirk K. Morr, Fahad Mahmood

Cosmology and Nongalactic Astrophysics

Lessons from double-plane lenses for galaxy profiles, the Hubble tension, and the mass-sheet degeneracy

Double-plane lenses, like single-plane lenses, are subject to the mass-sheet degeneracy, a fundamental limit to our knowledge of the deflector mass profiles from lensing. However, with a prior on the relative expansion history, unlike in the single-plane case, this degeneracy can be constrained with the modelling of lens images, without the need for time delays, velocity dispersion measurements or a prior on the Hubble constant $H_0$. We show that double-plane lenses are sensitive to a specific combination of the mass-sheet components of the two lens planes. Using literature values of the cosmological scaling factor from the double-plane lenses J0946 and AGEL0353, a measurement from the point-source images of J1721 (the “Einstein zig-zag” lens), and a reference value of $\Omega_\mathrm{m}$, we show the constraints that can be placed on the mass-sheet component of each of these systems. We show that the additional information brought by images of a second source can improve measurements of $H_0$ from individual time-delay lenses by up to $\sim50\%$, depending on certain key assumptions. Using forecasted Euclid and LSST double-plane lens samples, we show that these samples will enable strong constraints on the population of mass sheets in lens galaxies. Such population constraints will have implications for a range of strong-lensing science applications, in particular in constraining $H_0$, dark energy, and the dark matter profiles of galaxies.
Read the full abstract

All authors Daniel P. Johnson, Martin Millon, Pierre Fleury, Giacomo Queirolo, Tian Li, Wolfgang J. R. Enzi, Daniel J. Ballard, Duncan Bowden, Thomas E. Collett, Daniel Gilman

Quantum Gases

Raman anyon-Hubbard model in a 1D Bose lattice gas

The anyon-Hubbard model is the paradigmatic lattice model of one-dimensional (1D) anyons, in which chiral interactions make bosons equivalent to particles with fractional statistics. Its continuum limit, the chiral BF theory, emerges as the low-energy description of a Raman-coupled Bose gas with imbalanced interactions. Here, we show that loading such a gas into a 1D optical lattice realizes the lattice counterpart of this theory, which we name the Raman anyon-Hubbard model. The Raman-dressed bosons acquire density-dependent Peierls phases when tunneling through occupied sites. For a small statistical phase, the model reduces to the anyon-Hubbard model, while beyond this limit it defines a distinct lattice anyon model with unbounded rather than periodic statistical interactions. Numerically, we show that the dressed gas displays the defining properties of 1D anyons, namely a density-dependent asymmetry of the ground-state momentum distribution, interaction-induced asymmetric transport and chiral bound states. Remarkably, these features are readily accessible with Raman-coupled $^{39}$K atoms, opening up the experimental exploration of 1D anyons beyond the few-particle regime.
Read the full abstract

All authors Josep Cabedo, Claudio Iacovelli, Leticia Tarruell, Alessio Celi

Image and Video Processing

Super-resolved digital terrain models for GEDI geolocation correction: agreement with airborne LiDAR DTM-based corrections at the beam level, not the footprint level

The Global Ecosystem Dynamics Investigation (GEDI) mission measures three-dimensional forest structure with a LiDAR aboard the International Space Station. GEDI observations have geolocation errors of about 10 m, which cause misalignment with other remote sensing data and reduce the accuracy of downstream tasks. Matching GEDI ground elevations against high-resolution digital terrain models (DTMs) from airborne LiDAR is one way to correct these errors, but such DTMs exist in few regions. To widen this method's applicability, we super-resolved a more widely available low-resolution DTM and used it for the matching. Our question is whether a super-resolved DTM yields the same correction as a 1 m-resolution airborne LiDAR DTM and, if not, why. In steep forests of western Japan, we super-resolved an existing 10 m-resolution DTM to 1 m. Improving its accuracy required fusion with a global 30 m-resolution DTM from a different source. Using the resulting DTM, we corrected GEDI observation positions first by a single shift per ground track, then per observation. For most tracks, the shifts nearly matched those from the airborne LiDAR DTM, likely because DTM errors canceled out across many observations. Per-observation corrections, however, did not match. At a single observation, the error does not cancel out and is too large for the correction. This error likely originates mainly from the input DTM. Therefore, even without a 1 m-resolution DTM, a roughly 10 m-resolution DTM, fused with a global DTM and super-resolved to 1 m, should yield track-wise corrections nearly matching those from a 1 m-resolution DTM.
Read the full abstract

Numerical Analysis

Efficient iterative methods for the nonlinear compressible Reynolds equation with Navier slip boundary condition

In this study, we present a comparison of iterative methods for the compressible Reynolds equation with Navier slip boundary conditions, based on a fully compressible cavitation model for lubricated bearings. We establish the existence and uniqueness of the solution to the two-dimensional problem. We introduce the Kirchhoff transformation to reformulate the nonlinear density-pressure relationship into a quasi-linear problem. We develop an efficient numerical approach to solve the two-dimensional nonlinear problem. The methods examined are the Gauss-Seidel method, the Newton-Raphson method and parallelized nonlinear Jacobi method, applied to one- and two-dimensional slider bearing configurations. The overall solution time is shown to depend strongly on the choice of method.
Read the full abstract

Astrophysics of Galaxies

The AIDA-TNG project: Galactic bar pattern speeds in self-interacting dark matter

Cosmological simulations within the standard cold dark matter (CDM) paradigm systematically predict higher values of the dimensionless bar-speed parameter, $\boldsymbol{R} \equiv R_{\mathrm{cr}}/R_{\mathrm{bar}}$ (where $R_{\mathrm{cr}}$ and $R_{\mathrm{bar}}$ denote the corotation radius and the bar radius, respectively), than those inferred from observations of local galaxies, a discrepancy known as the fast-bar tension. Since bar slowdown is primarily driven by dynamical friction with the dark matter halo, bar dynamics is inherently sensitive to the properties of dark matter particles. We aim to determine whether dark matter self-interactions alleviate or exacerbate the fast-bar tension, and to identify the physical mechanism behind the differences in bar dynamics between CDM and self-interacting dark matter (SIDM) models. We use the $51.7\,\mathrm{Mpc}$ box of the AIDA-TNG cosmological simulations, run with three different dark matter models: standard CDM, and two SIMD models one with a constant cross-section model (SIDM1), and the other with a velocity-dependent model (vSIDM). We identify barred disk galaxies with stellar mass $M_* > 10^{10}\,M_{\odot}$ at redshifts $z=0.2,\,1.0,\,1.5,$ and $2.0$, and measure the bar fraction, bar strength, bar length, pattern speed $\Omega_{\mathrm{p}}$, and the ratio $\boldsymbol{R}=R_{\mathrm{cr}}/R_{\mathrm{bar}}$. The pattern speeds in the SIDM1 and vSIDM models are generally lower than in CDM at nearly all redshifts probed in this study. The subsequent slowdown proceeds with nearly 10% more effective in the SIDM models likely due to velocity-anisotropy. Consequently, at low redshift the mean value of $\boldsymbol{R}$ is significantly higher in the SIDM models, indicating that the fas-bar tension is further exacerbated in these scenarios.
Read the full abstract

All authors Asiyeh Habibi, Mahmood Roshan, Giulia Despali, Mark Vogelsberger, Amir Ghari, Lauro Moscardini

Computer Vision

Health-System-Wide Mapping of Demographic Information in Routine Clinical Histopathology

The extent to which demographic-associated information is detectable across organs and disease states in routine clinical histopathology remains unclear. Here, we analyzed 1.3 million hematoxylin and eosin whole-slide images from 257,322 patients across 25 organs, using pathology foundation-model representations and multiple-instance learning to systematically map information associated with sex, age group, and self-reported race. Across 69 organ-demographic combinations, demographic attributes were broadly recoverable, with pan-organ mean AUROCs of 0.828 for sex, 0.733 for age group, and 0.662 for self-reported race, and substantial variation across organs. Demographic-class balancing produced relatively small changes in performance, while evaluation within diagnosis-text clusters showed modest attenuation. Under training with diagnosis-based grouping, demographic-associated signals remained detectable across multiple organs, including sex in gastric tissue (AUROC 0.865), age group in breast (0.816), and self-reported race in skin (0.728). Literature-guided analysis of these settings generated candidate histomorphological and contextual explanations for further investigation. Together, these findings provide a systematic map of demographic-associated information in routine histopathology, informing investigations of its sources, assessments of potential shortcut learning, and the development of more interpretable computational pathology models.
Read the full abstract

All authors Shengjia Chen, Ruchika Verma, Kevin Clare, Paula Restrepo, Jannes Jegminat, Eugenia Alleva, Kuan-lin Huang, Aryeh Stock, Brandon Veremis, Thomas Fuchs, Gabriele Campanella

Chemical Physics

Classical control of quantum-state relaxation in confined molecular hydrogen

Quantum-state conversion between the ortho and para nuclear-spin states of H$_2$ is forbidden in the isolated molecule but these states can become connected through interactions with a surrounding environment. How that same environment governs the dynamics of the newly connected states remains poorly understood. Here, we investigate this problem using H$_2$ confined in crystalline molecular solids, where the host environment selectively enables conversion between a pair of otherwise disconnected ortho-H$_2$ and para-H$_2$ states. Rotationally resolved infrared spectroscopy directly determines the state-specific conversion dynamics in crystalline CO$_2$ and N$_2$O at 10 K, revealing that increasing H$_2$ concentration suppresses nuclear-spin conversion by nearly an order of magnitude, reaching complete suppression by 20% H$_2$ loading. Molecular dynamics and Monte Carlo simulations identify two complementary origins of this concentration dependence: changes in the thermally populated phonon modes available to mediate relaxation, and strain-driven H$_2$ clustering that progressively removes molecules from the host crystal field responsible for enabling conversion. The resulting picture connects a classical collective response of the host to microscopic quantum-state dynamics: H$_2$ loading generates strain that reorganizes the local molecular environment, thereby modifying both the phonon bath and the molecular coupling to that bath. These results establish a multiscale mechanism by which condensed-phase environments can govern the connectivity and dynamics of an effective two-level system.
Read the full abstract

Rings and Algebras

Deformations and Local Geometry of Lie Triple System Morphisms

In this paper, first we study the deformations, stability and rigidity of Lie triple system morphisms by means of higher derived brackets and augmented Yamaguti cohomology. The vanishing of the first cohomology gives formal and local inner rigidity. The vanishing of the second cohomology removes the formal extension obstructions and, in finite dimension over $\mathbb{R}$, gives local smoothness and stability under simultaneous perturbations of the two products. Next, we establish the relationship with Lie algebra homomorphisms in each part of the theory. The universal enveloping Lie algebra identifies a Lie triple morphism with a grading-preserving Lie homomorphism. We obtain the corresponding first cohomology isomorphism, the equivalence of formal and smooth deformations, and a stability result for graded Lie bracket perturbations. For Lie triple products induced by binary Lie brackets, we construct a comparison of differentials and obstruction classes. Under perfectness and a zero-centralizer condition, the local morphism spaces coincide. Finally, we relate triple subsystem graphs to graded Lie subalgebra graphs and compare their normal differentials.
Read the full abstract

Information Theory

How far does a hash-only sender go? A pre-registered map of sender memory versus bits for short text messages

Can a sender that only computes hashes, with a language model only at the receiver, transmit short human text more cheaply than a classical sender trained on the same domain? We answer with one pre-registered run on 1,000 English user utterances from Taskmaster-1. The primary endpoint is the cost per message to deliver 100% of the messages (hash bits, plus a 2-byte negative acknowledgement and the raw text when the first attempt fails), paired against an in-domain PPM order-1 sender. The hash-only sender delivered 857 of 1,000 messages on the first attempt and cost 192.8 bits per message, 18.4 bits more than PPM order 1 (95% bootstrap interval 10.9 to 26.3); by the pre-registered rule it spends more and is not equivalent within one byte. As secondary results we map measured sender memory against bits. The hash-only sender (8,190 B on a host build, keyed BLAKE2b tag included, as in every method) was the cheapest point in bits from 8,190 B up to, but excluding, PPM order 1 at 24,844 B, 45 to 46 bits below smaz2 and Unishox2; from PPM order 1 upwards, in-domain senders spend fewer bits, except order 2 pruned to 16 KiB. In airtime it loses to PPM order 1 on all five links priced, to Unishox2 on four, and to raw text with a tag on two, because each failure costs an acknowledgement frame and a fallback frame. A wider receiver search recovers half of the failures and makes the cost 7.7 bits lower than PPM order 1, at about three minutes of receiver time per failure, and still loses in airtime on four of five links. We report the measurement as a delimitation, with every deviation of the run.
Read the full abstract

Machine Learning

Computer-Use Agents Are Embodied Generalists

In many areas of AI, general-purpose models have increasingly replaced handcrafted methods and specialized architectures. We argue that embodied AI is poised for a similar shift. Our key claim is that embodied tasks can be formulated as interaction with digital software. Specifically, we build software that renders an agent's view of a physical environment on a screen and provides controls for acting in it. A computer-use agent (CUA) can then perform embodied tasks through the same screenshots and keyboard and mouse actions that it uses to operate ordinary software. Using a single CUA design and changing only the software, our entries win first-, second-, third-, and sixth-place prizes across four ECCV workshop competitions, without any competition-specific agent tuning. These results suggest that the skills CUAs learn on software, such as visual grounding, precise clicking, and exploration, already carry over to embodied tasks, giving CUAs a far broader scope in embodied AI than previously realized. We therefore advocate that computer-use agents can be embodied generalists. Further, we advocate that, by following the training strategy that made computer-use agents general, embodied AI can converge on a single general model as well, ushering in a new era of general embodied AI. Website: https://pranjal2041.github.io/cua-embodied
Read the full abstract

Artificial Intelligence

Ensemble Methodology for Reference-Free Confidence Estimation in LLM-based Code Summarization

This paper describes an approach and use case to estimate the correctness of LLM-generated COBOL summaries without access to human-authored references. The core intuition is that inter-model agreement on semantics is informative about truth, and that an LLM-as-judge can provide higher-fidelity pairwise assessments than traditional string- or embedding-based metrics. We operationalise this intuition with (i) controlled summary generation from diverse models; (ii) a LLM evaluator that compares pairs of summaries and emits structured judgments; and (iii) a calibrated aggregation rule that combines supportive evidence while penalising only corroborated contradictions.
Read the full abstract

All authors Marco Colombo, Balaji A Jayaram, Sowmyakumari Selvaraj, Sumit Tokkar, Animesh Kumar, Aravinda Prasad Sistla, Ravishankar Shetty, Rigel Gjomemo