Peer Revenue Past citations are no guarantee of future returns

Papers

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

All of arXiv

Tuesday 6 October

Computer Vision

RACER: Reflective Agent Coupling Query Interpretation and Tool-Based Retrieval for Frame Selection in Long Video Understanding

Yiyang Huang, Yitian Zhang, Yizhou Wang et al. (+6 more)

Video large language models (Vid-LLMs) excel at diverse video-language tasks by reasoning over selected frames. However, frame selection for long videos remains challenging, as it requires retrieving relevant frames distributed across segments from a large candidate pool given complex queries. This paper investigates dominant approaches to long-video frame selection from a task-decomposition perspective, identifying two key challenges: the Query Comprehension Gap in similarity-based methods and the Interpretation–Selection Gap in judgment-based methods. To address them, we propose RACER, a training-free reflective agentic framework that decomposes long-video frame selection into query interpretation driven by a lightweight Vid-LLM and evidence localization supported by an embedding model serving as a retrieval tool. Specifically, the Vid-LLM is responsible solely for reformulating the complex query into sub-queries that make implicit information requirements explicit, mitigating the Query Comprehension Gap. Meanwhile, the retrieval tool leverages these sub-queries to localize relevant evidence, relieving the Vid-LLM of direct frame selection and thus addressing the Interpretation–Selection Gap. Finally, the retrieved frames are fed back to the Vid-LLM for sub-query refinement, forming a reflection loop that iteratively improves query interpretation and frame selection. Experiments across multiple benchmarks show that RACER consistently improves long video understanding. Notably, RACER achieves effective frame selection even with limited-capability components, demonstrating that agentic integration enables these components to enhance more capable Vid-LLMs.
Read the full abstract

All authors Yiyang Huang, Yitian Zhang, Yizhou Wang, Jianglin Lu, Qihua Dong, Hailing Wang, Huimin Zeng, Mingyuan Zhang, Yun Fu

Quantum Physics

An efficient variational polaron master equation for non-Markovian spin-boson dynamics: Transformed initial states and general observables

Chien-Chang Chen and Hsi-Sheng Goan

We investigate the effects of initial states on the dynamics of a quantum spin-boson model using an efficient variational polaron master equation approach. The transformation of a factorized system-environment initial state into the variational polaron frame introduces inhomogeneous terms in the master equation, essential for accurately describing non-Markovian dynamics beyond the weak coupling regime. To efficiently account for these effects, we formulate the dynamics in an extended auxiliary Liouville space, converting the resulting integro-differential equations into coupled ordinary differential equations. We further introduce a simplified variational polaron technique for computing the time evolution of expectation values of general observables, including those that do not commute with the polaron transformation, overcoming a major limitation of previous approaches and enabling the unified treatment of coherence and population dynamics. The effects of transformed initial states generated from arbitrary factorized initial states are systematically characterized through expectation values and trace-distance analysis, providing insight into the associated bath-relaxation dynamics over a broad range of tunneling amplitudes, temperatures, and coupling strengths. Comparison with the weak-coupling master equation reveals a delay phenomenon associated with multi-bath-quanta processes, pronounced at intermediate coupling strengths and tunneling amplitudes. We further show that coherent superposition initial states induce stronger nonequilibrium effects than localized states, with inhomogeneous terms playing a crucial role in the short-time dynamics. The proposed formalism applies to both undriven and resonantly driven spin-boson models, providing a computationally efficient framework for open quantum-system dynamics with arbitrary factorized initial states and general physical observables.
Read the full abstract

Solar and Stellar Astrophysics

Turbulent viscosity in line-driven stellar winds

Matías Montesinos, Michel Cure, Ignacio Araya et al. (+1 more)

Radiation-driven winds from massive stars generate powerful outflows known to be intrinsically turbulent. Despite this, standard hydrodynamic models have historically restricted the application of macroscopic turbulent viscosity to rotationally supported disks, neglecting its dynamic role in the supersonic expansion regime of the wind. We evaluated the dynamical relevance of macroscopic turbulent momentum transport in accelerating line-driven stellar outflows of rotating B supergiants. We introduced a generalized strain-limited viscous prescription within a stable, time-dependent 1D m-CAK hydrodynamic framework using the FARGO3D code. We compared inviscid, constant-viscosity, and strain-limited models to analyze their effects on the radial velocity, density stratification, mass-loss rate, and angular momentum transport. The strain-limited prescription produces a radially structured viscosity that primarily acts as a broad radial braking term. With the viscosity bounded through a prescribed maximum mixing length, the viscous acceleration remains subdominant to the total radiative acceleration but is dynamically relevant to the advective acceleration. In the strongest-viscosity model, the terminal velocity decreased by $\sim 39\%$ and the mass-loss rate by $\sim 58\%$ relative to the inviscid model in the stationary state. The flow remains radiatively driven, and angular momentum transport remains dominated by advection, maintaining a quasi-conserving $v_\phi \propto r^{-1}$ scaling. Turbulent viscosity exerts a measurable dynamic effect on supersonic line-driven winds. When bounded by the local macroscopic strain, it modifies the radial momentum balance and the resulting mass-loss rate without forcing the wind into a disk-like regime of strong viscous angular momentum redistribution.
Read the full abstract

All authors Matías Montesinos, Michel Cure, Ignacio Araya, Roberto O. J. Venero

Cryptography and Security

ASPIRE: Agentic Safety & Prompt Injection Red-teaming Engine

Pengfei He, Deep Mitra, Vishesh Sharma et al. (+4 more)

LLM agents retrieve untrusted content and act through tools, creating indirect prompt-injection risks that can cause unauthorized actions or persistent state changes. Existing automated red-teaming largely optimizes payloads for pre-specified scenarios, leaving latent vulnerabilities across the agent's behavior space unexplored. We present ASPIRE, an Agentic Safety & Prompt Injection Red-teaming Engine for open-ended, behavior-level vulnerability discovery. ASPIRE maintains an evolving Agent Security Behavior Graph and uses complementary Explore and Exploit experts to discover, verify, and generalize consequence-centric tests. Trajectory evidence updates the graph and diagnoses partial or failed attempts, while cross-run memory transfers useful red-team strategies. Experiments on various benchmarks show that ASPIRE substantially expands coverage across consequences, injection methods, environments, and behavior paths while maintaining strong attack success.
Read the full abstract

All authors Pengfei He, Deep Mitra, Vishesh Sharma, Jiliang Tang, Vinay S Rao, Tomas Pfister, Long T. Le

High Energy Physics (Phenomenology)

Resonance-aware electroweak Sudakov reweighting for multijet-merged predictions

Lenart Jerala, Rikkert Frederix, Miha Muškinja et al. (+3 more)

We present a resonance-aware extension of electroweak Sudakov-logarithm reweighting in MadGraph5_aMC@NLO for NLO QCD-matched and FxFx multijet-merged event samples. The method identifies the hard production process underlying each event and evaluates its electroweak Sudakov correction while accounting for the spin correlations between unstable particles and their decay products. Photonic logarithms are removed from the virtual weight to avoid overlap with QED shower evolution. We validate the method in FxFx-merged $Z$+jets production and fully leptonic $W^+W^-$ production. The reweighting yields the expected growing weak suppression in high-energy tails and is stable under variations of the clustering and merging parameters. In $W^+W^-$ production, the density-matrix and spin-summed prescriptions are statistically compatible for the observables studied, indicating only a small additional helicity dependence of the Sudakov correction in this channel. The method provides a practical virtual weak-Sudakov approximation for matched and merged simulations, but does not constitute an exact NLO electroweak calculation.
Read the full abstract

All authors Lenart Jerala, Rikkert Frederix, Miha Muškinja, Davide Pagani, Timea Vitos, Marco Zaro

Combinatorics

Primitive periods in greedy chip-firing games

Zheng Huang and Anyuan Tian

For a finite irreducible rational transition matrix $P=(P_{ij})$ on $n$ states, the hunger game introduced by Li and Propp is a greedy chip-firing process in which each state $i$ carries a real-valued hunger $h_i$. At each step, a state $i$ of maximal hunger is selected, with ties resolved by a fixed order on the states. Firing $i$ subtracts one from its hunger and then adds $P_{ij}$ to the hunger of each state $j$. Let $\pi$ be the stationary distribution of $P$, and let $T$ be the least positive integer such that $T\pi$ is integral. We prove that every periodic orbit has least period $T$, with state $i$ firing $T\pi_i$ times, and that the set of periodic states of total hunger zero tiles the corresponding hyperplane by lattice translations. These results settle two conjectures of Li and Propp. We also prove an analogous least-period theorem for nontrivial periodic orbits of fixed-priority chip-firing on finite strongly connected digraphs, another greedy chip-firing model. We then find a connection between the hunger game and the chairman assignment problem: the greedy rule that always chooses the state furthest below its proportional target is precisely a rank-one hunger game. Motivated by this application and related allocation rules, we generalize the hunger game by assigning each state a score that is a nondecreasing function of its hunger and firing a state of maximal score each time. For these monotone-score hunger games, we prove that every periodic orbit still has least period $T$ and give a sufficient condition for eventual periodicity from every initial state.
Read the full abstract

Systems and Control

Quantifying Grid-Forming Requirements for System Strength: How Much and Where?

Ambuj Gupta, Balarko Chaudhuri and Mark O'Malley

As power systems transition toward inverter-based resource (IBR)-dominated grids, Grid-ForMing (GFM) IBRs are increasingly expected to provide the voltage-source characteristics required to maintain adequate system strength. However, how many GFM IBRs are required, and where they should be located in the grid, remains unclear. This paper investigates the impact of the number and location of GFM IBRs on small-signal system strength. The Jacobian Voltage Stiffness Metric (JVSM), which quantifies the voltage-magnitude and voltage-phase stiffness from frequency-domain Jacobian characteristics, is used as a system-strength measure. The analysis demonstrates that increasing the number of GFM IBRs generally improves system strength; however, configurations with the same number of GFM IBRs can exhibit significantly different bus-wise system-strength characteristics depending on their locations. Thus, this paper provides a quantitative basis for specifying GFM requirements in terms of the resulting system strength, rather than solely in terms of penetration level or the number of GFM IBRs. Case studies on a modified IEEE 39-bus system demonstrate how JVSM can be used to quantify and compare the system-strength impact of different numbers and locations of GFM IBRs, with the resulting strength assessments validated against small-signal oscillatory stability.
Read the full abstract

Cosmology and Nongalactic Astrophysics

Measuring the Clustering of tSZ-Selected Galaxy Clusters with SPT-3G

E. S. Martsen, D. Anbajagane, L. E. Bleem et al. (+96 more)

We present a measurement of the spatial clustering of galaxy clusters selected via the thermal Sunyaev-Zel'dovich (tSZ) signal. Using a cluster catalog derived from five years of SPT-3G Main field survey data, we define a sample spanning $0.25 < z < 1.1$, as well as two subsamples spanning $0.25 < z < 0.67$ and $0.67 < z < 1.1$. We detect cluster clustering at $12.8\sigma$, the highest significance detection in a tSZ-selected sample to date. We detail our analytic modeling and theoretical covariance models, including validations performed on synthetic data created from numerical simulations. The measurements and inference pass a variety of null tests, and show minimal sensitivity to common spatially varying systematics that impact optical clustering measurements, such as depth, stellar density, and extinction. We use our pipeline to place constraints on the halo bias of our sample, $b_{\rm{h}} = 5.30\pm{0.24}$, and the normalization of our mass-observable relation, $\rm{lnA}_{sz} = 0.43^{+0.17}_{-0.21}$. We also demonstrate the useful synergies between cluster clustering and abundance measurements, which manifest in the breaking of high-dimensional parameter degeneracies. Samples of clusters selected using the tSZ effect serve as a unique probe of clustering that is insensitive to common systematics, relative to conventional galaxy datasets, and the detection significance of this measurement will increase substantially with existing data from the SPT-3G Ext-10k survey and upcoming data from SPT-3G+ and the Simons Observatory.
Read the full abstract

All authors E. S. Martsen, D. Anbajagane, L. E. Bleem, B. A. Benson, E. Anderes, A. J. Anderson, B. Ansarinejad, M. Archipley, L. Balkenhol, D. R. Barron, P. S. Barry, K. Benabed, A. N. Bender, F. Bianchini, S. Bocquet, F. R. Bouchet, E. Camphuis, M. G. Campitiello, J. E. Carlstrom, J. Carron, C. L. Chang, P. M. Chichura, A. Chokshi, T.-L. Chou, A. Coerver, T. M. Crawford, C. Daley, T. de Haan, K. R. Dibert, M. A. Dobbs, M. Doohan, D. Dutcher, C. Feng, K. R. Ferguson, N. C. Ferree, K. Fichman, A. Foster, N. Frontiere, S. Galli, A. E. Gambrel, A. K. Gao, F. Ge, S. Gradis, F. Guidi, S. Guns, N. W. Halverson, E. Hivon, G. P. Holder, W. L. Holzapfel, J. C. Hood, A. Hryciuk, N. Huang, T. Jhaveri, F. Kéruzoré, A. R. Khalife, M. Klein, L. Knox, K. Kornoelje, C.-L. Kuo, P. Larsen, K. Levy, Y. Li, A. E. Lowitz, C. Lu, G. P. Lynch, T. J. Maccarone, A. S. Maniyar, F. Menanteau, M. Millea, J. Montgomery, Y. Nakato, T. Natoli, Y. Omori, A. Ouellette, Z. Pan, P. Paschos, K. A. Phadke, K. Prabhu, W. Quan, M. Rahimi, A. Rahlin, C. L. Reichardt, M. Rouble, J. E. Ruhl, A. C. Silva Oliveira, A. Simpson, J. A. Sobrin, A. A. Stark, J. Stephen, C. Tandoi, C. Trendafilova, J. D. Vieira, A. G. Vieregg, A. Vitrier, Y. Wan, N. Whitehorn, W. L. K. Wu, M. R. Young, J. A. Zebrowski

Computational Engineering

MarketRank: ranking stocks by the stationary probabilities of a Markov chain over estimated money flows, with a topological landscape view

B. Kaan Karamete

Every trading day, money leaves some stocks and moves into others. MarketRank asks where that money concentrates and how the concentration shifts over time. It treats the market as a Markov chain whose states are stocks. The chance of moving from one stock to another is the share of the first stock's outgoing dollars that goes to the second, and a stock's MarketRank score is its long run (stationary) probability in that chain, computed by damped power iteration on a sparse transition matrix. PageRank ranks web pages with the same idea, using links instead of dollars. Who sold what to buy what is not public, so the flows are estimated from daily prices and volumes with a correlation tilted gravity model, in a deterministic parallel C++ engine. On a universe of 10K US stocks and ETFs, the ranking is led by the large semiconductor and technology names, and estimated flows stay within a sector more often than chance. The level of the score, however, does not predict next day returns: it shows where money sits, not where prices go next. A 3D landscape view, Fluxscape, turns the ranking into terrain: Louvain communities of the recent flow graph, reclustered on every bar, become territories, ordered by recursive spectral bisection along a generalized Hilbert curve, interpolated by inverse distance weighting, smoothed by a centroidal Voronoi style relaxation with portfolio holdings pinned to their exact values, and rendered as a Gouraud shaded mesh. The surface is for reading trends; the exact ranked table remains the reference. The practical question is whether MarketRank can improve real portfolio moves, and we answer it with tests fixed in this study.
Read the full abstract

Quantum Physics

Efficiently computable bounds on the energy-constrained quantum reading capacity

Vishal Singh, Zixin Huang and Mark M. Wilde

In quantum reading, classical messages are encoded in sequences of quantum channels and recovered by probing the channels and processing their outputs. We study the reading capacity of a finite family of finite-dimensional channels under an average constraint on the probing energy, allowing arbitrary adaptive operations between channel uses. Using an input-dependent chain rule for the Belavkin-Staszewski relative entropy, we derive a converse bound expressed as an optimization involving channel Choi operators and the operator relative entropy. This formulation admits semidefinite approximations and reduces, without an energy constraint, to a channel information radius. To obtain achievable rates, we construct bilinear semidefinite lower approximations to a standard non-adaptive reading bound. Alternating optimization produces feasible probe states and encoding distributions, whose Holevo information gives a directly evaluable achievable rate. For jointly classical-quantum channels, we derive a separate converse based on the Umegaki relative entropy. Without an energy constraint, this converse matches a non-adaptive achievable rate, recovering a recent result of Pascual Abraldes and Winter: non-adaptive protocols suffice to achieve the unconstrained reading capacity of jointly classical-quantum channels.
Read the full abstract

Computer Vision

One-Slide Calibration of Pathology Foundation Models

Ming Ren Hou and Tianyi Huang

Scanner variation changes how pathology foundation models represent the same tissue. We introduce SlideRuler, which uses other regions within a slide as internal controls to predict a query region's scanner-induced shift. A transfer map learned from paired rescans enables calibration from a single scan at inference while keeping the foundation model fixed. Across two encoders and five SCORPION scanners, learned transfer reduces mean target-to-source embedding distance by 16.3-38.5% relative to raw embeddings. Replacing controls with unrelated same-scanner regions reveals a positive same-slide contribution in all four evaluation settings, including scanner holdout. A source-anchored variant reduces source-feature displacement by 47.7-83.6% relative to learned transfer while retaining most of its alignment gain. By drawing calibration information from the slide itself, SlideRuler offers a path toward more consistent use of frozen pathology models across imaging systems.
Read the full abstract

Human-Computer Interaction

"I'm Very Happy for It to Start Hallucinating a Little Bit": Using ClayFlect to Negotiate Multimodal AI Representations in Material Meaning-Making

Kellie Yu Hui Sim, Quoc-Nam Nguyen, Shuenn Yuen Han et al. (+1 more)

As AI enters reflection and emotional support, understanding how it can participate in personal meaning-making while preserving users' authority over interpretation is increasingly important. We present ClayFlect, a novel MLLM-powered system integrating tactile clay-making with conversational and visual generative AI, and report a mixed-methods study with 50 participants. Reflection developed across material, conversational, and generated forms rather than through AI interaction alone. Participants treated AI representations as provisional: they compared, redirected, reinterpreted, selectively incorporated, or left them aside as their artefacts and meanings evolved. Clay provided a directly manipulable space in which participants could continue developing meaning independently of the AI, while generated representations externalised possibilities beyond what they could readily make or visualise. We show how generative AI can participate through representations that remain negotiable, and derive implications for supporting movement across representations, preserving parallel sites of control, and allowing AI support to recede or deepen as reflection unfolds.
Read the full abstract

All authors Kellie Yu Hui Sim, Quoc-Nam Nguyen, Shuenn Yuen Han, Kenny Tsu Wei Choo

Quantum Gases

Matter wave bistability with a momentum chirped Bose-Einstein condensate

Mikhail Mamaev, Nicholas Mantella, Noah P. Baker et al. (+2 more)

While a Fabry-Perot (FP) cavity interferometer is a standard tool in modern optics, analogous physics can also be observed with matter waves such as Bose-Einstein condensates (BECs). Interatomic collisions in BECs can induce nonlinear features such as multi-valued transmission spectra for interferometry experiments. Observation of interaction effects in the matter-wave analogue of FP interferometry has not yet been experimentally achieved, mostly due to broadening of transmission peaks by interaction-induced shifts of cavity resonances. In this work, we theoretically show that this limitation can be overcome by engineering a momentum chirp in a BEC wavepacket. By balancing this chirp against the dynamical mean-field energy shift of the cavity during transmission, we predict enhanced resonant spectra and bistability features due to interaction-assisted tunneling that would be otherwise inaccessible. We show that this physics is readily observable with quasi-one-dimensional BECs of cold atoms colliding with a pair of light-induced Gaussian potential barriers forming the cavity. We also benchmark the requisite optimum chirp needed and characterize the contrast of near-discontinuous FP transmission spectra at an interacting resonance. Our findings enable the realization of technologies like matter-wave switching protocols, and high-precision velocity measurements; estimates of achievable sensitivity for the latter are provided.
Read the full abstract

All authors Mikhail Mamaev, Nicholas Mantella, Noah P. Baker, Joseph McGowan IV, Aephraim M. Steinberg

Computer Vision

SPW-Nav: A Streaming Panoramic World Model for Language-Guided Navigation

Yunheng Liu, Ziqi Cai, Siqi Yang et al. (+7 more)

Language-guided panoramic video generation benefits various downstream applications, such as interactive 3D scene exploration, virtual reality experiences, and embodied agent training. Existing panoramic generators follow predefined trajectories, and interactive world models act through low-level actions in perspective views. We propose SPW-Nav, a streaming panoramic world model that understands movement instructions and streams one minute of 2K 360-degree video in real time from a single panorama. SPW-Nav interprets each instruction in the previously generated panorama as camera motion. Spherical rotation decoupling applies rotation exactly on the sphere, pose-aligned conditioning keeps translation inputs bounded over long streams, and a multi-term memory with a few-step generator continues the scene as instructions change. We also build SPW-NavSet, panoramic videos with camera trajectories and verified instructions. Driven by language, SPW-Nav outperforms prior panoramic generators in camera-following accuracy and video quality, and supports on-the-fly instruction switching.
Read the full abstract

All authors Yunheng Liu, Ziqi Cai, Siqi Yang, Yimu Wang, Minggui Teng, Jiaming Tan, Shuchen Weng, Erwin Wu, Kaipeng Zhang, Boxin Shi

High Energy Astrophysical Phenomena

Predictions for Short Gamma-Ray Burst Detections by COSI

Eliza Neights, Eric Burns, Nicolò Parmiggiani et al. (+21 more)

The Compton Spectrometer and Imager (COSI) is a soft gamma-ray telescope scheduled to launch in 2027. COSI will trigger onboard the spacecraft in response to short-duration transients, including gamma-ray bursts, and it will promptly disseminate alerts which include localizations to the community. This will enable rapid follow-up observations of the afterglow and associations with transients identified by other instruments. By simulating COSI observations of short gamma-ray bursts (sGRBs) and processing them with the trigger algorithm which will run onboard the spacecraft, we estimate the sensitivity of COSI to sGRBs. We then convolve this with a simulated set of sGRBs representing the intrinsic population in the universe to predict the rate that will be detected by COSI. We find that COSI is expected to trigger on $\sim$47-89 sGRBs per year with $\sim$7-10 localized to within 2.5{\deg}. With the gravitational wave sensitivity anticipated in the O5 run, we predict 0-3 joint detections with gravitational waves.
Read the full abstract

All authors Eliza Neights, Eric Burns, Nicolò Parmiggiani, Yong Sheng, Savitri Gallego, Aldo Aguilar-Nadalini, Aaron C. Trigg, Sylvain Guiriec, Koothodil Abhijith Augustine, Matthew G. Baring, Anna L. Butterworth, Hsiang-Kuang Chang, Che-Yen Chu, Niccolò Di Lalla, Christopher Fryer, Hadar Lazar, Israel Martinez-Castellanos, Marion Sudvarg, Péter Veres, Steven E. Boggs, Dieter H. Hartmann, Carolyn A. Kierans, John A. Tomsick, Andreas Zoglauer

High Energy Physics (Theory)

Non-supersymmetric M-theory vacua and dark bubbles

Vincent Menet, Alessandro Tomasiello and Vincent Van Hemelryck

An expanding bubble in AdS$_5$ has a dS$_4$ geometry; the dark bubble scenario explores whether this can be our Universe. Motivated by this, we consider new non-supersymmetric AdS$_5$ compactifications of M-theory and investigate their non-perturbative instabilities. We focus on a compactification on the flag manifold. In certain regions of the flux landscape, the non-supersymmetric vacua become unstable under the nucleation of M5-brane bubbles wrapping an internal two-cycle, thus realising the dark bubble scenario. The cosmological constant on the bubble can be made parametrically small by tuning the internal fluxes, making the M5-branes close to the extremality bound. The solution relates the induced cosmological constant on the brane and the typical largest length scale of the extra dimensions through $R^{-1}\sim \Lambda_4^{1/4}$ in 4d Planck units, allowing for extra dimensions of the order of tens of microns. We also consider various alternatives, which for various reasons are currently not as successful.
Read the full abstract

High Energy Physics (Theory)

Root systems and Minimal RG flows to c=1 CFTs

Changrim Ahn, Sara Kostadinovic and Francesco Ravanini

We derive exact massless S-matrices from algebraic root-system geometry and derive thermodynamic Bethe ansatz (TBA) equations that generate minimal renormalization group (RG) flows. We show how P. Dorey's geometric construction, originally developed for massive S-matrices, can be extended to massless cases and prove that these new $S$-matrices straightforwardly satisfy the bootstrap equations. We focus on minimal D_n coset CFTs, which are c=1 free-boson CFTs with compactification radius \sqrt{n/2}, and find their UV completions by analyzing the TBA.
Read the full abstract

Biomolecules

A Shortcut to Structure in AlphaFold 3

Jonathan Feldman and Jeffrey Skolnick

AlphaFold 3 predicts protein structures with remarkable accuracy, yet how structural information emerges within the model remains poorly understood. Here, through causal interventions on internal representations and direct probing of every Pairformer block, we trace the formation of global protein geometry and identify the multiple sequence alignment (MSA) as a structural shortcut to the fold. Removing the MSA largely preserves local secondary structure while disrupting the long-range relationships that define global topology. Restoring the MSA-enriched pair representation at only forty residues recovers most of this lost organization, including at pairs never directly modified. This contribution depends on the detailed direction of the MSA module's output rather than its magnitude. The Pairformer rapidly converts this signal into global geometry: the final fold becomes recoverable by approximately block 9 of 48 for a majority of proteins, roughly twenty-seven blocks before the model's decoder can render it, whereas without the MSA it remains inaccessible for most proteins throughout the pass. Which homologs are supplied shapes this trajectory more strongly than which query is supplied; it persists for a designed query that never evolved but collapses for a shuffled sequence. Most importantly, an alignment built for a different protein that shares the fold, supplied only at the structurally corresponding columns, raises the median TM-score against experiment from 0.44 to 0.72, while the same alignment shifted a few residues along the chain performs worse than supplying no alignment at all. What AlphaFold 3 reads from an alignment is therefore a description of the fold itself, transferable between proteins that share one, rather than the query's own evolutionary history. This explains both its accuracy and the limits of what it has solved.
Read the full abstract

Computer Vision

VCR-Bench: A Modular Open-Source Benchmark for Video Classification Robustness

Maksim Plinskiy, Aleksandr Gushchin, Sergey Lavrushkin et al. (+2 more)

Robustness of image classification has several benchmarks, but their video counterparts are absent. In video classification temporal dimension introduces additional degrees of freedom for adversarial attacks, defenses, and preprocessing. Temporal sampling, perturbation budgets, and metric aggregation also interact in ways with no direct analogue in the image setting. Therefore, robustness for video classifiers is studied across scattered, incompatible implementations, making reported numbers hard to reproduce and analyze. We introduce VCR-Bench, a modular open-source benchmark framework that standardizes video loading, wrappers for classifiers, adversarial attacks and defenses, perceptual metrics, configuration presets, and result logging. VCR-Bench currently integrates 30 video classification models, 14 adversarial attacks, and 10 defense wrappers under a common evaluation protocol. We evaluate representative video classifiers, attacks, and defenses on Kinetics-400 subset, reporting clean accuracy, attack success rate, perceptual quality, runtime, and memory usage. VCR-Bench is released with documented installation, reproducible run presets, component-extension interfaces, and scripts for reproducing the reported results at https://github.com/msu-video-group/vcr-bench.
Read the full abstract

All authors Maksim Plinskiy, Aleksandr Gushchin, Sergey Lavrushkin, Dmitriy S. Vatolin, Anastasia Antsiferova

Algebraic Geometry

The logarithmic generalised Kummer locus and its tropicalisation

Patrick Kennedy-Hunt

We construct a separated logarithmic compactification of a divisor in the moduli of polarised irreducible holomorphic symplectic (IHS) varieties of generalised Kummer type in dimension at least four. This compactification carries a universal family of logarithmic IHS varieties, and its tropicalisation carries a universal family of tropical generalised Kummer varieties. The compactification is built from a more general construction of explicit, projective, and logarithmically smooth models for type~II and type~III degenerations of generalised Kummer varieties of arbitrary dimension and over base logarithmic schemes of any dimension. Over a base of any logarithmic rank, these degenerations come with a tropicalisation map to a family of integral affine manifolds with singularities, and with boundary in the type II case. This tropical family is constructed through a tropical generalised Kummer construction. Our approach uses logarithmic Hilbert schemes to extend the conventional generalised Kummer construction to families of logarithmic abelian varieties. We describe fibres of our degenerations, and thus every point in the boundary of our compactification. Proper models correspond to polyhedral structures on the associated tropical generalised Kummer variety, with the resulting polyhedral complex serving as the dual intersection complex; each irreducible component is constructed by applying the generalised Kummer construction to a product of logarithmic group compactifications of semi-abelian varieties.
Read the full abstract

Superconductivity

Remote magnon sensing detects the universal phase stiffness jump in Bi$_2$Sr$_2$CaCu$_2$O$_{8+\delta}$

Yue Sun, Mai Nguyen, Naiyuan James Zhang et al. (+14 more)

Phase stiffness governs macroscopic quantum coherence in superfluids and superconductors, providing a sensitive probe of condensate depletion, fluctuations, and phase transitions. However quantitative and spatially resolved measurements remain challenging. Here we report a contactless, local, all-optical measurement of superconducting phase stiffness using propagating antiferromagnetic magnons as remote sensors. In CrSBr, strong exciton-magnon coupling enables optical excitation and detection of magnon wavepackets and precise determination of their group velocity. When CrSBr is interfaced with underdoped Bi$_2$Sr$_2$CaCu$_2$O$_{8+\delta}$, we find that the magnon group velocity is substantially enhanced below the superconducting transition and quantitatively encodes the phase stiffness through Meissner screening currents. Using this capability, we identify a parallel-field-induced Berezinskii-Kosterlitz-Thouless transition through direct observation of the universal phase-stiffness jump, revealing a field-controlled route to expose intrinsic two-dimensional phase-ordering physics in a bulk cuprate. Our work establishes propagating magnons as local probes of superconducting electrodynamics and phase-coherent phenomena in low-dimensional quantum materials.
Read the full abstract

All authors Yue Sun, Mai Nguyen, Naiyuan James Zhang, Dingbin Huang, Shuai Yuan, Aljoscha Söll, Arun Ramanathan, Takashi Taniguchi, Kenji Watanabe, Xavier Roy, Zdeněk Sofer, Shigeyuki Ishida, Hiroshi Eisaki, Makoto Hashimoto, Zhi-Xun Shen, Joseph Orenstein, Xiaodong Xu

Robotics

Toward Evidence-Driven Human-Agent-Robot Teaming for Earth-Independent Anomaly Triage

Ignacio G Lopez-Francos, Alexis Gallagher and Samira Shalal

Deep-space crews cannot rely on real-time ground support for urgent off-nominal events. Initial alerts may underdetermine cause, while discriminating evidence may reside in crew observations or at locations that are unsafe, costly, or unavailable for crew inspection. We present an evidence-driven architecture for human-agent-robot teaming in Earth-independent anomaly triage. Agentic AI is treated as a stateful coordinator over bounded, inspectable services rather than as a fully autonomous vehicle controller. A triage state manager maintains hypotheses, evidence provenance, uncertainty, operational context, and tool status; a crew-facing embodied agent elicits observations and explains assessment changes; and a mobile robot acquires targeted, localized evidence. Typed interfaces separate dialogue and orchestration from monitoring, robot command, context retrieval, and safety-critical control. Two scenarios illustrate the architecture: a crewed deep-space mission based on an actual ISS ammonia false alarm, where suspected contamination restricts crew access, and a power-interface anomaly at a crewed lunar base, where robotic inspection distinguishes a local connector fault from other causes ambiguous in remote telemetry. Our main contribution is an authority-bounded closed evidence-loop architecture, exercised in a hardware-in-the-loop integration prototype using Reachy Mini and an Innate MARS mobile robot.
Read the full abstract

Information Theory

A Unified Framework for Characterizing General MIMO Channels

Zeyan Zhuang, Anzheng Tang, Xin Zhang et al. (+2 more)

Modern MIMO systems are evolving toward higher-dimensional and more flexible architectures, such as distributed and holographic MIMO, offering substantial benefits while giving rise to increasingly complex channel correlation structures. This growing architectural diversity makes it difficult to characterize the fundamental limits of different MIMO systems on a case-by-case basis, motivating a unified analytical framework applicable across a broad range of architectures. This paper addresses this research gap by investigating a generally correlated Rayleigh channel whose vectorized form follows a general Gaussian distribution with an arbitrary covariance matrix. For that purpose, we first prove that the spectral norm of the channel matrix is bounded in the asymptotic regime where the numbers of transmit and receive antennas grow proportionally. We then derive a deterministic approximation for the ergodic mutual information, with an explicit convergence rate governed by the structure of the channel correlation. The approximation is characterized by a pair of matrix-valued self-consistent equations, for which we establish the existence and uniqueness of the solution and propose an iterative numerical algorithm. Furthermore, we develop a framework based on positive linear maps to analyze the stability of these equations, which can facilitate the spectral analysis of broader classes of random matrices. The proposed characterization unifies many existing results for structured MIMO channels as special cases while remaining applicable to a broad class of general MIMO architectures that are difficult to evaluate using existing analytical frameworks. To demonstrate its utility, we apply the developed theory to two representative systems, namely downlink distributed MIMO and uplink heterogeneous MIMO. Numerical results confirm the accuracy of the derived deterministic approximations.
Read the full abstract

All authors Zeyan Zhuang, Anzheng Tang, Xin Zhang, Dongfang Xu, Shenghui Song

Astrophysics of Galaxies

Early onset of the hot circumgalactic medium around Milky Way galaxies

Víctor Rufo-Pastor, Oscar Agertz, Santi Roca-Fábrega et al. (+6 more)

Hot coronae are key regulators in Milky Way (MW)-type galaxy formation and are thought to emerge once a galaxy's dark matter halo crosses a critical mass threshold, $M_{\mathrm{vir}}\sim10^{12} {\mathrm{M}}_\odot$, so that accretion shocks can heat infalling gas to virial temperatures, $T_{\mathrm{vir}}$. Here we investigate the roles of stellar feedback and accretion shocks in establishing the circumgalactic medium (CGM) around MW-mass galaxies using the VINTERGATAN cosmological simulation suite. We find that stellar feedback drives the early formation of a hot CGM, with $T_{\mathrm{CGM}}\sim T_{\mathrm{vir}}$. By contrast, in feedback-free simulations, accretion shocks alone do not produce a CGM with comparably high average temperatures at early epochs ($z>2$), independent of gas metallicity. We show that CGM virialisation (defined when halo-gas cooling time exceeds the free-fall time) begins with the emergence of a hot corona ($T>10^5$ K) that can appear as early as $z\approx3$ at virial radius, $R_{\mathrm{vir}}$, when feedback is included. This is much earlier than predicted by accretion-shock theory or feedback-free models ($z\lesssim1$). While accretion shocks alone produce an inside-out virialisation pattern, feedback can alter this behaviour, driving a rapid transition ($\lesssim0.5$ Gyr) that is either outside-in or nearly simultaneous throughout the halo. Finally, these early-forming coronae are largely stable from formation through to the present day, but can appear to evolve simply because the definition of $R_{\mathrm{vir}}$ grows with time. This 'CGM pseudo-evolution' is analogous to the pseudo-evolution of dark matter haloes and should be taken into account when interpreting the evolution of hot coronae.
Read the full abstract

All authors Víctor Rufo-Pastor, Oscar Agertz, Santi Roca-Fábrega, Martin P. Rey, Joss Bland-Hawthorn, Justin I. Read, Thor Tepper-García, Corentin Cadiou, Anatole Storck

Solar and Stellar Astrophysics

Re-calibrated analytic stellar evolution formulae for the main sequence: Extension to massive stars and variations in $[\alpha/\mathrm{Fe}]$

Conor M Byrne, Elizabeth R Stanway and Jan J Eldridge

Analytic formulae are an efficient way to approximate the evolution of a star of a given mass and chemical composition, and are widely used in the literature to compute large grids of evolutionary models for stellar population synthesis. Such formulae rely on calibration to detailed stellar evolution models, and existing versions are based on a sparse, limited grid in comparison to modern simulations. By calibrating against the latest generation of single-star evolution models calculated within the Binary Population and Spectral Synthesis (\bpass) framework, we revise and extend pre-existing equations to include the evolution of massive ($M\gtrsim50\,$\msun) stars and changes in chemical abundances in the form of variations in \afe. We introduce additional terms which allow for a better approximation of the main-sequence lifetimes, and the luminosities and radii at the beginning and end of the main sequence. These revised equations describe the main-sequence age, zero-age and terminal-age radius and luminosity with mean errors of less than 10 per cent or 0.05 dex, for all except the terminal age radius, which shows RMS errors of 20 per cent and 0.1 dex. For \afe\ enriched stellar populations, we advise using a matched-iron value for metallicity in the equations, consistent with the expectation that iron is the dominant opacity source affecting the evolution of main-sequence stars.
Read the full abstract

2610.08930 · 6 Oct · Accepted for publication in MNRAS. 18 pages, 13 figures, 1 table, plus appendices (12 pages, 7 tables 12 figures) Python functions to evaluate the equations are available at https://warwick.ac.uk/fac/sci/physics/research/astro/people/conorbyrne/byrne26_equations.py and a Fortran version is available at https://warwick.ac.uk/fac/sci/physics/research/astro/people/conorbyrne/byrne26_equations.f90

Abstract PDF $EVO ×6$STAR ×5$STELR ×4$MTLC ×1

Distributed Computing

Part of the Strassen algorithm can speed up matrix multiplication in a parallel pebbling game

Aaron Dayton and Thomas E. Baker

We present a novel variant of the Strassen algorithm called the Partial Strassen algorithm, which uses a fraction of the Strassen steps to perform matrix multiplication in less time than traditional implementations. The memory footprint required to implement this algorithm is provably small whether used with a single thread or in a multi-threaded context. Data comparing the Partial Strassen algorithm at depths one, two, and three with BLAS matrix multiplication show that the three-level Partial Strassen algorithm is able to perform matrix multiplication in $80\%$ the time of BLAS with $2.25$ times the memory requirement, with a theoretical improvement of $75\%$ for sufficiently large matrices. The Partial Strassen algorithm is also compared against an implementation of the traditional Strassen algorithm, and is shown to be more performant with increasing matrix size due to decreased memory consumption. An open source implementation of the Partial Strassen algorithm for arbitrary depth and rectangular matrix multiplication is also included.
Read the full abstract

Cryptography and Security

Multi-Aspect Runtime Verification for Simulation-Based V&V of LLM-Enabled Autonomous Agents

Nikolaos Kekatos, Dimitrios Nikou, Anastasios Temperekidis et al. (+4 more)

LLM-based agents are entering decision-support roles in defence staff work, where the obligations they must respect are already written down and binding, and where retraining is not available as a control because models arrive as procured components. What can be placed under engineering control is the interface between the agent and the systems it acts on. Those obligations are at once spatial, temporal and text-semantic, and a violation typically lives in the composition of a multi-step interaction, which is why per-event guardrails miss sequential tool-attack chains. We present a multi-aspect runtime-verification framework that decomposes a natural-language policy clause into a typed spatial/temporal/semantic triple over one canonical event stream, checks each aspect with its own monitoring specification, and fuses the verdicts through a four-valued algebra that carries provenance. The spatial aspect is interpreted over a weighted two-sorted location graph in which mission geometry and information-release topology are one object; we show that these spatial obligations are not in general subsumed by a first-order temporal specification. The past-time aspect runs on the unmodified MonPoly engine, which agrees with our reference monitor at every time point. Across two mission domains, casualty evacuation and contested sustainment, and one civil domain, composition under the precautionary blocking policy drives attack success to zero with no observed false positives and microsecond-scale per-event cost, while every single aspect and every pair leaves a substantial share of attacks succeeding. In a closed-loop experiment a policy-naive planner reaches a violating state in most unshielded missions and in none when shielded, and four refused episodes in five still recover to a compliant outcome.
Read the full abstract

All authors Nikolaos Kekatos, Dimitrios Nikou, Anastasios Temperekidis, Alexios Lekidis, Nikolaos Kolokotronis, Panagiotis Katsaros, Stylianos Basagiannis

Artificial Intelligence

Can AI Agents Make Open-Ended Scientific Discovery? Evidence from Station

Wenyu Du and Stephen Chung

Recent AI systems have made rapid progress in scientific discovery when given well-defined metrics, but whether they can autonomously undertake open-ended scientific discovery remains unclear. We investigate AI's ability to tackle open-ended tasks in Station, an open-world environment in which multiple agents simulate a scientific ecosystem. To tackle challenges specific to open-ended tasks, we propose augmenting Station with two mechanisms: a Supervisor mechanism and periodic Meta Reflection, which encourage persistent exploration even when intermediate metrics are lacking. We construct open-ended tasks from three recent oral papers presented at ICLR. We give agents the main research question studied in each paper while withholding the paper's results and disabling web access. We then measure how many of the original findings-partitioned into individual criteria-agents rediscover. We find that Station rediscovers 62.7% of the criteria on average, compared with 15.4% for Codex Multiagent-v2 and 14.4-20.6% for AI Scientist-v2. Ablation and behavioral analyses indicate that adding the two mechanisms together improves research coverage and continuity. We further evaluate Station on two open-ended tasks without oracle papers and find that some of the discoveries made by the agents closely match discoveries reported by researchers after the knowledge cutoff date. Together, these results indicate that a suitable environment can enable agents to autonomously make meaningful progress in open-ended scientific discovery.
Read the full abstract

Quantum Physics

High-Fidelity Inter-Species Rydberg Gates with Two-Photon Driving

Shuan Wang and Kevin Singh

Recently, neutral atom arrays have become known for their promising potential in quantum information processing. High-fidelity entangling operations are typically achieved by exciting atoms to high-lying Rydberg states through two-photon processes. To improve the performance of these systems at the hardware level, dual-species strategies have been applied to enhance readout, error correction, and reloading in neutral atom systems. However, the influence of laser crosstalk arising from two-photon excitation for dual-species systems has remained largely unexplored in the literature. Here, we develop a model that incorporates the influence of crosstalk in inter-species Rydberg gates with two-photon driving. We demonstrate that the influence of this crosstalk can be suppressed to arbitrarily small levels through appropriate gate design, with gate fidelities greater than $99.8\%$ achievable with conservative experimental parameters in the near-detuned case of an $^{85}$Rb/$^{87}$Rb system. We also outline how this fidelity can be further enhanced through improvements to the Rydberg excitation.
Read the full abstract

High Energy Physics (Theory)

Dimensional uplifts of thermal two-point functions and higher-derivative CFTs

Songyuan Li and Anastasios C. Petkou

We explore the idea that the mass parameter $m$ can be used as a diagnostic tool for identifying the critical points of thermal two-point functions of massive free scalar fields in odd $d$ dimensions. For general values of $m$ these correlators admit an expansion in thermal conformal blocks associated with symmetric traceless operators of definite scaling dimension. For each spin the higher-twist contributions can be resummed into a closed-form dressing factor of the leading-twist coefficient. At special values of $m$ for which the thermal trace of the energy-momentum tensor vanishes they are identified as two-point functions of thermal CFTs. Through a dimensional uplift, valid for all $m$, we map these correlators to thermal two-point functions of higher-derivative massive theories in $d+2$ dimensions. We show that the thermal trace of the energy-momentum tensor of the $d+2$-dimensional theory is proportional to the free energy of the original $d$-dimensional theory. When that thermal trace vanishes for $m\neq 0$ the uplifted correlators are identified as two-point functions of a higher-derivative nonunitary thermal CFT$_{d+2}$, but they differ from the uplifted nontrivial thermal two-point functions of the original unitary CFT$_d$.
Read the full abstract