Peer Revenue Equity in science

Papers

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

All of arXiv

Thursday 5 November

High Energy Physics (Theory)

Dressed Newton-Cartan geometry and Euclidean thermal configurations in 3D

We report a new class of solutions of three-dimensional Galilean Anti-de Sitter gravity. The theory can be formulated as a topological gauge theory invariant under the doubly extended Newton-Hooke group in three dimensions. The solution can be interpreted as a dressed Newton-Cartan geometry, which is endowed with four conserved charges: energy, angular momentum, spin and rest mass. Interestingly, the magnitude of the spin turns out to be bounded from above by the rest mass. The Euclidean continuation of the solution exhibits well–defined thermal properties in terms of its gauge theoretical description. Temperature, angular velocity and other chemical potentials are obtained from suitable regularity conditions associated with the topology of a solid torus.
Read the full abstract

All authors Luis Avilés, Oscar Fuentealba, Diego Hidalgo, Pablo Rodríguez

Systems and Control

Joint Electricity Price and Load Forecasting for Data Center Battery Dispatch Using Polynomial Diffusion Models

The rapid growth of large-scale data centers (DCs) is increasing electricity demand, market volatility, and operating costs. Battery energy storage systems (BESSs) can mitigate these challenges by shifting electricity purchases from high-price to low-price periods. This paper proposes a real-time BESS dispatch framework using polynomial diffusion models (PDMs) to forecast both electricity prices and DC loads. Leveraging the closed-form conditional expectations of PDMs, the forecasts are integrated into a dynamic programming (DP) framework to minimize electricity procurement costs. Across the evaluated BESS configurations, the proposed approach achieves 2.63 times the cost savings of a rule-based (RB) strategy.
Read the full abstract

Machine Learning

Same Behavior, Different Mechanism: What Finite Experiments Can Certify About Integrated Information

Interpretability, AI auditing and assessments of machine consciousness increasingly rest on mechanism-level quantities: functionals of a system's causal transition law that describe how it computes, not what it outputs. Experiments see that law only through finite, partial and noisy interventions. We ask when such a quantity is identifiable from an experiment's access, continuous in the mechanism and certifiable from finite data, and answer exactly for a deliberately demanding case - the integrated information $\varphi$ of integrated information theory - in stochastic networks of four and five units with known ground truth. Behavior can be silent: a recurrent and a feed-forward network with identical input-output laws at their response delay carry 0.28 and 0 bits, yet internal recordings separate them within 4-53 trials. Partial access can be just as blind: in our examples, units left out of a one-step recording impose precision floors of 0.30-1.02 bits that no sample size lowers, and in all-but-self parity networks, recordings of any length see several hidden units only through the product of their fidelities - a symmetry that one intervention broke in exact checks. The measure has sharp edges: because $\varphi$ selects a partition, and the cause and effect it specifies, by one criterion and reports another, infinitesimal changes of mechanism that cross an edge change it by 25-100% at 99% of the edges we found in random networks, and about 1% of surrogate models that fit interventional data almost perfectly misreport it by more than 20%. One-sided claims survive: the same data certify a lower bound on both sides of an edge, and we prove $\varphi>1.004$ bits throughout a 64-dimensional region of mechanisms and certify it from about $10^7$ controlled transitions with false-certificate probability at most $\alpha$. Measurability is thus a design question with computable answers.
Read the full abstract

Optimization and Control

An Accelerated Proximal Gradient Method for Bures-Wasserstein Space

We study composite optimization over the Bures–Wasserstein (BW) space of Gaussian measures. Our main observation is that a useful class of BW problems can be lifted into the Euclidean space of affine transport maps, such that the lifted problems can be solved efficiently using accelerated proximal gradient methods. For objectives whose smooth and nonsmooth components admit convex smooth and convex proximable affine lifts, respectively, our algorithm BW-FISTA achieves minimax-optimal objective gap $O(k^{-2})$ after $k$ iterations. We verify these regularity conditions for potential and interaction energies together with common spectral regularizers, and we treat stochastic gradients under a suitable noise model. As a principal application, we consider Gaussian variational inference with a log-concave, log-smooth target. Here, the resulting entropy term is handled by an efficient, exact proximal step, and we improve the previous rate of $O(k^{-1})$ to $O(k^{-2})$. We numerically verify this predicted boost with synthetic experiments.
Read the full abstract

All authors Jakob Nylöf, Sloan Nietert, Julien Pallage, Giancarlo Ferrari-Trecate, Daniel Kuhn, John Lygeros

Probability

Almost sure bounds and tail phases for critical non-Gaussian holomorphic chaos

Gu and Zhang asked for almost sure upper and lower bounds on the Taylor coefficients $a_n$ of $\exp(\sum_{k\ge1}X_kz^k/\sqrt{k})$ in the unit disc, and for the dependence of these bounds on the input tail. For independent, identically distributed, rotationally invariant complex inputs with $\mathbb E|X_1|^2=1$, we prove the universal almost sure bound $|a_n|=O((\log n)^{1/4})$. Almost surely, $\limsup_{n\to\infty}|a_n|/(\log n)^{1/4}$ is positive and finite exactly when $\mathbb E[|X_1|^2\log^+|X_1|^2]<\infty$, and is zero otherwise. For each input law, we define a deterministic scale such that, almost surely, the coefficients are bounded exactly when the scale is bounded, and tend to zero exactly when it vanishes. We also determine the almost sure logarithmic growth exponent as the maximum of contributions from collective fluctuations and individual large inputs. Explicit tails exhibit distinct thresholds for boundedness and for upper bounds at the scale in the phase criteria. Examples show that neither the leading tail nor the set of finite logarithmic moments determines the almost sure phase.
Read the full abstract

2611.08467 · 5 Nov · In my judgment, the argument in Proposition 5.11 for summing the negative Fourier modes introduced by the mask across dyadic scales is incorrect as written. This step is needed for (5.4), which controls the conditional variance in Section 5.8 and supports the almost-sure upper bound in Theorem 1.1. I request withdrawal pending further review

Data Structures and Algorithms

Canonical Cut Hierarchy in Almost-Linear Time via Reduction to Max-Flow

The canonical cut hierarchy recursively partitions a graph along Tutte-Nash-Williams partitions, choosing the partition with the most number of parts at each step. We give a randomized almost linear-time algorithm for computing this hierarchy in undirected graphs with polynomially bounded capacities via reduction to max-flow. This yields randomized almost linear-time algorithms for computing partition connectivity, strength, both arboricity and fractional arboricity, both integral and fractional spanning tree packing numbers, and implicit representations of the principal partition sequence and a maximum fractional spanning tree packing.
Read the full abstract

Machine Learning

Towards interpreting any residual stream direction by scaling activation inversion

To interpret a direction in a language model's residual stream, we can look for text that strongly activates it. But when such text is found by scanning a corpus, it is limited to what that corpus contains. We develop Maximally-Activating Example Metamodels (MAEMs), language models which map residual stream directions from a subject model to text exemplars that strongly activate those directions. Our objective is to maximise the cosine similarity of a supplied residual stream direction with the activations of the text generated by the MAEM. This label-free RL objective helps to scale MAEMs on diverse sources of residual stream directions. Empirically, we find the resultant metamodel produces informative exemplars: a Qwen3.6-27B MAEM inverts held-out activations better than corpus search or a natural-language autoencoder (NLA), beats corpus search (at 10M tokens) on a majority of held-out languages/domains, and activates 89% of held-out SAE features. It also identifies concepts behind steering vectors about as well as NLAs, recovers the payloads of LoRA backdoors more often than corpus search, and can even verbalise some model workspace inner monologue. Unlike explanations produced by other metamodels (e.g. NLAs), MAEM exemplars can be verified with a forward pass on the subject model, making the MAEM objective a grounded alternative for scaling the training of metamodels.
Read the full abstract

All authors Celeste De Schamphelaere, Juan J. Vazquez, Ari Spiesberger, Tomáš Gavenčiak, Gavin Leech, Euan Ong, Jiahai Feng

Representation Theory

Symplectic singularities in Relative Langlands duality for products of general linear groups

It was shown by Braverman-Finkelberg-Nakajima that the regular perverse sheaf on the affine Grassmannian of type A is isomorphic to the ring object associated to a Gelfand-Tsetlin quiver. We use this result to identify, for any finite dimensional representation $\mathbf{N}$ of a product of general linear groups, the relative Langlands dual $(T^*\mathbf{N})^\vee$ of $T^*\mathbf{N}$ with the Coulomb branch of a suitable representation. By a result of Bellamy, this implies that $(T^*\mathbf{N})^\vee$ has symplectic singularities.
Read the full abstract

High Energy Physics (Phenomenology)

Joint Bayesian Inference for Extraction of Charmonium Long Distance Matrix Elements Using PYTHIA

We present a joint Bayesian extraction of the nonrelativistic QCD long-distance matrix elements (LDMEs) for $J/\psi$, $\psi(2S)$, and $\chi_c$ production, using short-distance coefficients and feed-down structure implemented in PYTHIA 8.317. Rather than fitting each charmonium state separately, we construct channel-resolved responses that link parent production to daughter feed-down and infer all LDMEs simultaneously from 286 prompt charmonium measurements in $pp$ collisions at $\sqrt{s}=5.02$-13 TeV at the LHC. The extracted LDMEs provide a good description of the data, with an overall $\chi^2/\mathrm{ndf}=1.18$. We also provide a positivity-constrained set that can be used directly as PYTHIA input. Using the extracted LDMEs, we predict prompt $J/\psi$ and $\psi(2S)$ cross sections at high transverse momentum in $pp$ collisions at $\sqrt{s}=200$ GeV, which can be tested with RHIC data.
Read the full abstract

Robotics

OopsieRecover: Learning Transferable VLA Corrections from Semantic Interventions

Vision-Language-Action (VLA) models can perform a wide range of robotic manipulation tasks, but often fail during deployment. Teaching them to recover from these failures requires substantial corrective data: existing approaches learn to recover from failures by directly associating corrective actions with the observations in which they were demonstrated. For instance, DAgger-like fine-tuning can require tens to hundreds of interactions, yet the resulting policy may still fail to generalize the learned corrections beyond the demonstrated scenarios. We hypothesize that semantics can capture the underlying reasons for failures and their corrections, enabling policies to learn from only a handful of corrective demonstrations and generalize those corrections to novel objects, and scene variations. We propose OopsieRecover, a framework that uses abduction to infer the intent behind a small set of corrective demonstrations and encodes the inferred knowledge as dense spatial representations and conditional costs. These costs capture when a behavior is undesirable and steer the policy toward more desirable action regions. We experiment across 6 tasks and variations in object poses, scenes, and related tasks, and find that only a handful of corrective demonstrations can enable a VLA to overcome failures across a broad range of novel situations, without requiring new corrective data for each variation. Please refer to https://robin-lab.cs.utexas.edu/oopsierecover for videos.
Read the full abstract

All authors Arpit Bahety, Skand Peri, Sriniket Ambatipudi, Chen Xu, Haruki Nishimura, Masha Itkina, Roberto Martín-Martín

Quantum Physics

Quantum Knowledge Distillation: State-of-the-art, Applications and Open Problems

Knowledge distillation transfers predictions, representations, or other supervisory information between learning models and is increasingly investigated in quantum machine learning. Its applications extend beyond compression to cross-architecture transfer, circuit approximation, distributed learning, and classical deployment of models trained using quantum-derived supervision. This survey reviews these approaches using two complementary classification axes: teacher-student configuration and the role of quantum resources in the distillation process. Quantum-inspired methods are discussed separately from methods that simulate quantum computations classically or execute them on quantum hardware. We explain how measurement probabilities, expectation values, classical readouts, state overlaps, and circuit-level objectives support different transfer mechanisms. We then critically assess the presented results against baseline design, execution conditions, uncertainty, and resource costs. The evidence shows that distillation can improve selected quantum students, but benefits are task- and architecture-dependent, and several studies find negligible gains or even performance degradation. Evaluations on physical quantum hardware remain limited, with some testing individual quantum components rather than complete learning pipelines. Heterogeneous protocols and incomplete resource reporting prevent general conclusions about practical quantum advantage. We identify priorities for controlled benchmarking, noise-aware learning, reproducible evaluation, and efficient transfer between quantum and classical models.
Read the full abstract

All authors Aldana Al-Sulaiti, Kuan-Cheng Chen, Muhammad Bilal Akram Dastagir, Ahmed Farouk, Saif Al-Kuwari

Quantum Physics

Room-Temperature Quantum Emission from Boron Nitride Nanotubes

Quantum information systems such as quantum key distribution and quantum communications require bright, high-purity, stable sources of single photons to act as carriers of quantum information. As such, methods for obtaining single photon emitters (SPEs) in solid-state devices capable of articulating with quantum photonic circuits are a necessary obstacle that must be overcome. Hexagonal boron nitride (hBN) flakes have been extensively studied as a host material for SPEs, though working in two dimensions is accompanied by engineering complexity. However, dimensionality can be reduced through the implementation of boron nitride nanotubes (BNNTs). Here we investigate the potential of BNNTs as room-temperature SPEs using confocal photoluminescence microscopy and Hanbury Brown-Twiss interferometry. Quantum emitters were found in BNNTs spanning the entirety of the visible spectrum, including spectral regions where SPEs within hBN have not been reported. Furthermore, each tube was found to possess multiple SPEs at varying emission wavelengths along the length, exhibiting lifetimes as fast as 0.5 ns, which is about two-fold faster than the fastest hBN SPEs. Additionally, we demonstrate that surface passivation/coating mitigates blinking events, therefore increasing the emitter stability. Thus, BNNTs can function as a single structure supporting a variety of multiwavelength SPE sources, while maintaining a compact radius of ~25 nm, making them ideally suited for deterministic placement and integration within nanophotonic devices such as waveguides and interferometric filters.
Read the full abstract

All authors Ke-Sean J.M. Peter, Ryan A. Kowalski, David A. Curie, Dakota Braaten, Matthew W. Day, Denis R. Candido, Deyu Li, Richard F. Haglund, Joshua D. Caldwell, Richarda Niemann

High Energy Physics (Phenomenology)

Testing the Spin of New Physics with Top Precision Measurements

Precision measurements can reveal not only the presence of new physics, but also information about the dynamics that generated it. We study this question in top-pair production by comparing three top-philic scenarios: a colored scalar top partner, a colored vector-like fermion, and a spin-one $Z^\prime$ mediator. The first two contribute through loop corrections, while the $Z^\prime$ contributes at tree level. We show that the scalar and vector-like fermion scenarios match onto the same effective operator structure at low energies,making their distinction unfeasible using the EFT approach. However, near or above threshold their full loop amplitudes develop distinct momentum-dependent interference patterns. Using the top transverse momentum spectrum we find that resonant and loop-induced effects are relatively easy to distinguish, whereas separating the scalar and vector-like hypotheses is substantially more challenging. Nevertheless, their different analytic structures leave measurable information in the shape of the distribution. These results illustrate how precision differential measurements can probe the spin and dynamics of virtual new states even when they are not directly reconstructed, and motivate more powerful multidimensional analyses.
Read the full abstract

Robotics

Latent RL Priors for Data-Efficient Dexterous Manipulation

Learning dexterous manipulation from limited visual demonstrations is challenging because policies must acquire both perceptual grounding and precise control, while reinforcement learning (RL) in simulation can acquire rich control behaviors from large-scale interaction with privileged information. We introduce Latent RL Prior, a two-stage framework that reuses this control knowledge for data-efficient visual policy learning. During privileged pretraining, we distill large-scale RL rollouts into a flow-matching action decoder conditioned on proprioception and latent task context. During visual adaptation, we freeze the decoder and learn a visual interface that maps images to its latent context using action supervision alone, without privileged labels or latent targets. The same decoder can further be shared across arms by learning coordinated visual contexts for bimanual manipulation. Across three simulated tasks, our method outperforms full-parameter fine-tuning of $\pi_{0.5}$ and GR00T N1.7 in the low-data regime; on door lifting and bimanual assembly, it surpasses full-data baselines using $10\times$ and $5\times$ fewer demonstrations, respectively. Further analysis shows that direct visual-to-latent conditioning improves adaptation performance and training stability over routing visual predictions through the pretrained state encoder. Robustness to unseen mass variation and greater linear accessibility of contact-force information in decoder representations further provide evidence of retained RL-acquired control knowledge. Project website: https://yixiaowang7.github.io/latent-rl-prior-page/.
Read the full abstract

All authors Yixiao Wang, Peiyi Chen, Mustafa Mukadam, Shreekant Gayaka, Masayoshi Tomizuka, Son Tran

Machine Learning

Adversarial Training with Large Step Sizes: Implicit Bias, Optimization, and Sharpness

Adversarial Training (AT) is an important method for improving model robustness. We study the dynamics of deterministic Gradient Descent (GD) in AT. Most existing works in this direction focus on the small or vanishing step sizes regime, where GD resembles a continuous-time flow. However, such step sizes can make training inefficient and are rarely used in practice. In this paper, we study the AT dynamics of GD with large constant step sizes. We first study GD in adversarial linear logistic regression, focusing on two aspects: where the model parameters converge (implicit bias) and how quickly the losses decrease (optimization). We then turn to neural networks training and study how the sharpness of losses evolve during AT. Our results advance the understanding of the behaviors of GD with large step sizes in AT and highlight three key differences from its behavior in Standard Training (ST): (i) the implicit bias gap between AT and ST, previously established only for small step sizes, persists for arbitrarily large constant step sizes, (ii) different from ST, there is no uniform rate of convergence for the optimization error in AT, and (iii) sharpness can exhibit two-stage dynamics that differ from the Edge-of-Stability behavior observed in ST.
Read the full abstract

Software Engineering

Cognitive Tours: Making Understanding a Deliverable in AI-Assisted Development

Generative AI and agentic coding are accelerating software development, leading to concerns about technical debt. But two less visible forms of debt, cognitive debt and intent debt, are emerging as key risks to software health. Cognitive debt is the absence or erosion of a team's shared understanding where no one can confidently explain how a system works or predict the impact of a change. Intent debt is the lack of clearly articulated goals, constraints, and rationale that explain what the system is for and guide how it should evolve. This paper presents Cognitive Tours, a tool designed to mitigate and repair cognitive and intent debt. Cognitive Tours is a web-based prototype for creating and following curated paths connecting user needs, design decisions, code, tests, and open questions. AI-assisted skills help audit existing waypoints and tours for possible problems or staleness and suggest candidates for new ones; humans decide what matters, verify relationships, and construct the narrative. This division of work deliberately preserves productive friction: the sensemaking involved in selecting, explaining, and revising a tour, which is intended to foster and share understanding. Our demonstration uses a tour of Cognitive Tours itself to introduce the tool and illustrate its design decisions and interaction mechanisms. A proposed evaluation agenda concludes the paper. A demo is available at https://cognitive-tours-demo.vercel.app/
Read the full abstract

Combinatorics

Tusnády's problem and online two-sample thinning

We determine the sharp order of discrepancy in Tusnády's problem in every fixed dimension $d\ge 2$. The largest combinatorial discrepancy of $N$ points in $\mathbb{R}^d$ with respect to axis-parallel boxes is $\Theta_d((\log N)^{d-1/2})$, extending Wei's planar result and matching Nikolov's upper bound. More strongly, with high probability, this lower bound holds simultaneously for every subset obtained by deleting at most a sufficiently small fixed fraction of a uniform $N$-point sample. We apply this geometric estimate to online thinning of two independent samples of size $n$ from an unknown common distribution, where observations must be kept or discarded upon arrival. For every fixed dimension $k\ge 1$ and expected deletion budget $\rho n$, with $0<\rho<2$, we prove that the minimax expected prefix discrepancy over lower orthants is $\Theta_{k,\rho}((\log n)^{k+1/2})$. In dimension one, the optimal terminal discrepancy is instead $\Theta_\rho((\log\log n)^{3/2})$, giving a sharp separation between the prefix and terminal objectives. For $k\ge 2$, we obtain terminal bounds differing by a factor of $\log n$. The lower bounds are based on Gaussian small-ball estimates for box-incidence matrices, while the upper bounds use stationary Metropolis thinning and, in one dimension, a terminal cancellation procedure.
Read the full abstract

Differential Geometry

On do Carmo's Question: The Properly Embedded Case

In this paper, we prove that every complete noncompact two-sided properly embedded constant mean curvature hypersurface of finite Morse index in Euclidean space is minimal. The result holds in every dimension. We assume no bounds on curvature or volume growth. This establishes the properly embedded case of the finite-index version of do Carmo's question. For nonzero mean curvature, we use the radii of inscribed balls to define a positive weight. This weight gives a polynomial bound on weighted volume, while its reciprocal satisfies a differential inequality involving the Jacobi operator. We combine these properties with stability outside a compact set to obtain finite total area. This leads to a contradiction.
Read the full abstract

Machine Learning

Unpacking the Compute Efficiency of Strong-to-Weak OPD against GRPO

On-policy distillation (OPD) trains a student on its own rollouts under a frozen teacher's per-token supervision. Does this dense signal reduce the training FLOPs needed to reach a target accuracy compared with reinforcement learning? We systematically compare vanilla reverse-KL OPD with Group Relative Policy Optimization (GRPO) across task difficulty, student and teacher size, domain, and optimization choices, measuring held-out accuracy against training FLOPs. With an external post-trained teacher, OPD generally loses to GRPO in accuracy per training FLOP at matched settings, even when the teacher is further trained with GRPO, and often suffers response-length inflation. Yet a different teacher changes the picture: when the teacher is the student's own GRPO checkpoint, OPD recovers that checkpoint's accuracy on math with $50\times$ fewer distillation FLOPs than GRPO needs to reach the same accuracy. Following the same teacher setting, we consistently observe this high efficiency with a larger student, on code, and in a second model family. On a single domain, however, the GRPO checkpoint is already the desired model, so distilling it adds to total compute. Across domains, separate GRPO checkpoints hold different capabilities, and no single checkpoint serves every domain. Multi-teacher OPD (DeepSeek-AI, 2026) consolidates per-domain GRPO experts into a single student that matches each expert's accuracy, with the distillation stage costing only 5.2% (on 2 domains) and 8.0% (on 4 domains) of the specialist training FLOPs. Counting specialist training and distillation together, this pipeline reaches joint GRPO's performance on the 2-domain mix with 42.0% fewer total FLOPs.
Read the full abstract

All authors Zeyu Leo Liu, Yang Li, Bo Pang, Emre Can Acikgoz, Xuan-Phi Nguyen, Anurag Koul, Srijan Bansal, Erik Nijkamp, Shafiq Joty, Semih Yavuz

Artificial Intelligence

ReproBridge: Building Reproducibility into ML Research Artifacts

Reproducing machine learning experiments requires reconstructing procedures from information scattered across papers, released code, and supporting artifacts. For coding agents, reproduction success can therefore depend on their ability to fill in missing details, rather than on the released artifacts alone. We ask whether preserving recovered procedures and decisions improves subsequent independent reproduction. We define a standardized artifact design comprising a refined codebase, environment setup, experiment launchers, and a reproduction guide, and introduce ReproBridge, an agent workflow for constructing these refined reproducible artifacts. ReproBridge-Bench provides 1,029 tasks from 387 experiments across 36 papers, anchored by five human-verified reference scores per task. Agents attempt tasks in fresh sessions with raw code or refined artifacts; a separate Auditor checks execution evidence. Across four evaluated models, refined artifacts raise mean reproduction success from 76.4% to 92.6% and reduce downstream API spending by 19.7% on average. The cross-model success-rate spread narrows from 21.6 to 3.9 percentage points, with all four models achieving mean success rates above 90%. Removing any component reduces success in the ablation study, with the largest losses from code and environment removal. These results suggest that preserving execution knowledge helps even models with lower raw-code success reproduce experiments more reliably and at lower API cost.
Read the full abstract

All authors Yubo Wang, Rui Meng, Jaehyun Nam, Tomas Pfister, Jinsung Yoon

Dynamical Systems

h-dichotomies via uniform noncriticality and expansiveness for noninvertible evolution families

We establish characterizations of h-dichotomies for noninvertible evolution families acting on Banach spaces, in terms of appropriate notions of expansiveness and uniform noncriticality. Our results extend the corresponding characterizations previously obtained for invertible evolution families. To the best of our knowledge, no such characterizations have previously been established for general noninvertible evolution families. A key difference is that noninvertibility makes the structure of the stable spaces more delicate: in our positive-half-line result, finite codimension must be imposed at every time, rather than only at the initial time. We show by an example that the latter, weaker condition is insufficient. We also study evolution families on the negative half-line, using duality to obtain corresponding characterizations in terms of dual expansiveness and dual uniform noncriticality. General growth rates are treated by a time-rescaling argument.
Read the full abstract

Differential Geometry

Improved Index and Intersection bounds for Min-Max Geodesics

In this note, we show that the sum of the Morse index and number of intersections of min-max geodesics achieving the p-widths of a closed surface are bounded above by $p$, improving a result from our previous work. A key idea is that deformations coming from vertex desingularization and eigenfunctions of the Jacobi operator on geodesics are sufficiently independent.
Read the full abstract

Robotics

VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding

Models for physical AI tasks, such as robotics and autonomous driving, need not reason solely in language. They can use visual plans–points, bounding boxes, and trajectories–to represent spatial goals and anticipate motion. However, autoregressive vision-language models generate these plans one coordinate token at a time, making richer predictions increasingly costly. We introduce VFlash, a method for fast, accurate visual planning through Flash Primitive Decoding. VFlash exploits the known structure of visual primitives to predict their coordinates jointly, using a lightweight drafter that generates an entire primitive in one forward pass. The layout adapts to the primitive's type, dimensionality, and length, supporting compact boxes and long, variable-length trajectories. Crucially, spatial accuracy need not require reproducing an autoregressive model's exact token sequence. VFlash therefore trains on annotated primitives and directly accepts structurally valid predictions without target-token agreement. Across the evaluated 2D and 3D robot trajectory tasks, VFlash consistently reduces prediction errors relative to autoregressive decoding. Preliminary single-rollout measurements indicate approximately 25x faster generation for 3D point motion forecasting. A world action model conditioned on VFlash plans achieves 70.2% average success across four MolmoSpaces task suites, compared with 62.6% for the baseline.
Read the full abstract

All authors Yu-Cheng Chou, Shihao Wang, Jin Huang, Chi-Pin Huang, Chiao-An Yang, Xianzhe Fan, Jing Wang, Zhiqi Li, Yuanguo Kuang, Ruijie Zheng, Johan Bjorck, Sifei Liu, Hang Su, Abhishek Badki, Lei Zhang, Alan Yuille, Linxi "Jim” Fan, Jan Kautz, Zhiding Yu

Functional Analysis

Gelfand numbers of Schatten class embeddings in the quasi-Banach range

We prove Conjecture 1.4 of Prochno and Strzelecki [J. Approx. Theory 277 (2022), 105736] on Gelfand numbers of embeddings between finite-dimensional Schatten classes. For $0<p\le1$ and indices bounded away from the ambient dimension by a fixed proportion, we obtain the conjectured lower bound, uniformly in the target exponent $2\le q\le\infty$. Together with the known upper estimate obtained by Hinrichs, Prochno and Vybíral [Math. Ann., 380(3-4):1563-1593, 2021], this gives the exact asymptotic order in the previously unresolved range. As a consequence, we obtain a complete all-index description by combining our result with the estimates of Hinrichs, Prochno and Vybíral and of Prochno and Strzelecki. The proof is deterministic and is based on Hilbert–Schmidt projections on rectangular matrix spaces.
Read the full abstract

Quantum Gases

Dynamical effects induced by synthetic SU(1,1) gauge fields

Compact gauge fields are powerful tools for controlling quantum dynamics and topological phases of matter. Non-compact gauge fields are particularly relevant for open systems, where particles or energy are exchanged with the environment. Here, we investigate the wavepacket and entanglement dynamics induced by an $\mathrm{SU}(1,1)$ gauge field. The wavepacket undergoes non-Abelian non-compact Bloch oscillations, in which the spectrum of an auxiliary one-dimensional quantum problem determines the transition between stable oscillations and unstable amplification. In the unstable adiabatic regime, the wavepacket probes an $\mathrm{SU}(1,1)$ topological defect, and its amplification is determined solely by a hyperbolic solid angle in Minkowski space. Threading an $\mathrm{SU}(1,1)$ gauge field through a bosonic tight-binding model yields the bosonic Kitaev model, whose volume or super-volume entanglement scaling is governed by the stability of the gauge field. We further propose an experimental realization of an $\mathrm{SU}(1,1)$ gauge field in a spinor condensate. Our results can be generalized to other non-compact groups and higher dimensions, providing a geometric framework for entanglement control, selective amplification, and engineering energy transfer.
Read the full abstract

Probability

A Central Limit Theorem For Random Variables Under A Constraint

We consider vectors chosen randomly from high-dimensional hypersurfaces of the form $\{x\in\mathbb{R}^n: \sum_{j=1}^n V_j(x_j)=an\}$. These in general asymmetric surfaces generalize classical spheres and include unbounded and disconnected surfaces. We show that under mild assumptions as $n\to\infty$, any weighted sum of the random coordinates is asymptotically Gaussian, provided that no weight dominates. An explicit reduction in the variance compared to the unconstrained case is observed. We also study a model with penalization of deviations from the constraint, interpolating between the unconstrained and the constrained case. Our approach involves an explicit representation of the characteristic function of the constrained random variables in terms of independent random variables.
Read the full abstract

Machine Learning

A Hidden Markov Preference Model for Online Prediction of Adversary Actions

A defender anticipating an attacker needs a probability distribution over the attacker's next action from information available before that action occurs. Two structures complicate this: the attacker's behavioral regime persists and can switch within an episode, and each action combines an operation with state-dependent arguments whose feasible values change over time. We capture both with a hidden Markov model whose latent state is the active regime and whose regime-conditioned action law is defined on a feasibility-masked tree of executable actions. The same likelihood forms the next-action distribution and updates the regime belief once the action is seen, and a partially observed prefix contributes its resolved local factors. Estimation separates into convex logistic fits that Baum-Welch reuses when regimes are unlabeled. We derive a reach-weighted nodewise KL expression for regime information, a sequence-wise log-loss bound for switching regime paths, and a uniform fallback guarantee. On the CAGE 2 cyber defense benchmark the filtered latent state is worth 0.8 nats per step, and after a controller switch the loss begins to fall in three steps against ten for a recurrent baseline.
Read the full abstract

All authors Angel Daruna, Michael Cogswell, Tixiao Shan, Yi Yao, Han-Pang Chiu, Ajay Divakaran

Quantum Physics

Polynomial-time classical simulation of boson and fermion sampling under thermal loss

Sampling experiments are a leading approach to demonstrating quantum computational advantage, yet their robustness to experimental noise is not completely understood. We develop a common framework for classical simulation of boson and fermion sampling under thermal loss, including pure loss. For Fock-state boson sampling, Gaussian boson sampling with fixed uniform squeezing, and fermion sampling with magic inputs, we obtain polynomial-time sampling at inverse-polynomial total variation error, for any fixed nonzero overall loss rate, and with high probability over Haar-random interferometers. A milder flatness assumption on the ideal output distribution suffices for quasipolynomial-time simulation without Haar randomness. With constant thermal loss per layer, the same flatness assumption gives polynomial-time simulation at logarithmic depth. Our bounds for Fock-state boson sampling cover both dilute and saturated regimes and become stronger near full input filling: when only a bounded number of input modes are empty, the noisy output for Haar-random interferometers converges, as the number of modes grows, to a distribution independent of the interferometer.
Read the full abstract

Human-Computer Interaction

Souffleur: A Glanceable Display for Mobile Voice Conversations

The speech capabilities of mobile devices have grown, but their displays still rely on a text-based chat. When resuming a voice chat after a context switch, the last messages may not convey the key points of the conversation. Autobiographical use and sessions with external participants led to design goals for a mobile-first, speech-driven, glanceable display that helps users continue their reasoning from where they left off. We implement these goals in Souffleur and evaluate it in a between-subjects study with 48 participants. Compared with a conventional chat transcript, we found no reliable difference in how easily participants followed and resumed short voice conversations, nor in effort, attention or recall, although Souffleur reduces the conversation to a few short phrases.
Read the full abstract

Software Engineering

SoK: Assessing Post-Quantum Cryptography Adoption in Software Development: Standards, Lifecycle Coverage, Deployment Maturity, and Evidence of Readiness

Standardised post-quantum cryptography (PQC) is critical for protecting software systems against future quantum threats, but standardised algorithms alone do not provide a pathway for their secure integration into software. This paper presents a Systematisation of Knowledge (SoK) of PQC integration in software development, synthesising existing approaches across implementation, migration, and developer education. We analyse these approaches against NIST PQC standards and PQC migration guidance, software development lifecycle coverage, cybersecurity practices, deployment maturity, and testing and evaluation evidence. Our analysis shows that existing work establishes important technical and organisational foundations but provides disconnected support across the software development lifecycle, with limited coordination between preparation, implementation, migration, deployment, maintenance, and developer practices. Evaluation evidence also remains concentrated in controlled or early-stage settings, providing stronger evidence of technical feasibility than operational readiness. We identify cross-cutting gaps and research directions for developing lifecycle-oriented, empirically validated approaches that connect technical, developer, and organisational requirements for secure PQC integration.
Read the full abstract