Peer Revenue Returns subject to peer review

Papers

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

All of arXiv

Saturday 19 September

Rings and Algebras

Associativity of the $\star$-product of skew braces

Andrea Sciandra

For braces, associativity of the $\star$-product is equivalent to two-sidedness. We extend this result to skew braces by proving that any two of the following conditions imply the third: two-sidedness, associativity of the $\star$-product, and centrality of $A\star A$ in the additive group. We then study two-sided skew braces with associative $\star$-product through the radical ring associated with their additive abelianization. This yields a canonical fibre-product description modulo an annihilator ideal, exact sequences relating higher $\star$-products to the powers of the radical ring, coincidence of the left, right and strong $\star$-series, and sharp bounds on their nilpotency indices. We also obtain an explicit description of the lower central series, classify the structures whose additive abelianization is infinite cyclic, and characterize when they are bi-skew.
Read the full abstract

Signal Processing

Full-Frame OFDM-ISAC: Joint Constellation, Pilot, and Receive-Filter Design

Dongil Yang, Kaitao Meng, Fan Liu et al. (+2 more)

Orthogonal frequency-division multiplexing (OFDM) integrated sensing and communication (ISAC) reuses the entire time-frequency frame, including pilots and data payloads, for radar sensing. The random data payload, however, reduces the sensing dynamic range and degrades the target detection and parameter estimation performance. Existing studies typically optimize the modulation constellation, the pilot pattern, the power allocation, and the receive filter in isolation, overlooking their coupled effects on the sensing ambiguity structure. To address this gap, we propose a joint design of the constellation, the pilot placement and power split, and the mismatched receive filter for full-frame OFDM-ISAC systems, which minimizes the residual interference inside declared regions of interest (ROIs) under a communication rate constraint. For a given transmit signaling scheme, the receive filter admits a closed-form solution. With this filter in place, we derive closed-form expressions for the receiver-specific interference floor and noise enhancement over the 2-D range-Doppler domain, which show that the floor is governed by a single effective statistic of the full-frame signaling. This law reduces the joint design to two steps. A balanced staggered pilot placement removes the deterministic grating lobes of pilot-power boosting, and the shaped constellation and the power split are then selected jointly on the boundary of the rate constraint under the channel estimation error, which attains the joint optimum over the constructed shaping levels to first order in the declared ROI size. Under a quality-of-service (QoS) constraint of $98\%$ of the rate of a conventional 16-QAM frame, the proposed design significantly lowers the interference floor, improves the target detection probability, and achieves better ranging accuracy compared to the conventional receive filter-aware design.
Read the full abstract

All authors Dongil Yang, Kaitao Meng, Fan Liu, Christos Masouros, Kawon Han

Human-Computer Interaction

Adaptive Scaffolding Needs Contingency: An AI Tutor That Escalates and Fades on What the Learner Does

Xinmeng Hou, Yuxuan Weng, Chin Hsien Yeh et al. (+5 more)

Coding assistants raise task performance, but learners plan and monitor less. Giving less away, the usual fix, conflates two things: how much work a system carries (cognitive load) and what the learner must decide before help arrives (metacognitive demand). Our principle, preserved metacognitive demand, holds the second constant and lets the first vary. CoMeT implements it: support rises when a learner fails at a decision point and fades on take-up. Within subjects, 131 adult learners used CoMeT, an unrestricted assistant and a question-only tutor on three Python tasks. CoMeT matched the question-only tutor's demand, delivered artifacts twice as often as the assistant, and frustrated learners less than the question-only tutor, with delegation and load unchanged. Learners often did not answer. Fading held when their turn addressed the decision under support, and CoMeT surrendered the full answer in one session in sixteen, against one in six for the question-only tutor.
Read the full abstract

All authors Xinmeng Hou, Yuxuan Weng, Chin Hsien Yeh, Ding Lin Lee, Lishan Zheng, Fang Li, Wuqi Wang, Yang Liu

Materials Science

From Local Atomic Motifs to Thermodynamic State: An Interpretable Physics-Informed Framework for Cu-Zr Metallic Glasses

Prashil S. Joshi

Machine-learning models that relate local atomic structure to the thermodynamic state of metallic glasses typically assess physical consistency after training rather than enforcing it during learning. Here, we develop a multi-task physics-informed neural network (PINN) that predicts temperature directly from Voronoi-motif population histograms while incorporating autograd-derived gradient constraints representing physically motivated relationships between structural motifs, quench rate, and temperature. The model simultaneously classifies each configuration as liquid, supercooled/transition, or glass through an auxiliary classification head. The classification task achieves 96.6% test accuracy with a macro-F1 score of 0.95, while the regression head yields a mean absolute error of 23.4 K on a trajectory-disjoint held-out test set. Five-fold trajectory-grouped cross-validation, deep-ensemble predictions, and Monte Carlo dropout are used to assess model robustness and predictive uncertainty. Sensitivity analyses demonstrate that physics constraints can be incorporated over a broad range of loss weights without compromising predictive accuracy, while substantially improving compliance with the prescribed physical trends. Benchmarking against conventional machine-learning regressors further demonstrates competitive predictive performance. SHAP analysis across the cross-validation ensemble identifies the coupled near-icosahedral motif family, particularly the full icosahedral motif and its single-atom-perturbed counterpart, as the dominant structural fingerprints of the glassy state. These results demonstrate that physically motivated constraints can be embedded directly into motif-based structure-property models, providing a pathway from post-hoc interpretability toward physically constrained machine learning for metallic glasses.
Read the full abstract

Distributed Computing

Silent Failures Beyond the 32-Bit Index Range: A Differential Characterization of Large-Tensor Matrix Multiplication in PyTorch's MPS Backend

Junichiro Niimi

Apple Silicon machines with large unified memory make it possible to hold large tensors on a desktop GPU. However, we found that PyTorch's Metal Performance Shaders (MPS) backend silently returns wrong results for batched matrix multiplication with more than $2^{32}$ elements. torch bmm, including its wrappers matmul and eager attention, returns relative errors above 1 without an exception or a warning in every PyTorch release tested (2.4.1 to 2.14.0). We sweep bmm over dtypes, memory layouts, shapes and batch sizes around $2^{31}$ and $2^{32}$ elements, and judge every result against a float64 computation on the CPU. Three rules account for every outcome on 2.14.0. When the output exceeds $2^{32}$ elements and an operand is a transposed view, the entire output is wrong and equals a computation that ignores that operand's strides. Otherwise, a view with at least $2^{31}$ elements raises an exception, and a contiguous input above $2^{32}$ elements makes exactly the batches beyond that point wrong, equal to a computation whose index wraps at $2^{32}$. A slightly larger problem can thus turn an explicit error into a silent failure. The rules extend to the backward pass, where a correct forward pass can return silently wrong gradients. A second machine with another chip, under two macOS versions, reproduces all 6156 results, including the wrong values, and the same sweeps on an NVIDIA A100 are correct in all 2530 runs. In a public sentiment classifier, one oversized batch corrupts a third of the outputs, which collapse onto one class. All findings come from observable behavior, without access to the backend's closed-source kernels; we release the harness, raw results and a guard that stops any MPS operation touching $2^{32}$ or more elements at jniimi/mps-silent-failures (https://github.com/jniimi/mps-silent-failures).
Read the full abstract

Machine Learning

On attention heads and bilinear forms

Andrew O'Desky

We study the symmetric and antisymmetric parts of bilinear forms in the attention heads of trained large language models. We introduce an orthogonally invariant profile map from real bilinear forms to a three-dimensional simplex and observe that profiles of trained bilinear forms accumulate near profiles of rank-one bilinear forms. We prove that the symmetric part of a bilinear form in an attention head is the sum of a hyperbolic form and a zero form for a Zariski-dense subset of query-key matrices.
Read the full abstract

Representation Theory

From $\tau$-tilting to $\tau_2$-tilting via splitting tilting triples

Zhongyuan Wang, Haicheng Zhang and Tiwei Zhao

Let $(A,T,B)$ be a splitting tilting triple. We provide a construction of $\tau_2$-tilting $B$-modules from $\tau$-tilting $A$-modules. On the other hand, if $A$ is hereditary, by considering the mutation of a certain two-term silting complex of $A$ and its image under the derived equivalence $\mathbf{R}{\rm Hom}_A(T,-):D^b(A)\to D^b(B)$, we get a three-term silting complex ${\bf W}$ of $B$. Under an explicit non-vanishing condition on the cones of morphisms between the two summands, we prove that ${\rm H}^0\boldsymbol{W}$ is a basic $2$-tilting $B$-module and that it coincides with the module obtained from the first construction. As an application, we generalize the construction of Peng from tilting modules to $\tau$-tilting modules. We also determine the endomorphism algebra of the constructed module as an explicit triangular matrix algebra, which is derived equivalent to $A$ and $B$ whenever the constructed module is $2$-tilting, and give a full semi-orthogonal decomposition of $K^b(\operatorname{proj}A)$.
Read the full abstract

Computation and Language

Rethinking Pivot Programming Languages in Code Language Models

Andor Diera, Lukas Galke Poech and Matthias Tichy

Multilingual code language models transfer skills across programming languages (PLs), but whether any PL occupies a privileged pivot position remains contested: geometric analyses point to C-family languages and Go, while behavioral evidence highlights Python. We revisit this question under controls for representational anisotropy and length variation across PLs, two confounds that compromise prior cosine-based analyses. Across three code models on multilingual competitive-programming data, we study three views of cross-PL organization: pairwise PL geometry, PL-English alignment, and pivoted retrieval through candidate PL representation spaces. The results are relation-dependent. Code-code geometry reveals structured language regions but no universal center; code-English alignment favors high-level scripting languages; and pivoted retrieval favors different intermediate spaces for code-to-code and English-to-code transfer. These findings suggest that Python's special role is better understood as English-facing affinity than as universal geometric centrality.
Read the full abstract

Artificial Intelligence

OptiSkill: A Hierarchical and Evolving SkillBank for LLM-Based Optimization Modeling

Ruiqing Zhao, Rui Liu, Yuan Zuo et al. (+3 more)

Automated operations research (OR) modeling requires LLMs to translate natural-language decision problems into correct mathematical programs. Existing methods can improve individual formulations, but they often solve problems in isolation, retaining little reusable experience and repeating similar formulation errors. Prior memory-based approaches store examples, thoughts, or insights as references, while OR modeling requires reusable formulation skills that transfer across problem narratives and guide concrete modeling decisions. We propose OptiSkill, a skill-augmented framework that builds a hierarchical and evolving SkillBank for LLM-based OR modeling. SkillBank stores solver-verified experience as reusable skills, with Global Strategies for problem-level formulation skeletons and Step Experiences for local error-prevention rules. It is further refined through stable batch-level test-time evolution, where candidate skills are incorporated only after validation. Experiments on eight OR modeling benchmarks show that OptiSkill improves formulation accuracy across LLM backbones, outperforms strong agentic baselines, and gains further by expanding SkillBank coverage and reliability. Code and data are available at https://github.com/rachhhhing/OptiSkill
Read the full abstract

All authors Ruiqing Zhao, Rui Liu, Yuan Zuo, Huarong Zhang, Xiao Han, Junjie Wu

Systems and Control

On a Closed-Loop Controller for the Coevolutionary Model of Actions and Opinions via Broadcasting Information

Roberta Raineri, Mengbin Ye and Lorenzo Zino

We deal with controlling a complex social network in which agents have actions and opinions that coevolve, mutually influencing one another. We consider an input consisting in broadcasting information to a target set of agents with the objective of steering the population, initially at a consensus, to a different consensus. For a constant input, we derive a monotone convergence result, building on which we design an algorithm that determines whether a target set is sufficient to achieve the objective and an effective heuristic to optimize the target set. Then, we introduce a feedback control law that, using information on the state of the system, dynamically revises the target set, reducing the effort needed to achieve the objective while guaranteeing convergence to the desired consensus state.
Read the full abstract

Statistical Methodology

Lorentz Hyperbolic Weighted Regression: Theory for Fixed and Estimated Representations

Bahad{\i}r Yüzbaş{\i} and Zühal Küçükarslan Yüzbaş{\i}

Many applications provide each observation with a meaningful representation in addition to an ordinary response and covariates. When that representation is hierarchical, hub-periphery structured, or network derived, Euclidean or geographic proximity may define the wrong peer groups. We present Lorentz hyperbolic weighted regression (LHWR) as a practical local regression method for this setting. Responses and predictors remain real valued; only locality is defined by distances between observations represented on the Lorentz model of hyperbolic space. We describe coordinate construction, adaptive bandwidth selection, prediction, local coefficient summaries, collinearity checks, and residual autocorrelation diagnostics. Theoretical results explain consistency, bias-variance tradeoffs, curvature effects, and the extra uncertainty caused by estimated representations. Simulations show that the Lorentz geometry is most useful for sharply localized coefficient surfaces, whereas tangent-plane approximations can be competitive for smooth surfaces. In a 141-country World Development Indicators illustration, economic similarity based on income and trade openness defines local peer groups. Representation-based locality predicts GDP growth better than global least squares and geographically weighted regression, although differences among Lorentz, Poincare, and tangent metrics are modest. The main practical lesson is that the representation should be chosen scientifically and the distance geometry should be checked rather than assumed.
Read the full abstract

Machine Learning

LPINNs: First-Layer Gated Localization for Physics-Informed Neural Networks

Lakshay Chawla and Hardik Jain

Physics-informed neural networks (PINNs) use one shared representation over the computational domain, which can become difficult to optimize on long domains and for high-order operators. We study a minimal alternative: multiply the first hidden activation of an otherwise unchanged dense PINN by input-dependent localization functions, giving first-layer units receptive fields without partitioning the domain or adding interface losses. We screen 13 families of localization functions, in up to three parameterizations each, on a nonlinear harmonic oscillator (HO), a heat equation on a long spatial interval, and a manufactured four-dimensional (4D) fourth-order problem, with ten paired seeds throughout. Three configurations give large reductions in solution error at matched budgets: (i) Fixed Gaussian localization functions on the $2\pi$ HO domain cut mean solution RMSE from $4.8369\times10^{-1}$ to $8.83\times10^{-3}$ at 3k epochs. (ii) The inverse-quadratic family with learnable centers and widths cuts it from $2.896\times10^{-1}$ to $3.06\times10^{-2}$ on the $8\pi$ heat domain at 10k epochs. (iii) Fixed bump localization functions cut it from $1.75947\times10^{1}$ to $2.260\times10^{-1}$ on the $4\pi$ 4D domain at 10k epochs. Every paired seed improves in these three comparisons. The screen also shows that the mechanism is not a free win: on HO only 2 of 13 families beat the baseline, and 10 of the remaining 11 are 9 to 23 times worse; on 4D four families are non-finite and five are more than three orders of magnitude worse than the baseline. The inverse-quadratic family is the only one that beats the baseline on all three equations. Overall, these results show that first-layer localization can provide measurable improvements to baseline PINNs on long-domain and high-order problems.
Read the full abstract

Robotics

Connectivity-Aware Exploration of Robotic Grasp Spaces

Maksim A Kazanskii

Robotic grasping is typically formulated as the problem of identifying successful actions from a space of candidate grasp poses. However, the organization of successful actions within this space has received less attention. We study the multiscale structure of viable robotic grasps in $SE(3)$ and investigate whether this structure can be exploited for more efficient exploration. Using a large-scale grasp dataset, we show that successful grasp sets exhibit heterogeneous and reproducible connectivity structure across objects. We then introduce a connectivity-aware sampling strategy that incrementally explores the currently observed grasp space by prioritizing potential bridges between components, structural frontiers, boundary extensions, and geometric novelty. In controlled reconstruction experiments, the method recovers the connectivity structure of successful grasp sets substantially more efficiently than random sampling and farthest-point sampling. We further evaluate whether connectivity acquired under hidden grasp viability can improve subsequent grasp discovery, and whether structural experience from previously explored objects can be retrieved and transferred to unseen objects. These results suggest that the spatial organization of viable actions provides information relevant to grasp-space exploration beyond the viability of individual candidate actions. More broadly, they motivate structure-aware exploration as a means of exploiting the geometry of viable action spaces in robotic manipulation.
Read the full abstract

Earth and Planetary Astrophysics

Optical Constants of Photochemical Haze Analogs in N2-CH4-CO Atmospheres from 0.4 to 28.6 {\mu}m

Zhengbo Yang, Chao He, Haixin Li et al. (+4 more)

Photochemical hazes play an important role in shaping the spectra and radiative balance of N2-dominated planetary atmospheres. We present newly acquired FTIR and retrieved optical constants (N=n+ik) of laboratory-generated haze analogs from N2/CH4 and N2/CH4/CO gas mixtures under plasma discharge conditions. The retrievals use particle densities newly measured for the CH4-series and previously published particle densities for the CO-series. The experiments systematically explored CH4 concentrations from 0.5% to 10% and CO concentrations from 0% to 5% with fixed 5% CH4. Using measured particle densities together with the Beer-Lambert law and subtractive Kramers-Kronig (SKK) relation, we derived optical constants over the 350-25000 cm-1 (0.4-28.6 {\mu}m) spectral range, with the 0.4-25 {\mu}m results presented in the main text. The infrared spectra reveal prominent absorption features associated with hydrocarbon-, nitrogen-, and oxygen-bearing functional groups. Increasing CH4 abundance enhances aliphatic hydrocarbon features and corresponds to decreasing particle density, whereas increasing CO abundance promotes oxygen incorporation, broader mid-infrared absorptions, and higher particle density. The derived k spectra exhibit strong absorptions near ~3 {\mu}m, ~4.6 {\mu}m, and ~6-10 {\mu}m, while the real refractive index n generally ranges from ~1.2 to 1.7. The controlled CH4- and CO-series establish composition-dependent variations in haze optical properties. A benchmark comparison among Titan-, Pluto-, and Triton-like haze analogs then uses these experimentally identified trends to interpret the optical differences among N2-dominated planetary haze compositions. The density and optical constants provide laboratory constraints for atmospheric radiative transfer models and for interpreting planetary and exoplanetary spectra from spacecraft and telescopes.
Read the full abstract

All authors Zhengbo Yang, Chao He, Haixin Li, Sai Wang, Xiao'ou Luo, Yu Liu, Sarah M. Hörst

Cryptography and Security

Dual-Locking Learned AI Models: A PIN-Based Sparse QIM Watermarking and Adaptive Index Permutation Approach

Iva Vasic, Jesús Muñoz-Cádiz and Bata Vasic

We present a dual-locking method for securing trained neural networks that combines key-driven index permutation with PIN-based watermarking based on Sparse Quantization Index Modulation (QIM). Cryptographic randomness is introduced by independently applying a uniform random permutation to each row of adaptively selected index vectors. A robust blind binary watermark is then embedded into the bias coefficients by modulating their quantized values, binding the network to a user-defined Personal Identification Number (PIN). Without the correct key, the network retains its architecture but becomes functionally impaired due to disrupted internal representations. Inverse permutation fully restores the original model accuracy, while the embedded watermark remains imperceptible and enables blind verification of key association and model authorship. To improve both locking effectiveness and recoverability, an adaptive key selection strategy redistributes high-magnitude weights to low-sensitivity positions and vice versa, increasing degradation in the locked state while preserving full recovery. Experiments on MNIST, CIFAR-10/100, and ImageNet-1K using fully connected networks, ResNet CNNs, and transformer architectures show that locking reduces accuracy below 10%, and even below 0.5% for CNNs, while the correct key fully restores performance. The watermark introduces no measurable accuracy degradation and reliably authenticates ownership. Analysis of embedding distributions across CNNs and transformers further indicates potential diagnostic value for identifying undertrained or suboptimally designed models. The proposed approach therefore provides simultaneous model protection, recovery, and ownership verification.
Read the full abstract

Analysis of PDEs

The Binary-ternary Boltzmann Equation Near Equilibrium

Xianshen Hu, Linjie Xiong and Zhihong Xu

The classical Boltzmann equation, formulated by Boltzmann in 1872, describes the statistical behavior of a dilute gas in a non-equilibrium state. While the global well-posedness of the classical Boltzmann equation has been extensively studied and is relatively well-understood, this model becomes inadequate when the gas is no longer sufficiently dilute. Therefore, it is necessary to investigate higher-order extensions of the Boltzmann equation that account for multi-particle interactions. Recently, Ampatzoglou and Pavlović provided the rigorous derivation of both the ternary [Comm. Math. Phys. 387 (2021), no. 2, 793-863.] and the binary-ternary [Forum Math. Sigma 13 (2025), Paper No. e52, 95 pp.] Boltzmann equations from a classical particle system, yielding an explicit form of the ternary collision operator. However, the global existence near equilibrium for the binary-ternary Boltzmann equation remains an interesting and fundamental question, a gap that this paper seeks to address as its primary contribution.
Read the full abstract

High Energy Physics (Experiment)

Mixture Density Networks for Neutrino Reconstruction at Hadron Colliders

Seungjin Yang, Jason S. H. Lee and Junghwan Goh

Neutrino momentum reconstruction at hadron colliders is intrinsically ambiguous because the longitudinal momentum is not directly observed. We study this problem in semileptonic $t\bar{t}$ events using \monster{} (Mixture of Neutrino Solutions with Transformer Event Representation), a mixture density network that predicts a multivariate normal mixture for the conditional distribution of neutrino momentum from reconstructed event objects. The model uses a Transformer-based event encoder and yields a sampling-free point estimate of the neutrino momentum from the closed-form density. On the public benchmark introduced with \nuflows, \monster{} reduces the 68th percentile of the three-momentum residual by 5% and the 95th percentile by 6% relative to the empirical-mode \nuflows{} baseline, with per-variable root-mean-square errors smaller by 2 to 9% at comparable bias, while running 2 to 8.5 times faster at inference, depending on the device and batch size. These results indicate that mixture density networks are a competitive alternative to normalizing flows for neutrino reconstruction.
Read the full abstract

Distributed Computing

DeepSeek Elastic Compute (DSec): A Sandbox Infrastructure for Effective Agentic Training at Scale

Jialiang Huang, Hongxuan Tang, Jingchang Chen et al. (+128 more)

Large-scale agentic training and evaluation with large language models (LLMs) rely on isolated, stateful execution environments in which models inspect repositories, invoke tools, execute commands, and interact with task-specific services. These workloads create sandboxes in large bursts, span heterogeneous functionality and isolation requirements, retain state across long interactions, and draw from large image corpora with limited reuse. Supporting them therefore requires an elastic execution platform rather than a single sandbox runtime. This report presents DeepSeek Elastic Compute (DSec), a production sandbox platform that exposes FnCall, container, microVM, and full-VM sandbox backends through a unified SDK. DSec coordinates placement and lifecycle management across the cluster, composes environments from independently versioned layers, combines memory sharing, reclamation, and CPU scheduling for high-density execution, and loads image data on demand from Fire-Flyer File System (3FS), a cluster-wide distributed filesystem. DSec is co-designed with the reinforcement learning (RL) framework, decouples stateful rollout execution from preemptible GPU training, coordinates sandbox lifecycle with training to preserve rollout state while reclaiming idle resources, and mitigates agent misbehavior such as reward hacking. A single production-scale unit of DSec spans around 160 nodes, serving about 3 million sandboxes per day; in production, it supports over 380,000 concurrent sandboxes and sustains over 5,000 sandbox creations per second. Our evaluation and deployment experience show that these mechanisms reduce environment setup and image-distribution overhead, improve memory efficiency, and preserve latency-sensitive performance under high-density overcommit.
Read the full abstract

All authors Jialiang Huang, Hongxuan Tang, Jingchang Chen, Yuxuan Liu, Yixiao Chen, Yuan Cheng, Yi Tao, Jingli Zhou, Yupeng Chen, Haoyu Chen, Jiarui Wang, Shengkai Lin, Chuqi Zhang, Bryan Lee Teng, Lian Guo, Zhe Fu, Wenjun Gao, Yisong Wang, Liang Zhao, Zehao Wang, Ziwei Xie, Yongqiang Guo, Peixin Cong, Ziyi Gao, Shuiping Yu, Hanwei Xu, Zuofan Wu, Zhizhou Ren, Yuyang Zhou, Bowei Zhang, Zhihuan Huang, Qihao Zhu, Lei Wang, Tianle Lin, Han Yu, Jiewen Hu, Dejian Yang, Shuo Yang, Shanghao Lu, Shaoyuan Chen, Junjie Qiu, Zhangli Sha, Yinmin Zhong, Yongtong Wu, Shiyu Wang, Wei Liu, Bingzheng Xu, Longhao Chen, Qiushi Du, Yuzhen Huang, Shirong Ma, Yaohui Wang, Mingshu Chen, Tongrui Xiong, Y.C. Yan, Haowen Luo, Haofen Liang, Xiaokang Zhang, Weihao Zeng, Runxin Xu, Peiyi Wang, Jinhua Zhu, Ruoyu Zhang, Wenkai Yang, Qi Tang, Jiping Yu, Tian Ye, Ruizhe Pan, Honghui Ding, Xiaodong Liu, Lingxiao Luo, Zhihong Shao, Yuhan Wu, Jibai Lu, Wen Liu, Haoling Zhang, Jingcheng Hu, Yaoyang Ye, Chaofan Lin, Zhaochen Zhang, Jianan Tong, Hengxu Wu, Zhihao Li, Yicheng Wang, Luyao Wang, Yuzhuo Bai, Lingyue Fu, Ruifan Xu, Y.Z. Wang, Zonglin Li, Mingqi Wei, Haiyang Shen, Chengyuan Zhang, Chao Jin, Zili Zhang, R.H. Yang, Xinbo Xu, Jian Zhou, Ruidong Zhu, Yuzhe Guo, Zelun Pan, Shaoheng Nie, Erhang Li, Shuhan Lin, Zheng Liu, Anshuo Chen, Zilong Lyu, Sinuo Cao, Rui Yu, Chuhao Wang, Junyi Guo, Junxiao Song, Kaifeng Chen, Menghao Ye, Junxian Li, Di Wu, Haiyang Ma, Yilun Wang, Haoran Yang, Yizai Cai, Shichun Liu, Yiping Wang, Junbo Sun, Shicheng Xu, Xiao Bi, Ying He, Yichao Zhang, Mingxing Zhang, Liyue Zhang, Panpan Huang, Wenfeng Liang

Machine Learning

Beyond Similarity: Coverage-Aware Prompt Selection for Time Series Forecasting with LLMs

Daeun Ji, Minkyoung Kim, Dongkuk Kim et al. (+3 more)

Similarity-based retrieval is the dominant rule for conditioning large language models (LLMs) in in-context learning, retrieval-augmented generation, and prompt-based time series forecasting. The rule concentrates on near-duplicate candidates, an issue that has motivated diversity-aware retrieval but remains unexamined in other retrieval-conditioned pipelines. We study this issue using prompt-based time series forecasting as a test bed, where a learned prompt pool is retrieved by similarity. Dominant methods in this setting retrieve top-K entries by cosine similarity without redundancy control, producing a bias toward dominant temporal patterns while overlooking rare but informative events. We propose CASP-LLM, a coverage-aware semantic prompting framework that addresses this prompt selection bias by combining usage-tracking and saturating-gate techniques into a coverage regularizer that adds no learnable parameters. On six long-term benchmarks and the M4 short-term benchmark, CASP-LLM matches or improves on similarity-based LLM forecasters on most dataset-horizon settings, with the exceptions of Electricity, M4-Monthly, and the few-shot long-horizon setting. A controlled study locates the failure mode at the cross-batch usage level rather than per-retrieval redundancy: within-retrieval diversification such as MMR does not help, whereas regularizing anchor usage across training does.
Read the full abstract

All authors Daeun Ji, Minkyoung Kim, Dongkuk Kim, Yohan Lee, Beomsoo Kim, Beakcheol Jang

Quantum Physics

Weak Values Beyond the Weak Limit: Measurement-Strength Invariance Under Post-Selection

Cosmin Andrei and Vlatko Vedral

Weak values are conventionally extracted in the limit of vanishing measurement strength. At finite strength, the corresponding post-selected conditional averages generally acquire nonlinear corrections and need not retain their weak value interpretation. Here we identify a class of measurement protocols for which this expectation fails: the conditional value reconstructed from the measuring probe is exactly independent of the probe strength and therefore coincides with the real part of the weak value. We first analyze this phenomenon in the measurement–disturbance protocol introduced by Lund and Wiseman. We then establish conditions under which this invariance persists, showing that it is determined by the interplay between the probe interaction, a subsequent measurement interaction with an ancillary apparatus, and the final post-selection. These results demonstrate that weak coupling, although generally sufficient for accessing weak values, is not always necessary.
Read the full abstract

General Mathematics

Mathematical Principles of a Generalized Tonal Practice

Drew Flieder

We develop mathematical principles for a generalized tonal practice. We begin with an algebraic theory of modes and scales, formalized as groups and torsors over cyclic groups, respectively. We identify three musically salient classes of mode isomorphism—translations, rotations, and deformations. Upon this basis we organize successive layers of harmonic structure into a hierarchy of Grothendieck fibrations, built from scale types (generalizing the diatonic scale), modes (generalizing the diatonic modes), orbit covers (generalizing their covering by tertian triads), deformation families (generalizing chromatic alteration, such as modal mixture and secondary chords), and harmonic-function assignments (generalizing the assignment of tonic, subdominant, and dominant function). Equipping the resulting structure with a generative syntax—generalizing Rohrmeier's generative grammar of tonal harmony—yields a formal definition of a harmonic practice: a harmonic vocabulary together with an assignment of harmonic functions and a generative syntax accommodating hierarchical harmonic organization. We show that harmonic practices form a category, with morphisms constituting an inheritance and enrichment of structure from one harmonic practice to another. The theory originates in the author's own compositional practice, developed initially through the series of compositions Op. 26–31.
Read the full abstract

Robotics

A Horizon-slicing Approach to Minimum Obstacle Displacement Planning for Robot Navigation

Antony Thomas, Giulio Ferro, Fulvio Mastrogiovanni et al. (+2 more)

In this paper, we investigate the Minimum Obstacle Displacement Planning problem from a robot motion planning perspective. The problem involves determining a feasible path to a goal location by displacing movable obstacles when no collision-free path initially exists. We show that this problem is computationally challenging and, in particular, NP-hard when obstacles are modeled as polygons in the plane. Besides an exact formulation of the minimum obstacle displacement problem generalizing other problems in the literature, and the associated optimal solution, this paper proposes an approximate solution that is less intensive from a computational standpoint, and differs from the optimal solution by a fraction of the optimal cost, being able to trade-off between path length and amount of obstacle displacements.
Read the full abstract

All authors Antony Thomas, Giulio Ferro, Fulvio Mastrogiovanni, Michela Robba, Marco Baglietto

Robotics

An Empirical Study and Open Testbed for Federated Fine-Tuning of Vision-Language-Action Models

Zhekai Duan, Kevin Ziyang Xie, Xinyu Tan et al. (+5 more)

Adapting a pretrained Vision-Language-Action (VLA) model to a new robot, environment, or task requires demonstrations that are collected locally and often discarded. Federated learning is a promising approach to exploiting such distributed demonstrations by learning a shared policy. However, whether it can adapt large pretrained VLAs remains an open question, and a lack of reproducible benchmarks for pretrained VLAs and reusable training frameworks makes existing results difficult to compare. In this paper, we conduct a systematic study of federated fine-tuning of three modern pretrained VLA policies on the 40 simulated tasks of the LIBERO manipulation benchmark, and on six real-world tasks in two real-robot experiments, with demonstrations collected across two and three sites, respectively. Our study analyzes the key choices in this setting, spanning multiple federated parameter scopes, three aggregation algorithms, and evaluation under distribution shift. Based on the study, we derive a series of lessons, including the dominance of the federated scope over the choice of aggregation algorithm and the difficulty of matching centralized fine-tuning on physical robots, where cross-site heterogeneity is stronger than simulation captures. We also highlight opportunities for federated VLA learning, such as the ability to match centralized fine-tuning on heterogeneous data, to remain at least as robust as centralized fine-tuning under distribution shift, and to personalize, with each client federating part of the policy and keeping the rest local, which helps where the policy's pretraining is weak but leaves no usable global model. We open-source \decentvla{}, the model- and runtime-agnostic testbed behind the study, to facilitate future research and fair comparisons in federated VLA learning.
Read the full abstract

All authors Zhekai Duan, Kevin Ziyang Xie, Xinyu Tan, Shikai Geng, Chengxu Zhou, Ramana Kompella, Gaowen Liu, Chris Xiaoxuan Lu

Combinatorics

A twelve-term exclusion for the small Davenport constant of $E_2\times C_3^r$

Andreas Volkmann

Let $E_2$ be the extraspecial group of order $3^5$ and exponent three. For every $r\ge1$, we prove that a product-one-free sequence of length $2r+11$ over $E_2\times C_3^r$ cannot contain exactly $2r-1$ central terms. Thus the critical layer with twelve noncentral terms is excluded. The proof combines a relative moment criterion for abelian normal subgroups with a finite theorem in a symplectic four-space over $\mathbb F_3$. Under explicit subspace occupancy bounds, the family of balanced triangles that can actually be completed to a nonfull zero-sum block admits edge weights summing to one on every triangle. A dual cycle argument reduces this assertion to a potential condition on branching edges. All remaining configurations contain one of eleven minimal frames; two separately implemented exhaustive checks verify all their admissible extensions. The unrestricted check has 9544046 leaves. Complete source code and execution records are supplied. The exact value of $\mathsf{d}(E_2\times C_3^r)$ for arbitrary $r$ is not determined by this result.
Read the full abstract

Computation and Language

Automatic multimodal UX improvement recommendations from LLM agent user simulations

Anu Chowdhury, Bin Wu, Hossein A. Rahmani et al. (+1 more)

Evaluating user experience (UX) on live websites through user testing is expensive, subjective, and difficult to scale. LLM agents offer a promising route to automating UX testing by simulating realistic user behaviour. However, existing simulation approaches typically lack multimodality and require time-consuming manual review to extract actionable insights. We formalise UX improvement recommendation from simulation data as a structured natural language generation and ranking problem, and establish an evaluation protocol using expert annotation and LLM-as-a-Judge. We present AMUSER, a multimodal framework which simulates user behaviour and automatically generates prioritised UX improvement recommendations from resulting data. We evaluate AMUSER on commercial websites and show that its recommendations substantially outperform those from text-only simulation (NDCG@3 = 0.758 versus 0.359) at an 89% lower simulation cost. Our results suggest an asymmetric role of multimodality: visual access during simulation improves recommendations through richer traces, while providing visual inputs during recommendation generation can modestly degrade quality. We also discuss practical deployment lessons from applying AMUSER to commercial websites.
Read the full abstract

All authors Anu Chowdhury, Bin Wu, Hossein A. Rahmani, Emine Yilmaz

High Energy Physics (Phenomenology)

CP violation in $B_{(s)}\to\phi K$ decays: Standard Model and isospin-dependent New Physics

Robert Fleischer, Jelle Groot and K. Keri Vos

The penguin-dominated $B\to\phi K$ decays provide sensitive probes of physics beyond the Standard Model. Interpreting them at increasing experimental precision requires control of doubly Cabibbo-suppressed hadronic contributions. Using a factorization approach, we obtain Standard Model benchmarks for the branching ratios and CP-violating observables, including the hadronic phase shift in $B_d^0\to\phi K_{\rm S}$. The $B_s^0\to\phi K_{\rm S}$ channel, in which the corresponding penguin effects are not Cabibbo-suppressed, offers additional insight into these contributions. Comparing the neutral and charged $B\to\phi K$ modes further allows the construction of isospin observables. We discuss their Standard-Model expectations and sensitivity to New Physics in the $I=0$ and $I=1$ sectors. The available data are consistent with the Standard Model, while leaving substantial room for New Physics contributions.
Read the full abstract

Image and Video Processing

Scalable SSIM Estimation from PSNR for Per-Title and Context-Adaptive Encoding Workflows

Luc Trudeau and Maria G. Martini

Modern streaming pipelines run hundreds of candidate encodes per asset to support per-title encoding, shot-based optimization, and context-adaptive ABR ladder construction. These techniques have moved perceptual quality metrics into the critical path: SSIM and VMAF now guide encoding decisions rather than passively monitor them. We measure that SSIM evaluation accounts for 7-35% of x264 encode time at production speed presets, with the cost ratio rising as encoders run faster. We propose ApproxSSIMate, a low-complexity method for estimating SSIM from PSNR combined with reference-sequence statistics computed once per sequence and reused across every candidate encode. This decouples quality estimation from the encode-decode-compare loop, enabling perceptual quality feedback in live encoding and amortizing quality measurement across candidate encodes in per-title workflows. We validate the approach across H.264/AVC, H.265/HEVC, and AV1 on the Objective-1-fast dataset and release the implementation as free and open-source software.
Read the full abstract

Optimization and Control

Searching for Primes: A Neural AlphaZero Approach to a Factoring Game

Marcel Crasmaru

We study a one-player token game on an $N\times N$ board where tokens slide along diagonals or duplicate onto neighbouring ones to form a combinatorial rectangle $R\times S$. A conserved integer weight $W'$ and a strict monovariant guarantee $O(N^2)$-length solutions, placing the game in $\mathsf{NP}$. We prove that reaching a final position factors this $2N$-bit $W'$ into two $N$-bit factors $V,M < 2^N$ that encode the rectangle's rows and columns. Consequently, solving the game for a balanced-semiprime target is equivalent to integer factoring. However, if the target rectangle is known, the solution reduces to two polynomial-time steps: a forced downward chip-flow and a $0/1$-polynomial factorisation leveraging Cohn's theorem. The game's entire difficulty is thus isolated to the initial number-theoretic split. Supplying the popcounts of the factors as a promise preserves this asymptotic hardness but bounds the target search space. We exploit this constrained space using a learned policy/value network and an AlphaZero-style Monte-Carlo tree search, empirically probing the limits of neural look-ahead on a factoring-equivalent environment.
Read the full abstract

Computer Vision

General Collaborative Intelligence: Architecting Cognition for Resilient Multi-Agent Ecosystems

Lei Zhang, Chun Ye, Le Yang et al. (+4 more)

Multi-agent unmanned systems are moving from isolated, ego-centric sensing toward collaborative intelligence, in which distributed agents exchange compact features to overcome a local observation trap that no single agent can escape: occlusions, finite sensor range, and environmental degradation. The field has matured across architectural, communication, embodied, resilience, and trust dimensions, yet existing surveys examine these dimensions in isolation and rarely expose their dependencies. This review offers a unified synthesis through two complementary lenses. The first is a five-dimensional taxonomy spanning collaboration stage, communication paradigm, fusion architecture, learning strategy, and application domain. The second is three cognitive synergy conditions, Semantic Disambiguation, Pragmatic Information Exchange, and Proactive Informational Foraging, that turn cognitive synergy into operational criteria. Across these lenses we survey collaboration architectures and topologies, neural-communication co-design that treats the channel as a differentiable pipeline component, embodied action-perception loops via multi-agent reinforcement learning, and resilience mechanisms for synchronization, uncertainty quantification, and label-efficient learning. We then map these advances onto four operational domains, V2X, unmanned aerial, industrial logistics, and smart cities, and onto the safety-privacy-utility triad. To counter benchmark saturation and evaluation fragmentation, we propose GCI-Bench, a five-pillar scoring protocol with a maturity model that makes the trade-offs of collaborative methods comparable across studies. A critical reflection on reproducibility, the sim-to-real gulf, and conditions under which collaboration degrades performance identifies open challenges and charts directions toward general collaborative intelligence under real-world uncertainty.
Read the full abstract

All authors Lei Zhang, Chun Ye, Le Yang, Zhaozhong Wang, Deng-Ping Fan, Hang Dai, Binglu Wang

Robotics

Transferring the Intelligence of VLMs to Robotic Control

Meng-Hao Guo, Zhe-Han Mo, Jia-Jun Wang et al. (+6 more)

Humans can seamlessly adapt to both physical and digital worlds, suggesting that while a digital-to-real gap exists in embodiment, environment and task, human intelligence itself may transfer across this gap. This naturally raises a fundamental question: can the intelligence of vision-language models (VLMs) similarly generalize from the digital world to the physical world for robotic control? We investigate this question through RoboDawn, a human-intuitive interface that exposes robotic control to an agentic VLM through a compact set of discrete translation, rotation, and gripper commands. Using this interface, the VLM controls a robot in a closed loop: it observes the current visual state, reasons about the next action, executes it, and adapts subsequent decisions to the resulting state. Furthermore, we introduce an in-context learning (ICL) scheme that uses a few demonstrations to ground the VLM in both interface usage and task-solving strategies. Experiments on RoboTwin 2.0 C2R and RoboDojo demonstrate that RoboDawn achieves strong performance without task-specific robot training. In the zero-shot setting, RoboDawn outperforms several strong policies trained on benchmarkspecific robot data, while a single in-context demonstration further yields substantial performance gains and establishes state-of-the-art (SOTA) results. On RoboTwin 2.0 C2R, the success rate increases from 53.2% zero-shot to 73.6% one-shot, exceeding the solid baseline {\pi}0.5 (46.0%). Similar gains are observed on RoboDojo, where success rate improves from 35.67% zero-shot to 47.17% one-shot. The same framework also transfers to real-world robots, performing block-in-basket and block stacking on Franka.
Read the full abstract

All authors Meng-Hao Guo, Zhe-Han Mo, Jia-Jun Wang, Yi Zhang, Kejin Wang, Yi-Xuan Deng, Jia-Peng Zhang, Yongming Rao, Shi-Min Hu