Peer Revenue Impact factor: volatile

Papers

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

All of arXiv

Wednesday 23 September

Software Engineering

Teach-to-Crash: A Closed-Loop Student-Teacher LLM Framework for Collision-Inducing Test Scenario Generation

Zaid Ghazal, Khouloud Gaaloul and Bruce Maxim

Validating Autonomous Driving Systems (ADS) in simulation requires testing architectures that can discover rare, safety-critical failures while generating scenarios that are executable, diverse, and useful for downstream failure analysis. We introduce Teach-to-Crash, a closed-loop testing framework that combines a constrained ego-centric scenario representation, stagnation-aware search control, and a dual-LLM architecture for adaptive failure discovery. A high-reasoning Teacher LLM acts as an adaptive search controller, while a low-reasoning Student LLM emits simulator-executable scenarios in a strict JSON schema. The Teacher intervenes only when rolling collision rate and time-to-collision metrics stagnate, providing strategic guidance to redirect the search. In a CARLA case study with two experimental setups that vary the ego vehicle's speed policy, Teach-to-Crash achieves the highest Collision Hit Rate (90.79%), the shortest mean Time-to-Collision (18.31 s), and a competitive Collision Discovery Rate (136.21). PAFOT attains a higher mean CDR (179.44), but with substantially larger variance. Teach-to-Crash also yields the highest diversity (0.547) and, averaged across both setups on the CARLA Traffic Manager controller, the highest avoidability-based usefulness proxy (60.04%) among the compared methods. These results, within the evaluated CARLA scope, provide evidence that closed-loop dual-LLM reasoning can steer adversarial simulation-based testing over a constrained executable program space, generating failures that are frequent, structurally diverse, and assessed as more frequently avoidable.
Read the full abstract

Quantum Algebra

A diagrammatic presentation for every pivotal pointed fusion category

Chumeng Di and Anup Poudel

We provide a generators and relations presentation of pivotal pointed fusion categories, $Vec(G,\omega,\pi)$. Unlike the well-known skeletal model, our presentation is strict and allows multiple isomorphic objects. Our main tool is skein theory, which allows us to apply topological tools to understand the relations of morphisms in the category. This expository paper provides details of skein-theoretic work in the setup that allows multiple isomorphic labels in the generators, which didn't previously exist in the literature.
Read the full abstract

Machine Learning

KITE: KV-Invariant Transformer Expansion for Efficient Agentic LLM Scaling

Zhiheng Hu, Yixun Wei, Jian Zhou et al. (+8 more)

Scaling a language model is not only a question of final quality: the architectural choice determines how much computation is spent during training, prompt processing, and autoregressive decoding to achieve certain model quality. An ideal model architecture should lower all above computation costs to facilitate scaling to a larger model, while ensure the larger model indeed outperforms smaller baselines. We introduce KV-Invariant Transformer Expansion (KITE), a scaling paradigm that achieves this goal. It trains the model from a smaller size to a larger size (i.e., saving training costs via upcycling), while places newly added parameters in regions that do not affect attention KV. Consequently, during inference, prefilling KV only relies on the smaller part of the model, so the inference costs are saved. As a concrete instantiation, we present Step Scale Transformer (SST), a two-tower decoder in which one tower produces KV and the other reads them. At comparable cumulative training compute, SST, a 67B MoE model with 2.15B active body parameters per decode token, achieves lower training loss than 47B and 63B MoE Transformers with 1.48B and 2.02B active body parameters, respectively, while reducing estimated inference cost by 6.7% and 31.6%.
Read the full abstract

All authors Zhiheng Hu, Yixun Wei, Jian Zhou, Yizhuang Zhou, Ji Li, Xing Chen, Yang Li, Bojun Wang, Yibo Zhu, Xiangyu Zhang, Daxin Jiang

Theoretical Economics

Singleton-Attainability and Transparent Access in Matching

Szilvia Pápai

Matching mechanisms differ in how much of an agent's preference ranking must be determined and reported to obtain a particular object. A mechanism is singleton-attainable (SA) if every object that an agent can obtain through some report can also be obtained by reporting only that object as acceptable. With an SA mechanism, once an attainable object has been identified, the agent need not rank or report any other object. Singleton-attainability identifies a distinct dimension in matching theory and market design: transparent access to attainable outcomes, separate from incentives, stability, welfare or equity. We establish general conditions for SA and derive its strategic implications. Top-lift invariance and truncation invariance together imply SA, while strategyproofness and stability each imply SA. By contrast, no Pareto improvement over a strategyproof, individually rational, and non-wasteful mechanism is SA. In particular, every Pareto improvement over Deferred Acceptance violates SA. We introduce report width, which measures how many acceptable objects may have to be reported to obtain an object. SA mechanisms have report width one. Report width is unbounded for a large class of efficient mechanisms that Pareto-improve Deferred Acceptance. Stable selection with report-induced priorities has width one when priorities are monotone and maximal width under reverse priority dominance. Rank-welfare maximization has width one when the outside-option rank is fixed and maximal width when it is report-dependent. These results reveal a structural divide, which we call the width dichotomy: across all mechanisms and families in our classification and all structural classes we study, report width is either one or unbounded.
Read the full abstract

Fluid Dynamics

Transition from a weak-elastic to an elastic-activated regime in two-dimensional polymeric turbulence

Haoyu Wang, Hangyu Zhu, Xiaoqiu He et al. (+1 more)

We conduct direct numerical simulations to investigate the effects of dilute polymers on flow structure, statistical properties and energy transfer in 2D turbulence. As the Weissenberg number (Wi) increases, the flow undergoes a sharp but continuous crossover between a weak-elastic regime and an elastic-activated regime. In the weak-elastic regime, the flow remains close to its Newtonian counterpart and polymer stresses have weak dynamical effects. In the elastic-activated regime, elastic stresses attenuate vortical motions and promote shear-dominated filaments. Scale-by-scale energy analysis shows that polymer addition weakens both the inverse kinetic-energy cascade and the forward enstrophy cascade in the two regimes, whereas the kinetic–elastic exchange and the inter-scale redistribution of elastic energy differ between them. Kinetic-to-elastic conversion dominates over the analysed range of scales in the weak-elastic regime, whereas a partial transfer from elastic to kinetic energy emerges at small scales in the elastic-activated regime. The inter-scale transfer of elastic energy is dominated by inverse transfer over the large-scale range in the weak-elastic regime but by forward transfer over the small-scale range in the elastic-activated regime. These results link the crossover to the onset of strong elastic feedback from the polymers and the associated reorganisation of the scale-by-scale transfer pathways. Downstream-decay statistics also show the same qualitative two-regime behaviour.
Read the full abstract

All authors Haoyu Wang, Hangyu Zhu, Xiaoqiu He, Yongliang Xiong

Machine Learning

NGN: Learning Neural Network Size as a Differentiable Count

Lixing Li

Neural network size is usually chosen before training, separating architecture selection from weight optimization. We introduce the Neurogenesis Network (NGN), a differentiable parameterization for learning how many ordered structural components a model should use. For each ordered component group, one learnable boundary selects an active prefix while the model parameters are trained. The boundary can grow from a compact initialization and can be deployed by discarding components beyond the learned boundary. Controlled experiments examine convergence of the learned boundary, the performance of deployed prefixes, and comparisons with fixed-size models and alternative approaches to learning capacity. We then apply the same mechanism to MLPs, convolutional and graph networks, Transformers, state-space models, LoRA, and adapters. Across these settings, deploying only the learned prefix usually changes performance little, and the selected architectures perform similarly to fixed models trained at the same size. These results show that structural capacity can be optimized directly as a count.
Read the full abstract

Artificial Intelligence

Sparse-Observation Atmospheric Thermal Forecasting with Physics-Informed Neural Networks for Climate-Aware Digital Twins

Tannaz Goodarzvand Chegini, Elyas Shivanian, Behzad Karimi et al. (+1 more)

Short-horizon forecasts of atmospheric temperature are needed to support climate-aware digital-twin systems, but such forecasts must be produced where thermal observations are incomplete. This study evaluates a physics-informed neural network for potential-temperature forecasting, constrained by a pressure-coordinate thermodynamic advection-source equation and a diabatic-source closure fit from the preceding 12-hour period and frozen before future-time training. Using hourly ERA5 reanalysis at three pressure levels, the model is evaluated as a conditional hindcast at lead times of one, two and three hours against persistence, local-trend, and two matched neural-network baselines, one of which receives the same future meteorological forcing as the PINN, helping distinguish the physical constraint from access to future forcing. In an Oklahoma development case, mean RMSE improvement over the strongest baseline grew from 8.1% at one hour to 23.8% at three hours; under an observation-density sweep down to 5% of candidate locations, this 3-hour advantage remained 14.6–16.9%, with no evidence that lower density improves performance. Under a fixed protocol transferred to an Alabama heat event with three virtual-observation layouts, three-hour improvement ranged 19.7-24.4% with consistent origin-level wins. A parallel Montana stress test, in which fixed pressure levels intersected complex terrain, produced a three-hour degradation of roughly 17.5%, identifying a terrain-related applicability limit of the formulation. Together, these results indicate that the physics constraint's benefit grows with forecast horizon, persists under severe observation sparsity, and transfers across regions, but is bounded by the validity of a fixed vertical-coordinate representation over complex terrain, evidence relevant to physics-constrained components of climate-aware forecasting and digital-twin systems.
Read the full abstract

All authors Tannaz Goodarzvand Chegini, Elyas Shivanian, Behzad Karimi, Faraz Dadgostari

Computation and Language

Ruby-ASR: Evidence-Preserving Supervision for Joint Orthographic and Lexical-Reading Recognition

Hao Shi, Yun Liu, Xuehao Yang et al. (+6 more)

Conventional Japanese automatic speech recognition (ASR) is supervised by an orthographic transcript, although the same written form can correspond to different lexical readings realized in speech. Such utterances receive an identical target, so their reading distinction is absent from the supervision interface and cannot be recovered reliably by post-hoc text-only grapheme-to-phoneme conversion. We present Ruby-ASR, which refines the conventional target into a span-bound orthographic–lexical-reading sequence. Unlike separate full-sentence orthographic and phonological outputs, the ruby representation locally binds each written span to its realized reading and permits deterministic recovery of both views. We instantiate the target under subtitle-style and verbatim-style transcription conventions using a Qwen3-ASR backbone; a mora-level CTC objective provides auxiliary monotonic reading supervision. The experimental results across five Japanese benchmarks show that refining the recognition target can improve lexical-reading recovery without sacrificing readable orthographic transcription. We release the checkpoints and inference code.
Read the full abstract

All authors Hao Shi, Yun Liu, Xuehao Yang, Jun Liu, Chuanbo Hua, Xuanjun Chen, Lianbo Liu, Shiao Zhu, Zixiong Su

Artificial Intelligence

PotARCin: Multi-Dimensional Evaluation of Skill Acquisition in Abstract Reasoning Tasks

Claas Beger, Ryan Yi and Melanie Mitchell

The Abstraction and Reasoning Corpus (ARC) has become a prominent benchmark for evaluating general abstract reasoning and fluid intelligence in AI models. Yet standard ARC evaluation considers only a single capability: producing the correct output grid for a test input. We argue that this narrow format fails to evaluate the diversity of abilities that genuine abstract skill acquisition should enable. We introduce PotARCin, a benchmark that extends ARC by assessing understanding of a task's underlying abstract rule across five dimensions: Definition, Classification, Constrained Generation, Editing, and Inversion. PotARCin employs programmatic methods to generate new task instances and transform given inputs for a given ARC task, enabling dynamic generative sampling beyond fixed input-output pairs. Across five state-of-the-art models evaluated on the ARC-AGI-1 training set, we observe a 25-52 percentage-point performance gap between standard ARC evaluation and evaluation on PotARCin, and find that multi-dimensional evaluation reorders models that standard accuracy ranks alike. We further investigate effects of generative sampling, difficulty of corruption types, and questions of self-consistency, showing that models frequently contradict their own formalized rule even where they have stated it correctly. We also introduce P-ARC, a held-out hand-crafted test set, on which models achieve 1-8% accuracy across all five dimensions, underscoring the importance of more holistic evaluations of abstract reasoning capabilities.
Read the full abstract

Machine Learning

SR-Fraud: An Outcome-Supervised Reflective LLM Agent Framework for Non-Stationary Payment Fraud Detection

Xuwei Tan, Yao Ma and Xueru Zhang

Real-time payment fraud detection is a non-stationary streaming prediction problem: adversaries adapt before supervised labels mature, and localized burst attacks can cause losses before retraining. Production systems typically rely on tabular classifiers and rules, which can struggle to capture these emerging sequential patterns before periodic retraining occurs. We present SR-Fraud, an outcome-supervised reflective LLM framework that decouples request-time decisions from offline adaptation. A frozen, stateless agent scores each transaction from a Hybrid Episodic Window to track behavioral shifts, while an offline reflection agent proposes boundary hypotheses from matured errors. A deterministic verifier then admits only supported hypotheses into an executable knowledge state. On a production payment-fraud benchmark, SR-Fraud improves all detection metrics over its frozen decision agent, obtains higher point estimates than static and periodically retrained CatBoost, and detects an emerging fraud burst.
Read the full abstract

Artificial Intelligence

Memory Control Signals Emerge Before Action in Long Horizon Agents

Mingxuan Wang, Guorun Yao, Fei Luo et al. (+6 more)

Long horizon language model agents continuously accumulate interaction history, increasing computational cost while making relevant information harder to preserve and reuse. Existing context management methods mainly focus on how to compress or retrieve history, but largely leave open whether the model itself already represents the need for these memory operations before they occur. We study the hidden state immediately before each agent action and find that compression and recall needs are already encoded in the model's internal representations. These signals cannot be explained by simple context length or interaction progress, and they exhibit distinct formation patterns across model depth. We further show that most memory decision information is preserved in a compact recent context, while selectively restored historical evidence complements the long range dependencies that recent context misses. Based on these findings, we propose Preaction Memory with Evidence Retrieval (PaMER), which combines state guided compression with external evidence retrieval. PaMER+ further introduces step level evidence selection to recover only the historical information required by the current task. Experiments on WorkBuddyBench, across multiple context management baselines and model backbones, show that our framework substantially reduces context consumption while maintaining competitive task performance.
Read the full abstract

All authors Mingxuan Wang, Guorun Yao, Fei Luo, Yinglong Guo, Chao Ning, Bo Wang, Hongyue Chen, Yanbiao Ma, Jungong Han

Game Theory

Pacing Equilibria in Abstract Mechanisms

Salam Afiouni and Christian Kroer

Digital platforms increasingly rely on automated budget-management systems to regulate participation across allocation opportunities. A common tool is multiplicative pacing, which scales buyers' bids so that campaign-level budgets are spent gradually. Although pacing is operationally simple, its aggregate behavior depends on the underlying mechanism. In first-price single-item auction markets, pacing enjoys strong structural and computational properties that often fail in second-price auctions. It is unclear whether these properties extend to richer platform mechanisms. We develop a unified theory of pacing equilibria across a hierarchy of mechanisms. We first show that, even in first-price position auctions, the market-equilibrium interpretation of the single-item benchmark can fail. We then identify bid-maximizing pay-your-bid mechanisms as a broad class in which pacing equilibria exist, are unique, and admit an Eisenberg-Gale-type convex-program characterization. This implies efficient computation, Pareto-efficiency, and liquid-welfare guarantees, enabling pacing integration into large-scale optimization-based allocation routines. Finally, for abstract mechanisms satisfying a payment monotonicity condition, we prove equilibrium existence by smoothing discontinuities. With additional conditions and appropriate tie-breaking, we obtain uniqueness, revenue maximality among budget-feasible pacing vectors, and shill-proof implementability, and provide convergent budget-adjustment dynamics for approximate equilibria. Overall, many favorable properties of first-price pacing extend beyond single-item auctions under appropriate mechanism-level conditions.
Read the full abstract

Artificial Intelligence

Hunyuan-A13B Technical Report

Tencent Hunyuan Team, Ao Liu, Botong Zhou et al. (+72 more)

We present Hunyuan-A13B, an open-source large language model based on a Mixture-of-Experts architecture. It contains 80 billion total parameters but activates only 13 billion during inference, balancing model capability, computational efficiency, and deployment cost. The model is pretrained on a rigorously filtered 20T-token corpus with enhanced STEM data curation, improving factual reliability and reasoning ability. High-quality supervised fine-tuning and large-scale reinforcement learning further enhance its overall performance. Hunyuan-A13B also introduces a dual-mode Chain-of-Thought framework that adapts reasoning depth to task complexity: fast thinking for routine queries and slow thinking for complex, multi-step problems. Evaluations show competitive performance across mathematics, science, programming, general language understanding, and agent tasks, often approaching that of much larger models. Its high inference throughput makes it suitable for latency-sensitive applications. We release Hunyuan-A13B to support open research and practical LLM deployment.
Read the full abstract

All authors Tencent Hunyuan Team, Ao Liu, Botong Zhou, Can Xu, Chayse Zhou, ChenChen Zhang, Chengcheng Xu, Chenhao Wang, Decheng Wu, Dengpeng Wu, Dian Jiao, Dong Du, Dong Wang, Feng Zhang, Fengzong Lian, Guanghui Xu, Guanwei Zhang, Hai Wang, Haipeng Luo, Han Hu, Huilin Xu, Jiajia Wu, Jianchen Zhu, Jianfeng Yan, Jiaqi Zhu, Jihong Zhang, Jinbao Xue, Jun Xia, Junqiang Zheng, Kai Liu, Kai Zhang, Kai Zheng, Kejiao Li, Keyao Wang, Lan Jiang, Lixin Liu, Lulu Wu, Mengyuan Huang, Peijie Yu, Peiqi Wang, Qian Wang, Qianbiao Xiang, Qibin Liu, Qingfeng Sun, Richard Guo, Ruobing Xie, Saiyong Yang, Shaohua Chen, Shihui Hu, Shuai Li, Shuaipeng Li, Shuang Chen, Suncong Zheng, Tao Yang, Tian Zhang, Tinghao Yu, Weidong Han, Weijie Liu, Weijin Zhou, Weikang Wang, Wesleye Chen, Xiao Feng, Xiaoqin Ren, Xingwu Sun, Xiong Kuang, Xuemeng Huang, Xun Cao, Yanfeng Chen, Yang Du, Zhen Yang, Yangyu Tao, Yaping Deng, Yi Shen, Yigeng Hong, Yiqi Chen

Statistics Applications

Bayesian calibration of adaptive-behavior SIR models for multi-wave COVID-19 incidence in New York City

Luis A. Barboza, Carlos Pasquier, Baltazar Espinoza et al. (+1 more)

Epidemic incidence reflects both transmission dynamics and adaptive human behavior, yet these mechanisms may be difficult to distinguish from aggregate case data alone. We calibrated four susceptible–infected–recovered (SIR) specifications to weekly confirmed COVID-19 incidence in New York City from June to December 2020, comparing a single continuous SIR trajectory, a wave-initialized SIR model, and two adaptive-behavior models with either shared or wave-specific transmission. Inference was performed using rejection Approximate Bayesian Computation (ABC), and in-sample reconstruction was assessed using root mean squared error (RMSE) and the weighted interval score (WIS). Reinitializing the epidemic state by wave produced the largest structural improvement over the continuous SIR trajectory, reducing mean-based RMSE by 48.6% and WIS by 17.6%. Adding delayed prevalence-dependent behavioral adaptation with shared transmission further reduced mean-based RMSE by 23.4%, but yielded essentially unchanged WIS relative to the wave-initialized SIR model. Allowing transmission to vary by wave did not provide a consistent additional advantage and produced strongly asymmetric posterior-simulation trajectories. Behavioral sensitivity, response midpoint, and delay remained only weakly to partially identified. The clearest posterior structure was a negative association between transmission intensity and the behavioral midpoint, indicating that higher transmission could be compensated by behavioral responses activated at lower prevalence. Sensitivity to ordered behavioral priors further showed that reconstruction and behavioral inference depend materially on structural prior assumptions. These results suggest that adaptive mechanisms can improve multi-wave incidence reconstruction, while aggregate incidence alone is insufficient to sharply separate transmission from behavioral adaptation.
Read the full abstract

All authors Luis A. Barboza, Carlos Pasquier, Baltazar Espinoza, Fabio Sanchez

High Energy Astrophysical Phenomena

A multi-telescope fit to the FRB population, allowing for FRB repetition

Jordan L. Hoffmann, Clancy W. James, Jason X. Prochaska et al. (+3 more)

Fast radio bursts (FRBs) are millisecond-duration radio pulses with extragalactic origins. A majority of FRBs have only had a single pulse detected from them; however, an increasing number have been identified as repeaters. Here we present a full redshift – dispersion measure ($z$-DM) analysis including a simple power-law model of repetition rates. We assume all FRBs are drawn from the same intrinsic population and fit these FRB population parameters using data from ASKAP, Murriyang (Parkes), DSA, FAST, MeerKAT and CHIME catalogue 1. We do not utilise data from CHIME catalogue 2 yet, as we cannot robustly identify all repeaters in the sample. We constrain the minimum repetition rate above $10^{39}$ erg to $R_{\mathrm{min}}=-4.23^{+1.10}_{-1.74}$ d$^{-1}$, the maximum repetition rate to $R_{\mathrm{max}} \gtrsim 1$ d$^{-1}$ and the repetition rate index to $R_{\gamma}=-2.20^{+0.12}_{-0.22}$. These parameters do not show significant correlation with any other FRB parameters. Our values are an improvement on existing results and are self-consistent with the DM distribution of CHIME and observed ratios of repeaters to non-repeaters across the sky. We also provide updated estimates of the FRB energy function and host galaxy parameters.
Read the full abstract

All authors Jordan L. Hoffmann, Clancy W. James, Jason X. Prochaska, Ines Pastor-Marazuela, Kritti Sharma, Liam Connor

Analysis of PDEs

Low Mach number limit around the planar diffusion wave for 3D Navier-Stokes-Fourier equations

Qiangchang Ju, Rui Li and Fanrui Meng

We investigate the low Mach number limit of the three-dimensional full compressible Navier-Stokes-Fourier (NSF) equations on \(\mathbb{R}\times\mathbb{T}^2\) for two different classes of initial data, corresponding respectively to a well-prepared regime and an ill-prepared regime. The density and temperature are allowed to approach different asymptotic states at infinity. For the well-prepared regime, the solutions of compressible NSF equations converge to a planar diffusion wave solution globally in time as the Mach number tends to zero, where the difference between the states at the far fields is small independently of the Mach number and the non-zero modes of the initial perturbations are exponentially small. Moreover, the optimal time-decay rate can be obtained. It can be viewed as the first global-in-time result on the low Mach number limit of three-dimensional NSF equations with large temperature variations. The corresponding local-in-time result is proved while the difference between the states at the far fields can be arbitrarily large for the ill-prepared data.
Read the full abstract

Machine Learning (Statistics)

Multitask Regression with Pairwise Fusion

Xiaodong Li and Zhentao Li

We study multitask regression when coefficient sharing can differ by predictor. For a given predictor, many tasks may have the same coefficient while a few differ, and the exceptional tasks need not be the same for another predictor. We describe this structure by two quantities: the number of active predictors and the total number of task coefficients that differ from the most common value for their predictor. We estimate the coefficient matrix by penalizing all pairwise coefficient differences across tasks, with an additional group penalty when predictor selection is needed. The resulting upper and lower bounds have the same dependence on these two quantities. We also consider the stronger setting in which a large set of tasks shares one entire coefficient vector. Under explicit sample-size conditions, the same pairwise estimator pools those tasks exactly, while allowing the remaining tasks to differ. Simulations and household energy data illustrate the transition between broad sharing and task-specific coefficients.
Read the full abstract

Artificial Intelligence

EnSIMem: Entity-Structured Indexing for Long-Term Agent Memory

Xuanyu Meng, Xing Fan, Xinyi Fan et al. (+3 more)

An agent that interacts with users over long periods must recall facts, preferences, events, and changes from a continuously growing interaction history. Existing memory systems often compress interactions into generic summaries or retrieve anonymous text chunks, making it difficult for an agent to identify the correct entity, property, and supporting evidence. We present EnSIMem, an entity-structured long-term memory architecture for an agent. During offline construction, the system organizes interactions into theme-coherent episodes and builds dialogue-grounded index entries of the form [entity][entity_type][property: value]. Each entry preserves its source turns, temporal information, and available multimodal fields. During online interaction, the agent's request is decomposed into evidence requirements whose properties are aligned with the memory index. Entity-property lookup and adaptive retrieval then collect the evidence needed for point, temporal, compositional, and aggregation reasoning. The agent generates its response from the preserved source evidence rather than from lossy memory summaries. On long-term agent-memory benchmarks, EnSIMem achieves high answer accuracy while maintaining compact, evidence-focused contexts. These results show that entity-structured indexing and episode-level provenance provide a reliable foundation for long-term memory in agents. The code of our model is available at https://github.com/RamonMeng/EnSIMem.
Read the full abstract

All authors Xuanyu Meng, Xing Fan, Xinyi Fan, Chenlei Guo, Yixuan Xie, Jiawei Han

Machine Learning

Graph Learning with Spectral Connectivity Priors for Scarce Data

Mingxiao Liu, Bahar Oveisgharan, Bingyan Zou et al. (+3 more)

Learning a sparse graph from scarce data is practically important but challenging. Motivated by the desirable combination of local sparsity and strong global connectivity exhibited by expander-like graphs, we propose spectral connectivity-regularized graph learning (SCoGL), a framework that incorporates a family of Laplacian spectral priors to explicitly promote global connectivity. Specifically, SCoGL augments a combinatorial-Laplacian-constrained graphical lasso (GLASSO) objective over a target adjacency matrix $\mathbf{W}$ with a general connectivity prior computed from Laplacian eigenvalues. We derive gradients for several representative connectivity priors and develop a projected gradient descent (PGD) algorithm with Armijo backtracking to efficiently optimize $\mathbf{W}$. Experiments show that the proposed SCoGL variants improve graph recovery and enhance downstream tasks such as graph signal denoising when signal observations are scarce.
Read the full abstract

All authors Mingxiao Liu, Bahar Oveisgharan, Bingyan Zou, Gene Cheung, H. Vicky Zhao, Feifei Gao

Artificial Intelligence

TimeEvo: Failure-Driven Self-Evolution of a Time Series Agent

Jie Yang, Yan Zheng, Jiarui Sun et al. (+8 more)

Time series agents answer analytical questions by calling external tools, and which tools they carry is decided by people before the agent runs. However, we identify two failures in this setup. Human-Agent Tool Misalignment: a library of 21 expert-curated tools helps on some tasks and hurts on others, dropping anomaly accuracy under every backbone we test. Silent Harm: one round of generic self-revision changes 147 answers and breaks 56 of them, while the final score moves by less than a point. Both follow from the same gap: whether a tool helps is decided question by question at runtime, while tools are supplied in advance and judged by a single average. To address this, we propose TimeEvo, which clusters an agent's diagnosed failures into capability gaps, plans a measurement for each, synthesizes evidence-only tools that fill them, and admits the candidate library only through a paired admission gate. Experiments on ten time series QA tasks and three backbones show that TimeEvo, starting from an empty library, improves accuracy on every task and every backbone, and that a library grown on a cheap model still gains when it is installed into stronger ones. Code is available at https://github.com/Muyiiiii/TimeEvo.
Read the full abstract

All authors Jie Yang, Yan Zheng, Jiarui Sun, Xiran Fan, Junpeng Wang, Liang Wang, Zelin Xu, Qinghua Liu, Zhengyu Fang, Yiwei Cai, Philip S. Yu

Artificial Intelligence

DRSR: Learning Set-Level Deletion Risk for Efficient Long-Horizon Agents

Mingxuan Wang, Bo Wang, Fei Luo et al. (+6 more)

Long-horizon language-model agents accumulate reasoning traces, tool exchanges, and observations whose relevance changes with the current decision. Existing compression strategies often score historical units independently, but the safety of deleting several units is generally not determined by their singleton scores: redundant evidence, accumulated small effects, and the information that remains after deletion all matter. We introduce Direct Relational Set-Risk Pruning (DRSR), which formulates agent-history compression as risk-constrained selection over deletion sets. Offline, DRSR constructs exact counterfactual supervision by jointly deleting protocol-valid history Blocks and measuring the change in teacher-forced likelihood of the same recorded next output. A lightweight scorer then predicts set-level harm from online-visible relations between candidate history and the current pre-action state, together with deleted-retained and pairwise set structure. At deployment, DRSR evaluates a small set of structurally valid deletion candidates with the lightweight scorer and removes the largest feasible set under recency, protocol, budget, and learned-risk constraints, abstaining when no set is sufficiently safe. On WorkBuddyBench Full260, DRSR increases mean reward from 0.699 to 0.802 while reducing total model tokens by 20.820%. On the fixed Eval40 comparison, it obtains 0.794 reward at 1.211M tokens per task, using 35.850% fewer tokens than the uncompressed agent. Mechanistic analyses and ablations further show that decision-conditioned relations, retained-context information, pair interactions, and abstention each contribute to reliable pruning.
Read the full abstract

All authors Mingxuan Wang, Bo Wang, Fei Luo, Guorun Yao, Chao Ning, Yinglong Guo, Hongyue Chen, Yanbiao Ma, Jungong Han

Robotics

BranchDrive: A Branch-Structured Dataset for Action-Conditioned Driving Prediction

Feeza Khan Khanzada, Sudarshan Sridhar and Jaerock Kwon

Most autonomous-driving datasets record only the action executed by a behavior policy and the single future that followed, providing limited supervision for comparing alternative ego decisions. We introduce BranchDrive, a branch-structured CARLA dataset and benchmark that pairs one canonical pre-decision history with one nominal expert future and twelve physically executed intervention futures spanning acceleration, braking, and left- and right-steering policies at three magnitudes. Each intervention lasts 2.5 s and is followed by expert recovery. Following control-compliance, modality-completeness, replay-fidelity, and action-leakage audits, the frozen benchmark contains 606 independent branch groups and 7,878 associated trajectories. We evaluate prediction of six continuous short-horizon outcomes and a ten-step ego trajectory using action-only, history-only, structured, visual, multimodal, and privileged bird's-eye-view models. On the held-out test split, the structured history-and-action model achieves a macro normalized mean absolute error of 0.5036 and an average displacement error of 2.2042 m, significantly outperforming both restricted baselines. In full-information offline evaluation, its outcome-derived selector increases balanced policy value from 0.5364 to 0.5704 and reduces normalized regret from 0.2674 to 0.1495 relative to the frozen action prior. However, a validation-calibrated minimum-separation guard rejects every intervention, showing that conservative execution remains unresolved. BranchDrive therefore supports action-conditioned short-horizon prediction and fixed-bank offline decision evaluation, but does not establish exact causal effects, binary safety prediction, or closed-loop safety improvement.
Read the full abstract

Computer Vision

High Dynamic Range Video Reconstruction from Single-Exposure Raw Sequences

Tao Zhang, Peixian Su, Xingyu Gao et al. (+6 more)

Due to the limited dynamic range of conventional image sensors, captured low dynamic range (LDR) video often suffers from highlight clipping and shadow detail loss, making high-quality high dynamic range (HDR) reconstruction from single-exposure sequences highly challenging without alternating exposures or extra hardware. Alternating-exposure HDR methods sacrifice frame rate and struggle with motion alignment, making them impractical for real-world capture. To address this, we propose RawHDRV, an end-to-end framework for single-exposure Raw video HDR reconstruction, that fundamentally exploits the linear response and channel-specific characteristics of Bayer data. Specifically, it features a channel-decomposition temporal alignment and fusion strategy that processes Bayer channels separately to exploit their distinct exposure characteristics, together with exposure-aware weighted fusion. It further incorporates an exposure complementarity mask-guided restoration module that leverages inter-frame exposure redundancy to adaptively fuse reliable information and suppress saturation artifacts, and introduces a mask-guided color loss that combines normalized error constraints with gradient smoothing to enhance highlight recovery. Furthermore, we construct a large-scale mobile Raw-HDR video dataset with per-frame HDR annotations. Experiments show that our method achieves the state-of-the-art results in all metrics, demonstrating superior spatial quality and temporal stability under extreme exposure conditions. The code is available at https://github.com/supeixian/RawHDRV.
Read the full abstract

All authors Tao Zhang, Peixian Su, Xingyu Gao, Yunhao Zou, Yu Lu, Zunjie Zhu, Bolun Zheng, Ying Fu, Chenggang Yan

Artificial Intelligence

CAVEAT: Towards Robust Computer-Use Agents in Incentive-Misaligned Environments

Yuxuan Li, Will Epperson, Wesley Deng et al. (+1 more)

Computer-use agents (CUAs) increasingly act on behalf of users online. What happens when the environments they operate in have incentives of their own? Online marketplaces, for example, may favor some products over others, steering agents away from the user's objective. Existing CUA benchmarks cover cooperative settings or explicit attacks, but do not test whether agents preserve user objectives when the environment itself has a stake in the outcome. We introduce CAVEAT, a controlled benchmark spanning nine marketplace environments and a taxonomy of eight common steering mechanisms. Across five model families, agents purchase the user-optimal product in 78.6% of matched-control episodes but only 17.3% when steering mechanisms are enabled. Larger models and more reasoning improve robustness, but substantial failures persist. Trajectory analysis and targeted ablations identify three weaknesses in how agents decide: they (1) prematurely narrow the set of alternatives they consider, (2) impose priorities the user never stated, and (3) commit before resolving decision-relevant evidence. Guided by this diagnosis, we develop CAVEAT-Harness, which targets these failures and raises the optimal purchase rate by up to 80.0 percentage points, and show that targeted post-training further improves a smaller open model. These results establish incentive robustness as a distinct challenge for delegated agents, diagnose failure modes, and show how targeted interventions can substantially improve robustness.
Read the full abstract

All authors Yuxuan Li, Will Epperson, Wesley Deng, Zezhou Huang

Numerical Analysis

Layer-wise discrete certificates in multicomponent reacting-flow discretizations

Jingchao Zhang

Multicomponent reacting-flow discretizations must conserve total species mass, preserve positive mass fractions, and dissipate a convex mixing potential. These properties are controlled at different stages of a solver, however, and a certificate established at one stage need not survive the next. We study this compatibility problem for mixture-averaged diffusion, non-orthogonal finite-volume reconstruction, and positivity-preserving Patankar time integration. The analysis is isothermal, uses prescribed density, and uses synthetic detailed-balance reaction networks for the chemistry tests. At the face level, the entropy-variable jump is shown to equal a log-mean Gibbs Hessian applied exactly to the mass-fraction jump. Requiring conservation and entropy neutrality of the mass-correction term on the simplex tangent space then yields a unique componentwise-positive face composition and an exact two-point entropy certificate. The arithmetic-weight face criterion is also satisfied by production neutral-air/carbon data and a quasi-neutral ambipolar-air control, but a face-level certificate does not automatically extend to subsequent discretization layers. A weighted least-squares non-orthogonal correction can, in the tested prototype, convert a dissipative two-point operator into an anti-dissipative multipoint operator. Likewise, Patankar rescaling can preserve positivity and conservation while increasing the free energy; an explicit finite-step rational counterexample establishes this failure independently of ensemble statistics. A fully discrete two-cell example further shows that a positive face entropy margin does not guarantee positivity of the fully discrete update. Thus conservation, positivity, and free-energy dissipation are layer-specific properties and require distinct discrete certificates.
Read the full abstract

Algebraic Geometry

Rigid and obstructed tangent bundles on Calabi–Yau threefolds

Xueyuan Wan

We address two questions of Huybrechts concerning rigidity and singular deformation spaces of tangent bundles on Calabi–Yau threefolds, without assuming simple connectedness. Our first example is a free $(\mathbb Z/2)^3$ quotient of a smooth intersection of four quadrics in $\mathbb P^7$. Its tangent bundle is stable and infinitesimally rigid, and it also provides a three-dimensional counterexample to the proposed classification in Peternell's Question~1.6. For a classical Igusa quotient of a product of three elliptic curves, we determine the analytic semiuniversal deformation germ of its polystable tangent bundle, keeping the underlying threefold fixed. This germ is isomorphic to the product of two copies of the union of the three coordinate axes in $\mathbb C^3$. It is therefore reduced and singular, with nine two-dimensional irreducible components.
Read the full abstract

Dynamical Systems

A note on bistability of a two-gene competitive system

Eduardo D. Sontag

Self-regulation together with mutual promoter competition provides a simple mechanism for bistability in gene-regulatory models. We study a two-gene system with regulatory terms of Hill exponent one, allowing distinct basal production rates and distinct degradation rates. Each gene product, when bound to its own promoter, may enhance or reduce production relative to the basal rate, while the two products compete through promoter occupancy. We show that the system has at least one and at most three equilibria in the positive quadrant. Exactly two positive equilibria can occur only if one nullcline intersection is degenerate; consequently, a configuration in which all positive nullcline intersections are transverse has either one or three positive equilibria. If there are exactly three distinct positive equilibria, then all three are automatically hyperbolic: the two outer equilibria are asymptotically stable nodes and the middle equilibrium is a saddle. Moreover, every positive solution converges to an equilibrium. Hence the positive quadrant is the disjoint union of the basins of attraction of the two stable nodes and the one-dimensional stable manifold of the saddle, yielding global bistability.
Read the full abstract

Robotics

Banana Kick: Response-Informed Skill Evolution for Humanoid Soccer

Hao E. Zhang, Ruize Geng, Raihan Haque et al. (+5 more)

Humanoid kicking requires coordinated whole-body motion and precise contact, while a banana kick demands contact mechanics that generate ball spin and aerodynamic curvature. Motion imitation provides a reliable ordinary-kick prior, but reinforcement learning may improve shot speed and placement accuracy without changing the underlying kicking technique. Adapting this prior to a qualitatively different contact-rich skill can fail even when the reward is dense and optimization remains stable. The failure occurs when the task objective is locally flat over the current policy's responses. We term this condition first-order learning starvation. To address it, we propose response-informed skill evolution (RISE), a closed-loop objective-continuation method for policy adaptation. RISE ranks bounded objective changes using response sensitivity estimated from cached rollouts and accepts updates only when they produce verified response progress while preserving kicking reliability. Our analysis shows that rescaling a saturated spin reward cannot recover first-order sensitivity at zero spin, whereas adapting coupled contact responses can provide a learnable path to spin generation. We integrate RISE into a humanoid kicking pipeline under calibrated contact and Magnus-force aerodynamics, and sim-to-real transfer. Experiments show that RISE evolves the ordinary kick into a high-spin curved kick with 11.55 rad/s mean ball spin, improves the mean evaluation score by 19.8% over a learning-progress curriculum, and raises joint target attainment from 15.2% to 50.9%. Ablations and response diagnostics support the mechanism, while 30 motion-capture-recorded physical trials demonstrate consistent hardware transfer of the learned curved kick. Project website: https://haozhang-thu.github.io/bananakick/
Read the full abstract

All authors Hao E. Zhang, Ruize Geng, Raihan Haque, Khalil Zbiss, Guanyang Luo, Hui-ping Wang, H. Eric Tseng, Ding Zhao

Rings and Algebras

Toeplitz multiplication and graded factorization of determinant recurrences

Max A. Alekseyev and Dmitry I. Khomovsky

We study how determinant recurrences of banded Toeplitz matrices behave when their Laurent symbols are multiplied. Two classical structures underlie the problem. Clean banded Toeplitz determinants have a root-product description of their recurrences, while a product of two finite Toeplitz sections differs from the finite section of the product by finite-rank corner corrections. We connect these structures by identifying multiplication-induced boundary defects with exterior-degree sectors of the determinant recurrence. To each polynomial core we associate recurrence polynomials for all exterior degrees. Multiplication of cores becomes a graded convolution via composed products, while complementary degrees are related by a scaled reciprocal duality. On the matrix side, one-sided factors produce a defect filtration whose successive pieces are exactly these sectors. For general two-sided factors, independently weighting the two corner corrections gives an interval filtration: increasing either defect order adds one adjacent sector, and the cumulative recurrences are generically minimal. The multiplication-induced same-corner minors used in these filtrations lie in the standard row-column state module, whereas cross-corner imbalance corresponds instead to Laurent recentering. Opposite Laurent recenterings realize the individual sectors directly, and for several factors the admissible recenterings form an explicit lattice polytope. The pentadiagonal case gives the basic $1+4+1$ decomposition. The resulting framework gives a factor-level description of how finite-section boundary effects generate the recurrence spectrum of structured banded Toeplitz products.
Read the full abstract

Probability

Long-time behavior of McKean–Vlasov stochastic systems with singular coefficients

Shangyun Hua, Tao Wang and Longjie Xie

We develop a quantitative framework for the long-time behavior of McKean–Vlasov stochastic differential equations with singular coefficients and possible phase transitions. The framework separates the existence of invariant measures from their uniqueness. For existence, we introduce a generalized Lyapunov condition with a one-level trapping mechanism, which requires dissipativity at only one admissible moment level rather than global contraction. This permits distribution dependence with supercritical growth and thus is more compatible with phase-transition models. For uniqueness and convergence, we establish an anchored uniqueness and quantitative ergodicity principle. An explicit \(L^1\)-smallness condition, expressed through a convolution kernel built from derivative estimates of the anchored frozen semigroup, yields uniqueness and transfers exponential or polynomial mixing rates of the frozen dynamics to the nonlinear McKean–Vlasov system. Moreover, the principle is sensitive to the choice of topology: different distances lead to different perturbation kernels and hence different stability thresholds. A local version gives local uniqueness and quantitative attraction near a prescribed equilibrium. We apply the framework to two representative models. For non-symmetric granular media dynamics, we derive two explicit uniqueness thresholds that reveal the topology-sensitive nature of the theory. For the dynamical Curie–Weiss model, the Wasserstein-1 criterion recovers the sharp bifurcation threshold up to the critical equality. We identify all invariant measures, establish basin-dependent exponential convergence in the phase-transition regime, and show that the loss of anchored smallness at criticality leads to polynomial slowing down.
Read the full abstract