| 1 | Nonequilibrium Phases of Repulsive Self-Attention: Chaos, Attention Condensation, and Emergent Locality | Qucheng Gao, Zuyi Yang, Xiao Chen | cond-mat.dis-nn | 2026-09-23 |
| 2 | Minimal-Norm Univariate Two-Layer ReLU Classification: Exact Solutions and Global Optimality with Skip Connections | Karolina Drabik, Ben Lewis, Antoni Puch +4 | cs.LG | 2026-09-23 |
| 3 | Repairability of Inexact Solvers in Recursive State Estimation with Machine Learning | Yanjun Ji, Dennis Willsch, Orkun Şensebat +8 | quant-ph | 2026-09-23 |
| #4 | From Agent Output to Authorized Transition | Christopher Koch | cs.SE | 2026-09-23 |
| #5 | Finite-Sample Probabilistic Safety Certification for AI-Based Grid-Edge Coordination | Yihong Zhou, Hanbin Yang, Thomas Morstyn | cs.AI | 2026-09-23 |
| #6 | Exact Feedback Is Not Control: Evaluating Text-based Closed-Loop Revision in LLMs | Haitong Jiang, Chunlin Liu, Yile Wang +1 | cs.CL | 2026-09-23 |
| #7 | Probabilistic and Geometry Aware Neural Surrogate of Scrape Off Layer Plasma Simulations | Gabriele Gianuzzo, Stefan Dasbach, Fleur Hendriks +2 | cs.LG | 2026-09-23 |
| #8 | Exact Quantile Balancing and Load-Error Injection for Mixture-of-Experts | Pit Neitemeier, Jiaze Li, Alessio Serra +2 | cs.LG | 2026-09-23 |
| #9 | Evaluating Open-Weight LLMs for Turkish Domain Documents Under Retrieval and Hardware Constraints | Imtiaz Ul Hassan, Öykü Akbulut, Onur Kaya +4 | cs.CL | 2026-09-23 |
| #10 | Shared Global KV with Layer-Specific Local History | Xinglang Xian | cs.LG | 2026-09-23 |
| #11 | Private Decentralized Optimization with Noise Reduction and Bias Correction | Yizhao Fan, Wenjian Luo, Jiaojiao Zhang | cs.LG | 2026-09-23 |
| #12 | VCMM: Variance-Calibrated Momentum for Multimodal Learning | Zhongjing Gu, Chenyang Huang, Yufa Feng +3 | cs.LG | 2026-09-23 |
| #13 | BEE: Intervention-Adaptive Real-World Reinforcement Learning with Vision-Language-Action Models | Weihui Zhao, Xiaohan Yan, Zunian Wan +14 | cs.RO | 2026-09-23 |
| #14 | Multitask Regression with Pairwise Fusion | Xiaodong Li, Zhentao Li | stat.ML | 2026-09-23 |
| #15 | DRSR: Learning Set-Level Deletion Risk for Efficient Long-Horizon Agents | Mingxuan Wang, Bo Wang, Fei Luo +6 | cs.AI | 2026-09-23 |
| #16 | The Risk-Sensitive Schrödinger Bridge: Is Not a KL Projection | Hamidreza Behjoo | cs.IT | 2026-09-23 |
| #17 | Learning Spectral Allocation: A Fractional Diffusion Framework for Adaptive Volumetric Segmentation | Yi-Hui Shen, Tie-Qiang Li | cs.CV | 2026-09-23 |
| #18 | Phonemizing User-Generated Text: A Benchmark, Taxonomy, and Compositional Approach | MinJu Jeon, Younghan Park, Han Sung Park +3 | cs.CL | 2026-09-23 |
| #19 | XLOG: A CUDA-Native Engine for Neurosymbolic Integration | Levi Dubrovin, Nikita Pospelov, Kirill Sabitov | cs.AI | 2026-09-23 |
| #20 | ZO-COSMO: Index-Free One-Hop Mixing for Decentralized Zeroth-Order Optimization | Shengjun Zhang, Tingyi Liu, Heng Zhang +1 | cs.LG | 2026-09-23 |
| #21 | Learning Risk Scores Robust to Unobserved Confounders | Ryan Edmonds, Yingxiao Ye, Sina Aghaei +3 | cs.LG | 2026-09-22 |
| #22 | Local Evidence and Geometric Readout Repair in Trained GNNs | Nadi Tomeh, Hugo Attali | cs.LG | 2026-09-22 |
| #23 | Policy-as-Skill: Governed LLM Decision Support with Evidence, Deterministic Control, and Audit | Kabeh Mohsenzadegan, Vahid Tavakkoli, Kyandoghere Kyamakya | cs.AI | 2026-09-22 |
| #24 | Math Reasoning in LLMs is Organized by Approach, Not Topic | Sajad Goudarzi, Samaneh Zamanifard, Moloud Nasiri +1 | cs.AI | 2026-09-22 |
| #25 | CVaR anchor regression protects against rare shifts | Malte Londschien | stat.ME | 2026-09-22 |
| #26 | GeoRVQ: Decoder-aware geometry for residual-token prediction in physiological signals | Bo Cui, Yaowen Zhang | cs.LG | 2026-09-22 |
| #27 | LEGO: Synergizing Expert GraphRAG and Expert Chain-of-Thought for Legal Reasoning | Qingjing Chen, Junkai Zhang, Shaochun Wang +7 | cs.CL | 2026-09-22 |
| #28 | Sharp Convergence of Wasserstein Gradient Flows for Spectrally Nonnegative Interaction Energies | Zhengjiang Lin, Philippe Rigollet | math.AP | 2026-09-22 |
| #29 | TinyUDE: Solver-Free Universal Differential Equations on Microcontrollers via Lie-Taylor Jet Matching | Pranavanath Balamurali, Hrishi Kamireddy | cs.LG | 2026-09-22 |
| #30 | TwinCheck: Evidence-Grounded Negative-Twin Verification for Stateful Tool Agents | Jiaxuan Dai, Tianyi Huang | cs.AI | 2026-09-22 |
| #31 | PR-Smoother: Simulator-Preserving Non-Gaussian Smoothing for Data Assimilation | Yuta Tarumi | cs.LG | 2026-09-22 |
| #32 | Type-Safe Is Not Error-Free: A Constrained Decision Head Follows the Option Name, Not the Rubric Bound to It | Yu Sun, Junhao Xu | cs.AI | 2026-09-22 |
| #33 | Metrics Failure in LLM-Based Code Vulnerability Repair: An Empirical Study and a Change-Aware Screen | Om Nepal, Sushant Aryal, Oluseyi Olukola +1 | cs.SE | 2026-09-22 |
| #34 | A Spectral Theory of Grokking: Weight Decay induces Feature Learning | Lenz Pracher, Pascal de Jong, Oskar Lieshaus +2 | cs.LG | 2026-09-22 |
| #35 | Marginally Correct Tool Caches Can Reverse Group-Normalized Policy Updates | Shivam Gupta | cs.LG | 2026-09-22 |
| #36 | Greedy Decoding Is Not Precision-Invariant: Cross-Precision Output Divergence in LLM Inference | Gaoyuan Du, Anam Nawaz Khan, Rex Zhou +4 | cs.LG | 2026-09-22 |
| #37 | When Recursive Models Finish Computing | Hare Krishna, Shubham Singh, Stephen Ebert +1 | cs.LG | 2026-09-22 |
| #38 | Reproducible AI Requires Reproducible Randomness | Anthony Bertrand, Tom Schmitt, Engelbert Mephu Nguifo +1 | cs.AI | 2026-09-22 |
| #39 | Target alignment, dilution and forecast selection when cross-sectional forecasts share a common target | Masoud Soleimani | econ.EM | 2026-09-22 |
| #40 | CompKV: Compensation-Aware KV Selection for Long-Context LLM Inference | Zhen Huang, Ruizhe Yao, Danyi Liu +8 | cs.LG | 2026-09-22 |
| #41 | Compiling Sufficient Governance Context from Declared Losses and Reachable States: Exact Observation-Contract Synthesis with Cardinality and Cost Objectives | Gaston Besanson | cs.SE | 2026-09-22 |
| #42 | Theory for groupoid equivariant neural networks: an approach for steerable CNNs on bounded domains | Alberto Ibort, Maria Jimenez-Vazquez, Juan M. Perez-Pardo | cs.LG | 2026-09-22 |
| #43 | CausalLoss-Fin: Attributing Financial-Agent Loss to Decisions and Infrastructure Faults | Abhishek Sharma | cs.AI | 2026-09-22 |
| #44 | Destination Support Restoration for Finite-Set Multimodal Trajectory Prediction | Fengrui Liu, Jiajun Peng, Duo Peng +1 | cs.RO | 2026-09-22 |
| #45 | Calibrating Retrieval Geometry: Reliability-Guided Training-Free Aggregation for Visual Place Recognition | Xin Li, Zhimin Mao, Shang Wang +2 | cs.CV | 2026-09-22 |
| #46 | Toward Responsible AI-Augmented Cyber Defense: Pattern Recognition, Defense-in-Depth, and the Case for Human-AI Collaboration | Mustafa S. Aljumaily, Hayder Kareem Abed, Nawar S. Alseelawi | cs.CR | 2026-09-22 |
| #47 | Neural Approximation by Function Composition: Rigidity and Doubly Exponential Convergence | Wentao Huang, Haizhang Zhang | cs.LG | 2026-09-22 |
| #48 | MemoryAthena: Adaptive Routing over Latent and Generated Memories | Mingyuan Li, Guangsheng Yu, Juyuan Zhang +4 | cs.CL | 2026-09-22 |
| #49 | Optimizing the Score, Losing Sight of the Task: Reward Hacking Across Weights, Selection, and Prompts | Vansh Wahi | cs.AI | 2026-09-22 |
| #50 | Gaussian Flow-Matching Schedules: Implications for Sampling and Training | Arsène Claustre, Hugo Negrel, Claire Boyer +2 | cs.LG | 2026-09-22 |