Large language models (LLMs) have exhibited remarkable performance on complex reasoning tasks, with reinforcement learning under verifiable rewards (RLVR) emerging as a principled framework for aligning model behavior with reasoning chains. Despite its promise, RLVR remains prohibitively resource-intensive, requiring extensive reward signals and incurring substantial rollout costs during training. In this work, we revisit the fundamental question of data and compute efficiency in RLVR. We first establish a theoretical lower bound on the sample complexity required to unlock reasoning capabilities, and empirically validate that strong performance can be achieved with a surprisingly small number of training instances. To tackle the computational burden, we propose Dynamic One-Shot Policy Refinement (DoPR), an uncertainty-aware RL strategy that dynamically selects a single informative training sample per batch for policy updates, guided by reward volatility and exploration-driven acquisition. DoPR reduces rollout overhead by nearly an order of magnitude while preserving competitive reasoning accuracy, offering a scalable and resource-efficient solution for LLM post-training. This approach offers a practical path toward more efficient and accessible RL-based training for reasoning-intensive LLM applications.
DiZiNER: Disagreement-guided Instruction Refinement via Pilot Annotation Simulation for Zero-shot Named Entity Recognition
Seltasquare Seoul, Korea（首尔韩国Seltasquare）；Seoul National University College of Medicine（首尔国立大学医学院）；Biomedical Research Institute, Seoul National University Hospital（首尔国立大学医院生物医学研究所）
World Models for Policy Refinement in StarCraft II
The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences（认知与决策智能复杂系统重点实验室，自动化研究所，中国科学院）；School of Artificial Intelligence, University of Chinese Academy of Sciences（人工智能学院，中国科学院大学）
Direct Alignment of Language Models via Quality-Aware Self-Refinement
Hong Kong University of Science and Technology（香港科技大学）；LightSpeed Studios, Tencent（腾讯光速工作室）；Centre for Artificial Intelligence and Robotics, Hong Kong Institute of Science & Innovation, CAS（中国科学院香港创新研究院人工智能与机器人中心）
LLM-HyPZ: Hardware Vulnerability Discovery using an LLM-Assisted Hybrid Platform for Zero-Shot Knowledge Extraction and Refinement
Department of Electrical and Computer Engineering, University of Arizona, Tucson, AZ, USA（电气与计算机工程系，亚利桑那大学）；Department of Systems and Industrial Engineering, University of Arizona, Tucson, AZ, USA（系统与工业工程系，亚利桑那大学）；College of Information Science, University of Arizona, Tucson, AZ, USA（信息科学学院，亚利桑那大学）；School of Electrical Engineering and Computer Science, University of North Dakota, Grand Forks, ND, USA（电气工程与计算机科学学院，北达科他大学）；Department of Computer Science, University of California Davis, CA, USA（计算机科学系，加州大学戴维斯分校）
RUFNet: Query-Guided Support Mask Refinement and Uncertainty Fusion based on Hybrid Mamba for Few-Shot Brain Tumor Segmentation
Department of Language Science and Technology, The Hong Kong Polytechnic University（香港理工大学语言科学与技术系）；School of Artificial Intelligence, Chongqing University of Technology（重庆理工大学人工智能学院）；Affiliated Lianyungang Hospital of Xuzhou Medical University（徐州医科大学附属连云港医院）；College of Mechatronics and Control Engineering, Shenzhen University（深圳大学机电与控制工程学院）
FREE: Feature Refinement for Generalized Zero-Shot Learning
Huazhong University of Science and Technology (HUST)（华中科技大学）；Southern University of Science and Technology (SUSTech)（南方科技大学）；Inception Institute of Artificial Intelligence (IIAI)（起源人工智能研究院）
ROAR: Retrieval Opportunity-Aware Refinement for Zero-Shot Time Series Forecasting
College of Computer Science and Technology, Jilin University（吉林大学计算机科学与技术学院）；College of Computing and Data Science, Nanyang Technological University（南洋理工大学计算机与数据科学学院）；City St George’s, University of London（伦敦大学城圣乔治学院）；Systems Research Institute, Polish Academy of Sciences（波兰科学院系统研究所）
PECAN: Leveraging Policy Ensemble for Context-Aware Zero-Shot Human-AI Coordination
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；King’s College London（伦敦国王学院）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；The University of Sydney（悉尼大学）；University College London（伦敦大学学院）
OPDSearch+: On-Policy Distillation with RL Refinement for Search-Augmented Reasoning
University of Chinese Academy of Sciences（中国科学院大学）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Macau（澳门大学）；Wuhan University（武汉大学）；Georgia Institute of Technology（佐治亚理工学院）；Northwestern Polytechnical University（西北工业大学）；University of Hong Kong（香港大学）
CourtGuard: A Model-Agnostic Framework for Zero-Shot Policy Adaptation in LLM Safety
Department of Computer Science, Virginia Tech（弗吉尼亚理工大学计算机科学系）；School of IT（信息技术学院）；Engineering, ADA University（工程学院，ADA大学）；School of Law, ADA University（法学院，ADA大学）
Prompting Decision Transformer for Few-Shot Policy Generalization
Carnegie Mellon University（卡内基梅隆大学）；University of Montreal（蒙特利尔大学）；Mila（米拉）；MIT-IBM Watson AI Lab（MIT-IBM沃森人工智能实验室）；Massachusetts Institute of Technology（麻省理工学院）；UMass Amherst（马萨诸塞大学阿默斯特分校）
