We introduce ThinkTwice, a simple two-phase framework that jointly optimizes LLMs to solve reasoning problems and refine the answers, based on Group Relative Policy Optimization (GRPO). In each pair of training steps, ThinkTwice first optimizes the model on solving reasoning problems, then optimizes it on refining its own solutions to the same problems, using the same binary correctness reward in both phases without correctness signals or critique annotations. Across five mathematical reasoning benchmarks and two model families including Qwen3-4B and Olmo3-7B, ThinkTwice substantially improves both reasoning and refinement performance over competitive online policy optimization baselines. Specifically, on Qwen3-4B, ThinkTwice outperforms GRPO on AIME by 5 percentage points before refinement and by 11.5 points after one self-refinement step, measured by pass@4. Analysis of the training dynamics of ThinkTwice reveals an implicit rectify-then-fortify curriculum: refinement predominantly corrects errors early in training and naturally shifts toward preserving already-correct solutions as the model improves, yielding a more rectified reward signal. Our work establishes joint training of reasoning and self-refinement as a principled and effective methodology for RLVR.
Knowing How to Edit: Reliable Evaluation Signals for Diagnosing and Optimizing Prompts at Query Level
School of Information Sciences, University of Illinois Urbana-Champaign（信息科学学院，伊利诺伊大学厄巴纳-香槟分校）；College of Engineering, Carnegie Mellon University（工程学院，卡内基梅隆大学）
Model selection meets clinical semantics: Optimizing ICD-10-CM prediction via LLM-as-Judge evaluation, redundancy-aware sampling, and section-aware fine-tuning
National Kaohsiung University of Science and Technology（高雄科技大学）；National Health Research Institutes（国家卫生研究院）；Kaohsiung Medical University（高雄医学大学）；Taitung MacKay Memorial Hospital（台东马偕纪念医院）；MacKay Medical University（马偕医科大学）；Kaohsiung Medical University Hospital（高雄医学大学附设医院）
Optimizing Spectral Prediction in MXene-Based Metasurfaces Through Multi-Channel Spectral Refinement and Savitzky-Golay Smoothing
organization= Department of Computer Engineering, College of Computing；Mathematics, King Fahd University of Petroleum \& Minerals , city= Dhahran , postcode= 31261 , country= Saudi Arabia；organization= SDAIA-KFUPM Joint Research Center for Artificial Intelligence, King Fahd University of Petroleum \& Minerals , city= Dhahran , postcode= 31261 , country= Saudi Arabia；organization= School of Physics；State Key Laboratory of Electronics Thin Films；Integrated Devices, University of Electronics
REVOLVE: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization
Hong Kong University of Science and Technology（香港科技大学）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Brown University（布朗大学）；University of Michigan - Ann Arbor（密歇根大学安娜堡分校）；Hong Kong Polytechnic University（香港理工大学）；Intel Lab（英特尔实验室）
RefineX: Learning to Refine Pre-training Data at Scale from Expert-Guided Programs
Institute of Computing Technology（计算技术研究所）；Key Laboratory of Network Data Science and Technology, ICT, CAS（网络数据科学与技术重点实验室）；State Key Laboratory of AI Safety（人工智能安全国家重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）；Alibaba Group（阿里巴巴集团）；University of California Merced（加州大学默塞德分校）；National University of Singapore（新加坡国立大学）
Visual Reasoning and Multi-Agent Approach in Multimodal Large Language Models (MLLMs): Solving TSP and mTSP Combinatorial Challenges
Queensland University of Technology（昆士兰科技大学）；Al-Ahliyya Amman University（安曼阿赫利亚大学）；Budapest University of Technology and Economics（布达佩斯技术与经济大学）；Arab American University（阿拉伯美国大学）；Columbia University（哥伦比亚大学）；Minia University（米尼亚大学）
