General reasoning represents a long-standing and formidable challenge in artificial intelligence. Recent breakthroughs, exemplified by large language models (LLMs) and chain-of-thought prompting, have achieved considerable success on foundational reasoning tasks. However, this success is heavily contingent upon extensive human-annotated demonstrations, and models' capabilities are still insufficient for more complex problems. Here we show that the reasoning abilities of LLMs can be incentivized through pure reinforcement learning (RL), obviating the need for human-labeled reasoning trajectories. The proposed RL framework facilitates the emergent development of advanced reasoning patterns, such as self-reflection, verification, and dynamic strategy adaptation. Consequently, the trained model achieves superior performance on verifiable tasks such as mathematics, coding competitions, and STEM fields, surpassing its counterparts trained via conventional supervised learning on human demonstrations. Moreover, the emergent reasoning patterns exhibited by these large-scale models can be systematically harnessed to guide and enhance the reasoning capabilities of smaller models.
R1-ShareVL: Incentivizing Reasoning Capability of Multimodal Large Language Models via Share-GRPO
Nanyang Technological University(南洋理工大学);ByteDance(字节跳动);Tsinghua University(清华大学);Beijing University of Posts and Telecommunications(北京邮电大学);The University of Sydney(悉尼大学)
KnowCoder-A1: Incentivizing Agentic Reasoning Capability with Outcome Supervision for KBQA
Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所);State Key Laboratory of AI Safety(人工智能安全国家重点实验室);School of Computer Science, University of Chinese Academy of Sciences(中国科学院大学计算机科学学院)
WebGen-R1: Incentivizing Large Language Models to Generate Functional and Aesthetic Websites with Reinforcement Learning
The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));The Hong Kong University of Science and Technology(香港科技大学);Tongyi Lab, Alibaba Group(阿里云实验室);Electronics and Telecommunications Research Institute(电子电信研究院);Ant Group(蚂蚁集团)
Visual Merit or Linguistic Crutch? A Close Look at DeepSeek-OCR
Chengdu Institute of Computer Applications, CAS(成都计算机应用研究所,中国科学院);University of Chinese Academy of Sciences(中国科学院大学);IMRAM, Tohoku University(东大理工学院IMRAM);China Tower Corporation Limited(中国塔公司);CSIS, Tohoku University(东大理工学院CSIS);National Institute for Materials Science(国家材料科学研究所);Shenzhen Institutes of Advanced Technology, CAS(深圳先进技术研究所,中国科学院);Artificial Intelligence Research Institute, Shenzhen University of Advanced Technology(深圳先进技术大学人工智能研究院)
Is PRM Necessary? Problem-Solving RL Implicitly Induces PRM Capability in LLMs
Huawei Technologies Ltd.(华为技术有限公司);Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation, Department of CSE, SUSTech(广东省脑启发智能计算重点实验室、信息科学部、南方科技大学)
TAROT: Test-driven and Capability-adaptive Curriculum Reinforcement Fine-tuning for Code Generation with Large Language Models
Electronics and Telecommunications Research Institute(电信研究所);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));The Hong Kong University of Science and Technology(香港科技大学);Hugging Face;Ant Group(蚂蚁集团)
Reinforcement Learning for Reasoning in Large Language Models with One Training Example
University of Washington(华盛顿大学);University of Southern California(南加州大学);Microsoft(微软);University of California, Santa Cruz(加州大学圣克鲁兹分校);Georgia Institute of Technology(佐治亚理工学院)
EvoCoT: Overcoming the Exploration Bottleneck in Reinforcement Learning
School of Computer Science, Peking University(北京大学计算机科学学院);College of AI, Tsinghua University(清华大学人工智能学院);Imperial College London(伦敦帝国理工学院);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所)
TL;DR: Too Long, Do Re-weighting for Efficient LLM Reasoning Compression
School of Artificial Intelligence, Chinese Academy of Sciences(中国科学院人工智能学院);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);University of California, Los Angeles(加州大学洛杉矶分校);Tsinghua University(清华大学);Microsoft(微软);Hong Kong University of Science and Technology(香港科技大学);Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州))
VersaTune: An Efficient Data Composition Framework for Training Multi-Capability LLMs
Center for Data Science, Academy for Advanced Interdisciplinary Studies, Peking University(数据科学中心、先进跨学科研究院、北京大学);School of CS & Key Lab of High Confidence Software Technologies (MOE), Peking University(计算机学院及高可信软件技术实验室(教育部));Baichuan Inc.(百川科技公司)