Parallel thinking has emerged as a new paradigm for large reasoning models (LRMs) in tackling complex problems. Recent methods leverage Reinforcement Learning (RL) to enhance parallel thinking, aiming to address the limitations in computational resources and effectiveness encountered with supervised fine-tuning. However, most existing studies primarily focus on optimizing the aggregation phase, with limited attention to the path exploration stage. In this paper, we theoretically analyze the optimization of parallel thinking under the Reinforcement Learning with Verifiable Rewards (RLVR) setting, and identify that the mutual information bottleneck among exploration paths fundamentally restricts overall performance. To address this, we propose Outline-Guided Path Exploration (OPE), which explicitly partitions the solution space by generating diverse reasoning outlines prior to parallel path reasoning, thereby reducing information redundancy and improving the diversity of information captured across exploration paths. We implement OPE with an iterative RL strategy that optimizes outline planning and outline-guided reasoning independently. Extensive experiments across multiple challenging mathematical benchmarks demonstrate that OPE effectively improves reasoning performance in different aggregation strategies, enabling LRMs to more reliably discover correct solutions.
Thinking Beyond Videos: Unifying Video Reasoning and Deep Research for Open-World Video Agents
Shandong University(山东大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Beihang University(北京航空航天大学);City University of Hong Kong(香港城市大学);Nanyang Technological University(南洋理工大学);Kuaishou Technology(快手科技);Southern University of Science and Technology(南方科技大学)
Thinking vs. Doing: Agents that Reason by Scaling Test-Time Interaction
Carnegie Mellon University(卡内基梅隆大学);Scribe(Scribe公司);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);University of Toronto(多伦多大学);University of California, Berkeley(加州大学伯克利分校);The AGI Company(AGI公司);New York University(纽约大学)
A Systematic Assessment of OpenAI o1-Preview for Higher Order Thinking in Education
AI4STEM Education Center(AI4STEM教育中心);University of Georgia(佐治亚大学);National GENIUS Center(国家GENIUS中心);School of Computing, University of Georgia(佐治亚大学计算学院);School of Teacher Education, Nanjing Normal University(南京师范大学教师教育学院);National Institute of Education, Singapore(新加坡国家教育学院)
Can OpenAI o1 outperform humans in higher-order cognitive thinking?
AI4STEM Education Center, University of Georgia(佐治亚大学AI4STEM教育中心);National GENIUS Center, University of Georgia(佐治亚大学国家GENIUS中心);School of Computing, University of Georgia(佐治亚大学计算学院);School of Teacher Education, Nanjing Normal University(南京师范大学教师教育学院);School of Social Sciences and Technology, Technical University of Munich(慕尼黑工业大学社会科学与技术学院);Faculty of Psychology, Beijing Normal University(北京师范大学心理学部)
Rethinking External Slow-Thinking: From Snowball Errors to Probability of Correct Reasoning
Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China(中国人民大学人工智能学院);Beijing Key Laboratory of Research on Large Models(北京大型模型研究关键实验室);Engineering Research Center of Next-Generation Intelligent Search(下一代智能搜索工程研究中心);College of Artificial Intelligence, Tianjin University of Science(天津科技大学人工智能学院)
PoTable: Towards Systematic Thinking via Plan-then-Execute Stage Reasoning on Tables
State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China(中国科学技术大学认知智能国家重点实验室);Shenzhen International Graduate School, Tsinghua University(清华大学深圳国际研究生院)
PTFA: An LLM-based Agent that Facilitates Online Consensus Building through Parallel Thinking
Nagoya Institute of Technology(名古屋技术大学);University of Southampton(南安普顿大学);University of Amsterdam(阿姆斯特丹大学);Japan Advanced Institute of Science and Technology(日本先进科学研究院)
Multi-Persona Thinking for Bias Mitigation in Large Language Models
Dept. Computing Science, Alberta Machine Intelligence Institute (Amii) University of Alberta, Canada(计算科学系,阿尔伯塔机器智能研究所(Amii),阿尔伯塔大学,加拿大);Canada CIFAR AI Chair, Amii(加拿大 CIFAR 人工智能主席,Amii)
Ontology-Guided Reverse Thinking Makes Large Language Models Stronger on Knowledge Graph Question Answering
Harbin Institute of Technology(哈尔滨工业大学);Beijing University of Posts and Telecommunications(北京邮电大学);Joint Laboratory of HIT and iFLYTEK(哈工大与科大讯飞联合实验室);University of Science and Technology of China(中国科学技术大学);Pengcheng Laboratory(鹏城实验室)
Entertain, then Cancel: How Thinking Makes Reasoning Models Read Some and Or Literally
The Hong Kong University of Science and Technology(香港科技大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));Peking University(北京大学)
Benchmarking the Thinking Mode of Multimodal Large Language Models in Clinical Tasks
Bytedance(字节跳动);Peking University(北京大学);The Chinese University of Hong Kong(香港中文大学);The University of Hong Kong(香港大学);Mohamed bin Zayed University of Artificial Intelligence(马尔代夫人工智能大学);Stanford University(斯坦福大学);University of Michigan(密歇根大学)