Cover Image

Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wang, Yue, Liu, Qiuzhi, Xu, Jiahao, Liang, Tian, Chen, Xingyu, He, Zhiwei, Song, Linfeng, Yu, Dian, Li, Juntao, Zhang, Zhuosheng, Wang, Rui, Tu, Zhaopeng, Mi, Haitao, Yu, Dong
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
by: Chen, Xingyu, et al.
Published: (2024)

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning
by: He, Zhiwei, et al.
Published: (2025)

Dancing with Critiques: Enhancing LLM Reasoning with Stepwise Natural Language Self-Critique
by: Li, Yansi, et al.
Published: (2025)

Teaching LLMs to Refine with Tools
by: Yu, Dian, et al.
Published: (2024)

DeepTheorem: Advancing LLM Reasoning for Theorem Proving Through Natural Language and Reinforcement Learning
by: Zhang, Ziyin, et al.
Published: (2025)

The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models
by: Ji, Ke, et al.
Published: (2025)

Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains
by: Su, Yi, et al.
Published: (2025)

Two Experts Are All You Need for Steering Thinking: Reinforcing Cognitive Effort in MoE Reasoning Models Without Additional Training
by: Wang, Mengru, et al.
Published: (2025)

Don't Get Lost in the Trees: Streamlining LLM Reasoning by Overcoming Tree Search Exploration Pitfalls
by: Wang, Ante, et al.
Published: (2025)

Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing
by: Tian, Ye, et al.
Published: (2024)

Insight Over Sight: Exploring the Vision-Knowledge Conflicts in Multimodal LLMs
by: Liu, Xiaoyuan, et al.
Published: (2024)

RaSA: Rank-Sharing Low-Rank Adaptation
by: He, Zhiwei, et al.
Published: (2025)

Trust, But Verify: A Self-Verification Approach to Reinforcement Learning with Verifiable Rewards
by: Liu, Xiaoyuan, et al.
Published: (2025)

OptimalThinkingBench: Evaluating Over and Underthinking in LLMs
by: Aggarwal, Pranjal, et al.
Published: (2025)

Towards Self-Improvement of LLMs via MCTS: Leveraging Stepwise Knowledge with Curriculum Preference Learning
by: Wang, Xiyao, et al.
Published: (2024)

SIaM: Self-Improving Code-Assisted Mathematical Reasoning of Large Language Models
by: Yu, Dian, et al.
Published: (2024)

Draft Model Knows When to Stop: Self-Verification Speculative Decoding for Long-Form Generation
by: Zhang, Ziyin, et al.
Published: (2024)

CDE: Curiosity-Driven Exploration for Efficient Reinforcement Learning in Large Language Models
by: Dai, Runpeng, et al.
Published: (2025)

DOTS: Learning to Reason Dynamically in LLMs via Optimal Reasoning Trajectories Search
by: Yue, Murong, et al.
Published: (2024)

Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning
by: Zhang, Yuheng, et al.
Published: (2024)

Every Question Has Its Own Value: Reinforcement Learning with Explicit Human Values
by: Yu, Dian, et al.
Published: (2025)

Improving Machine Translation with Human Feedback: An Exploration of Quality Estimation as a Reward Model
by: He, Zhiwei, et al.
Published: (2024)

LiteSearch: Efficacious Tree Search for LLM
by: Wang, Ante, et al.
Published: (2024)

DeepCompress: A Dual Reward Strategy for Dynamically Exploring and Compressing Reasoning Chains
by: Liang, Tian, et al.
Published: (2025)

Can Watermarks Survive Translation? On the Cross-lingual Consistency of Text Watermark for Large Language Models
by: He, Zhiwei, et al.
Published: (2024)

SmartSwitch: Advancing LLM Reasoning by Overcoming Underthinking via Promoting Deeper Thought Exploration
by: Zhang, Xichen, et al.
Published: (2025)

Avoiding Overthinking and Underthinking: Curriculum-Aware Budget Scheduling for LLMs
by: Rahman, Amirul, et al.
Published: (2026)

Scaling Synthetic Data Creation with 1,000,000,000 Personas
by: Ge, Tao, et al.
Published: (2024)

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability
by: Lin, Zicheng, et al.
Published: (2024)

Improving LLM General Preference Alignment via Optimistic Online Mirror Descent
by: Zhang, Yuheng, et al.
Published: (2025)

CLUE: Non-parametric Verification from Experience via Hidden-State Clustering
by: Liang, Zhenwen, et al.
Published: (2025)

Collaborative decoding of critical tokens for boosting factuality of large language models
by: Jin, Lifeng, et al.
Published: (2024)

Between Underthinking and Overthinking: An Empirical Study of Reasoning Length and correctness in LLMs
by: Su, Jinyan, et al.
Published: (2025)

Inconsistent dialogue responses and how to recover from them
by: Zhang, Mian, et al.
Published: (2024)

Measuring Bargaining Abilities of LLMs: A Benchmark and A Buyer-Enhancement Method
by: Xia, Tian, et al.
Published: (2024)

Dual-Uncertainty Guided Policy Learning for Multimodal Reasoning
by: Liu, Rui, et al.
Published: (2025)

The Hunger Game Debate: On the Emergence of Over-Competition in Multi-Agent Systems
by: Ma, Xinbei, et al.
Published: (2025)

Reasoning Like an Economist: Post-Training on Economic Problems Induces Strategic Generalization in LLMs
by: Zhou, Yufa, et al.
Published: (2025)

Evolving Language Models without Labels: Majority Drives Selection, Novelty Promotes Variation
by: Zhou, Yujun, et al.
Published: (2025)

Chain-of-Jailbreak Attack for Image Generation Models via Editing Step by Step
by: Wang, Wenxuan, et al.
Published: (2024)