EMNLP Findings2025
When to Continue Thinking: Adaptive Thinking Mode Switching for Efficient Reasoning
Xiaoyun Zhang, Jingqing Ruan, Xing Ma, Yawen Zhu, Haodong Zhao, Hao Li, Jiansong Chen, Ke Zeng, Xunliang Cai
Abstract
Large reasoning models (LRMs) achieve remarkable performance via long reasoning chains, but often incur excessive computational overhead due to redundant reasoning, especially on simple tasks. In this work, we systematically quantify the upper bounds of LRMs under both Long-Thinking and No-Thinking modes, and uncover the phenomenon of "Internal Self-Recovery Mechanism" where models implicitly supplement reasoning during answer generation. Building on this insight, we propose Adaptive Self-Recovery Reasoning (ASRR), a framework that suppresses unnecessary reasoning and enables implicit recovery. By introducing accuracy-aware length reward regulation, ASRR adaptively allocates reasoning effort according to problem difficulty, achieving high efficiency with negligible performance sacrifice. Experiments across multiple benchmarks and models show that, compared with GRPO, ASRR reduces reasoning budget by up to 32.5% (1.5B) and 25.7% (7B) with minimal accuracy loss (1.2% and 0.6% pass@1), and significantly boosts harmless rates on safety benchmarks (up to +21.7%). Our results highlight the potential of ASRR for enabling efficient, adaptive, and safer reasoning in LRMs.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
12 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- PaTaRM: Bridging Pairwise and Pointwise Signals via Preference-Aware Task-Adaptive Reward Modeling cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- Tree of Thoughts: Deliberate Problem Solving with Large Language Models cites Least-to-Most Prompting Enables Complex Reasoning in Large Language Models
- Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model? cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- Does Thinking More Always Help? Mirage of Test-Time Scaling in Reasoning Models cites Let's Verify Step by Step
Cited by top-tier papers12
- Does Thinking More Always Help? Mirage of Test-Time Scaling in Reasoning ModelsSoumya Suvra Ghosal, Souradip Chakraborty, Avinash Reddy, Yifu Lu et al.NeurIPS 2025 · 43 citations
- CodeDance: A Dynamic Tool-integrated MLLM for Executable Visual ReasoningQi Song, Honglin Li, Yingchen Yu, Haoyi Zhou et al.CVPR 2026 · 16 citations
- Test-Time Scaling in Diffusion LLMS via Hidden Semi-Autoregressive ExpertsJihoon Lee, Hoyeon Moon, Kevin Zhai, Arun Kumar Chithanar et al.ICLR 2026 · 7 citations
- AdaNav: Adaptive Reasoning with Uncertainty for Vision-Language NavigationXin Ding, Jianyu Wei, Yifan Yang, Shiqi Jiang et al.ICML 2026 · 6 citations
- PaTaRM: Bridging Pairwise and Pointwise Signals via Preference-Aware Task-Adaptive Reward ModelingAi Jian, Jingqing Ruan, Xing Ma, Dailin Li et al.ACL 2026 · 5 citations
Builds on13
- Direct Preference Optimization: Your Language Model is Secretly a Reward ModelRafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning et al.NeurIPS 2023 · 10,924 citations
- Tree of Thoughts: Deliberate Problem Solving with Large Language ModelsShunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran et al.NeurIPS 2023 · 5,068 citations
- Let's Verify Step by StepHunter Lightman, Vineet Kosaraju, Yuri Burda, Harrison Edwards et al.ICLR 2024 · 3,045 citations
- Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?Zhiqi Chen, Rui Lu, Andrew Zhao, Zhaokai Wang et al.NeurIPS 2025 · 1,109 citations
- Least-to-Most Prompting Enables Complex Reasoning in Large Language ModelsDenny Zhou, Nathanael Schärli, Le Hou, Jason Wei et al.ICLR 2023 · 318 citations
Related papers
- Incentivizing Dual Process Thinking for Efficient Large Language Model ReasoningXiaoxue Cheng, Junyi Li, Zhenduo Zhang, Xinyu Tang et al.NeurIPS 2025 · 25 citations
- SAT: Balancing Reasoning Accuracy and Efficiency with Stepwise Adaptive ThinkingWeiyang Huang, Xuefeng Bai, Kehai Chen, Xinyang Chen et al.ACL 2026 · 3 citations
- When Simple Problems Wear Complex Costumes: Improving Efficiency in LRM's Adaptive ReasoningJunnan Ren, Yan Zhang, Qian Chen, Yunhang Shen et al.ICML 2026
- Let LRMs Break Free from Overthinking via Self-Braking TuningHaoran Zhao, Yuchen Yan, Yongliang Shen, Haolei Xu et al.NeurIPS 2025 · 31 citations
- SuCo: Sufficiency-guided Continuous Adaptive ReasoningJiahao Wang, Bingyu Liang, Chenhao Hu, Longhui Zhang et al.ICML 2026
