# AdaptThink: Reasoning Models Can Learn When to Think

arXiv: https://arxiv.org/abs/2505.13417 · HTML: https://mlsanity.com/paper/2505.13417/

## Key facts

- Cited by 147 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 17 of its 28 references that are in the map.
- First posted on arXiv 19 May 2025.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 17 of 17)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Kimi k1.5: Scaling Reinforcement Learning with LLMs](https://mlsanity.com/paper/2501.12599/) (2025)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs](https://mlsanity.com/paper/2412.21187/) (2024)
- [Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models](https://mlsanity.com/paper/2503.16419/) (2025)
- [L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning](https://mlsanity.com/paper/2503.04697/) (2025)
- [Chain of Draft: Thinking Faster by Writing Less](https://mlsanity.com/paper/2502.18600/) (2025)
- [Training Language Models to Reason Efficiently](https://mlsanity.com/paper/2502.04463/) (2025)
- [Reasoning Models Can Be Effective Without Thinking](https://mlsanity.com/paper/2504.09858/) (2025)
- [Imitate, Explore, and Self-Improve: A Reproduction Report on Slow-thinking Reasoning Systems](https://mlsanity.com/paper/2412.09413/) (2024)
- [DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models](https://mlsanity.com/paper/2503.04472/) (2025)
- [Unlocking Efficient Long-to-Short LLM Reasoning with Model Merging](https://mlsanity.com/paper/2503.20641/) (2025)

## Built on by (most cited first, 30 of 147)

- [Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition](https://mlsanity.com/paper/2505.22375/) (2025)
- [AdaThinkDrive: Adaptive Thinking via Reinforcement Learning for Autonomous Driving](https://mlsanity.com/paper/2509.13769/) (2025)
- [Does Thinking More always Help? Mirage of Test-Time Scaling in Reasoning Models](https://mlsanity.com/paper/2506.04210/) (2025)
- [Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Models](https://mlsanity.com/paper/2508.02120/) (2025)
- [OptimalThinkingBench: Evaluating Over and Underthinking in LLMs](https://mlsanity.com/paper/2508.13141/) (2025)
- [Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle](https://mlsanity.com/paper/2509.16679/) (2025)
- [DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learning](https://mlsanity.com/paper/2510.15110/) (2025)
- [Promoting Efficient Reasoning with Verifiable Stepwise Reward](https://mlsanity.com/paper/2508.10293/) (2025)
- [LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving](https://mlsanity.com/paper/2603.01928/) (2026)
- [Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoning](https://mlsanity.com/paper/2512.24426/) (2025)
- [VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice](https://mlsanity.com/paper/2601.05175/) (2026)
- [Audio-Thinker: Guiding Audio Language Model When and How to Think via Reinforcement Learning](https://mlsanity.com/paper/2508.08039/) (2025)
- [Does Your Reasoning Model Implicitly Know When to Stop Thinking?](https://mlsanity.com/paper/2602.08354/) (2026)
- [Dynamic Model Routing and Cascading for Efficient LLM Inference: A Survey](https://mlsanity.com/paper/2603.04445/) (2026)
- [The Price of a Second Thought: On the Evaluation of Reasoning Efficiency in Large Language Models](https://mlsanity.com/paper/2505.22017/) (2025)
- [Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey](https://mlsanity.com/paper/2507.09662/) (2025)
- [R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning](https://mlsanity.com/paper/2508.21113/) (2025)
- [CodeDance: A Dynamic Tool-integrated MLLM for Executable Visual Reasoning](https://mlsanity.com/paper/2512.17312/) (2025)
- [Do Thinking Tokens Help or Trap? Towards More Efficient Large Reasoning Model](https://mlsanity.com/paper/2506.23840/) (2025)
- [Squeeze the Soaked Sponge: Efficient Off-policy Reinforcement Finetuning for Large Language Model](https://mlsanity.com/paper/2507.06892/) (2025)
- [NeuReasoner: Towards Explainable, Controllable, and Unified Reasoning via Mixture-of-Neurons](https://mlsanity.com/paper/2604.02972/) (2026)
- [Aware First, Think Less: Dynamic Boundary Self-Awareness Drives Extreme Reasoning Efficiency in Large Language Models](https://mlsanity.com/paper/2508.11582/) (2025)
- [Self-Compression of Chain-of-Thought via Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2601.21919/) (2026)
- [KAT-V1: Kwai-AutoThink Technical Report](https://mlsanity.com/paper/2507.08297/) (2025)
- [LAPO: Internalizing Reasoning Efficiency via Length-Adaptive Policy Optimization](https://mlsanity.com/paper/2507.15758/) (2025)
- [ThinkDial: An Open Recipe for Controlling Reasoning Effort in Large Language Models](https://mlsanity.com/paper/2508.18773/) (2025)
- [Thinker: Learning to Think Fast and Slow](https://mlsanity.com/paper/2505.21097/) (2025)
- [ConciseHint: Boosting Efficient Reasoning via Continuous Concise Hints during Generation](https://mlsanity.com/paper/2506.18810/) (2025)
- [Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking](https://mlsanity.com/paper/2509.23392/) (2025)
- [DeepPrune: Parallel Scaling without Inter-trace Redundancy](https://mlsanity.com/paper/2510.08483/) (2025)

## Abstract (arXiv)

Recently, large reasoning models have achieved impressive performance on various tasks by employing human-like deep thinking. However, the lengthy thinking process substantially increases inference overhead, making efficiency a critical bottleneck. In this work, we first demonstrate that NoThinking, which prompts the reasoning model to skip thinking and directly generate the final solution, is a better choice for relatively simple tasks in terms of both performance and efficiency. Motivated by this, we propose AdaptThink, a novel RL algorithm to teach reasoning models to choose the optimal thinking mode adaptively based on problem difficulty. Specifically, AdaptThink features two core components: (1) a constrained optimization objective that encourages the model to choose NoThinking while maintaining the overall performance; (2) an importance sampling strategy that balances Thinking and NoThinking samples during on-policy training, thereby enabling cold start and allowing the model to explore and exploit both thinking modes throughout the training process. Our experiments indicate that AdaptThink significantly reduces the inference costs while further enhancing performance. Notably, on three math datasets, AdaptThink reduces the average response length of DeepSeek-R1-Distill-Qwen-1.5B by 53% and improves its accuracy by 2.4%, highlighting the promise of adaptive thinking-mode selection for optimizing the balance between reasoning quality and efficiency. Our codes and models are available at https://github.com/THU-KEG/AdaptThink.