# AdaReasoner: Adaptive Reasoning Enables More Flexible Thinking in Large Language Models

arXiv: https://arxiv.org/abs/2505.17312 · HTML: https://mlsanity.com/paper/2505.17312/

## Key facts

- Cited by 17 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 30 of its 62 references that are in the map.
- First posted on arXiv 22 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, 30 of 30)

- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [Least-to-Most Prompting Enables Complex Reasoning in Large Language Models](https://mlsanity.com/paper/2205.10625/) (2022)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [An Overview of Multi-Task Learning in Deep Neural Networks](https://mlsanity.com/paper/1706.05098/) (2017)
- [Automatic Chain of Thought Prompting in Large Language Models](https://mlsanity.com/paper/2210.03493/) (2022)
- [Eureka: Human-Level Reward Design via Coding Large Language Models](https://mlsanity.com/paper/2310.12931/) (2023)
- [To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning](https://mlsanity.com/paper/2409.12183/) (2024)
- [Reward Design with Language Models](https://mlsanity.com/paper/2303.00001/) (2023)
- [SciBench: Evaluating College-Level Scientific Problem-Solving Abilities of Large Language Models](https://mlsanity.com/paper/2307.10635/) (2023)
- [Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning](https://mlsanity.com/paper/2310.12921/) (2023)
- [The Impact of Reasoning Step Length on Large Language Models](https://mlsanity.com/paper/2401.04925/) (2024)
- [Automatic Prompt Augmentation and Selection with Chain-of-Thought from Labeled Data](https://mlsanity.com/paper/2302.12822/) (2023)
- [Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse](https://mlsanity.com/paper/2410.21333/) (2024)
- [Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability](https://mlsanity.com/paper/2411.19943/) (2024)
- [Towards Mitigating Hallucination in Large Language Models via Self-Reflection](https://mlsanity.com/paper/2310.06271/) (2023)
- [LogiQA: A Challenge Dataset for Machine Reading Comprehension with Logical Reasoning](https://mlsanity.com/paper/2007.08124/) (2020)
- [Large Language Models Can be Lazy Learners: Analyze Shortcuts in In-Context Learning](https://mlsanity.com/paper/2305.17256/) (2023)
- [MedExQA: Medical Question Answering Benchmark with Multiple Explanations](https://mlsanity.com/paper/2406.06331/) (2024)
- [How to think step-by-step: A mechanistic understanding of chain-of-thought reasoning](https://mlsanity.com/paper/2402.18312/) (2024)
- [Metaphor Understanding Challenge Dataset for LLMs](https://mlsanity.com/paper/2403.11810/) (2024)
- [BRIGHTER: BRIdging the Gap in Human-Annotated Textual Emotion Recognition Datasets for 28 Languages](https://mlsanity.com/paper/2502.11926/) (2025)
- [Fewer is More: Boosting LLM Reasoning with Reinforced Context Pruning](https://mlsanity.com/paper/2312.08901/) (2023)
- [A Tutorial on LLM Reasoning: Relevant Methods behind ChatGPT o1](https://mlsanity.com/paper/2502.10867/) (2025)
- [In-context Learning Generalizes, But Not Always Robustly: The Case of Syntax](https://mlsanity.com/paper/2311.07811/) (2023)
- [Reinforcement Learning with Dynamic Boltzmann Softmax Updates](https://mlsanity.com/paper/1903.05926/) (2019)
- [CRoW: Benchmarking Commonsense Reasoning in Real-World Tasks](https://mlsanity.com/paper/2310.15239/) (2023)
- [Larger Language Models Don't Care How You Think: Why Chain-of-Thought Prompting Fails in Subjective Tasks](https://mlsanity.com/paper/2409.06173/) (2024)
- [LLM can Achieve Self-Regulation via Hyperparameter Aware Generation](https://mlsanity.com/paper/2402.11251/) (2024)

## Built on by (most cited first, 17 of 17)

- [Evolving Language Models without Labels: Majority Drives Selection, Novelty Promotes Variation](https://mlsanity.com/paper/2509.15194/) (2025)
- [Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs](https://mlsanity.com/paper/2507.02076/) (2025)
- [Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoning](https://mlsanity.com/paper/2512.24426/) (2025)
- [Dissecting Logical Reasoning in LLMs: A Fine-Grained Evaluation and Supervision Study](https://mlsanity.com/paper/2506.04810/) (2025)
- [From Efficiency to Adaptivity: A Deeper Look at Adaptive Reasoning in Large Language Models](https://mlsanity.com/paper/2511.10788/) (2025)
- [RADAR: Reasoning-Ability and Difficulty-Aware Routing for Reasoning LLMs](https://mlsanity.com/paper/2509.25426/) (2025)
- [Putting on the Thinking Hats: A Survey on Chain of Thought Fine-tuning from the Perspective of Human Reasoning Mechanism](https://mlsanity.com/paper/2510.13170/) (2025)
- [Synthetic Interaction Data for Scalable Personalization in Large Language Models](https://mlsanity.com/paper/2602.12394/) (2026)
- [Distribution-Aware Reward Estimation for Test-Time Reinforcement Learning](https://mlsanity.com/paper/2601.21804/) (2026)
- [Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data](https://mlsanity.com/paper/2604.18493/) (2026)
- [CIMR: Contextualized Iterative Multimodal Reasoning for Robust Instruction Following in LVLMs](https://mlsanity.com/paper/2507.22074/) (2025)
- [ChainV: Atomic Visual Hints Make Multimodal Reasoning Shorter and Better](https://mlsanity.com/paper/2511.17106/) (2025)
- [Thinking Hard, Not Smart: Reasoning Models Fail to Ration Test-Time Compute Across Questions](https://mlsanity.com/paper/2608.07968/) (2026)
- [HyPER: Bridging Exploration and Exploitation for Scalable LLM Reasoning with Hypothesis Path Expansion and Reduction](https://mlsanity.com/paper/2602.06527/) (2026)
- [GTA: Graph Theory Agent and Benchmark for Algorithmic Graph Reasoning with LLMs](https://mlsanity.com/paper/2609.12265/) (2026)
- [FORGE: Form-Optimal Routing of Grounded Evidence for Frozen LLM Agents](https://mlsanity.com/paper/2609.34358/) (2026)
- [Efficiently Learning To Reason or Not to Reason: Root-token Policy Optimization for Adaptive Thinking](https://mlsanity.com/paper/2610.17767/) (2026)

## Abstract (arXiv)

LLMs often need effective configurations, like temperature and reasoning steps, to handle tasks requiring sophisticated reasoning and problem-solving, ranging from joke generation to mathematical reasoning. Existing prompting approaches usually adopt general-purpose, fixed configurations that work 'well enough' across tasks but seldom achieve task-specific optimality. To address this gap, we introduce AdaReasoner, an LLM-agnostic plugin designed for any LLM to automate adaptive reasoning configurations for tasks requiring different types of thinking. AdaReasoner is trained using a reinforcement learning (RL) framework, combining a factorized action space with a targeted exploration strategy, along with a pretrained reward model to optimize the policy model for reasoning configurations with only a few-shot guide. AdaReasoner is backed by theoretical guarantees and experiments of fast convergence and a sublinear policy gap. Across six different LLMs and a variety of reasoning tasks, it consistently outperforms standard baselines, preserves out-of-distribution robustness, and yield gains on knowledge-intensive tasks through tailored prompts.