# Metacognitive Reuse: Turning Recurring LLM Reasoning Into Concise Behaviors

arXiv: https://arxiv.org/abs/2509.13237 · HTML: https://mlsanity.com/paper/2509.13237/

## Key facts

- Cited by 30 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 21 of its 36 references that are in the map.
- First posted on arXiv 16 Sep 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, 21 of 21)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [Kimi k1.5: Scaling Reinforcement Learning with LLMs](https://mlsanity.com/paper/2501.12599/) (2025)
- [M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation](https://mlsanity.com/paper/2402.03216/) (2024)
- [Tulu 3: Pushing Frontiers in Open Language Model Post-Training](https://mlsanity.com/paper/2411.15124/) (2024)
- [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://mlsanity.com/paper/2212.10509/) (2022)
- [REPLUG: Retrieval-Augmented Black-Box Language Models](https://mlsanity.com/paper/2301.12652/) (2023)
- [SimpleRL-Zoo: Investigating and Taming Zero Reinforcement Learning for Open Base Models in the Wild](https://mlsanity.com/paper/2503.18892/) (2025)
- [LIMO: Less is More for Reasoning](https://mlsanity.com/paper/2502.03387/) (2025)
- [TokenSkip: Controllable Chain-of-Thought Compression in LLMs](https://mlsanity.com/paper/2502.12067/) (2025)
- [Efficient Nearest Neighbor Language Models](https://mlsanity.com/paper/2109.04212/) (2021)
- [On Designing Effective RL Reward at Training Time for LLM Reasoning](https://mlsanity.com/paper/2410.15115/) (2024)
- [Efficiently Scaling LLM Reasoning with Certaindex](https://mlsanity.com/paper/2412.20993/) (2024)
- [AI-Assisted Generation of Difficult Math Questions](https://mlsanity.com/paper/2407.21009/) (2024)
- [Done Is Better than Perfect: Unlocking Efficient Reasoning by Structured Multi-Turn Decomposition](https://mlsanity.com/paper/2505.19788/) (2025)
- [AdaptMI: Adaptive Skill-based In-context Math Instruction for Small Language Models](https://mlsanity.com/paper/2505.00147/) (2025)

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

- [Privileged Information Distillation for Language Models](https://mlsanity.com/paper/2602.04942/) (2026)
- [Self-Distillation Zero: Self-Revision Turns Binary Rewards into Dense Supervision](https://mlsanity.com/paper/2604.12002/) (2026)
- [Rethinking Thinking Tokens: LLMs as Improvement Operators](https://mlsanity.com/paper/2510.01123/) (2025)
- [Vero: An Open RL Recipe for General Visual Reasoning](https://mlsanity.com/paper/2604.04917/) (2026)
- [UMEM: Unified Memory Extraction and Management Framework for Generalizable Memory](https://mlsanity.com/paper/2602.10652/) (2026)
- [Learning Next Action Predictors from Human-Computer Interaction](https://mlsanity.com/paper/2603.05923/) (2026)
- [Measuring Weak-to-Strong Legibility of Reasoning Models](https://mlsanity.com/paper/2603.20508/) (2026)
- [Verifying Meta-Awareness via Predictive Rewards in Reasoning Models](https://mlsanity.com/paper/2510.03259/) (2025)
- [Scaling Small Agents Through Strategy Auctions](https://mlsanity.com/paper/2602.02751/) (2026)
- [Self-Verification Dilemma: Experience-Driven Suppression of Overused Checking in LLM Reasoning](https://mlsanity.com/paper/2602.03485/) (2026)
- [Federation over Text: Insight Sharing for Multi-Agent Reasoning](https://mlsanity.com/paper/2604.16778/) (2026)
- [Training-Free Spectral Fingerprints of Voice Processing in Transformers](https://mlsanity.com/paper/2510.19131/) (2025)
- [Towards Effective Experiential Learning: Dual Guidance for Utilization and Internalization](https://mlsanity.com/paper/2603.24093/) (2026)
- [Neural Garbage Collection: Learning to Forget while Learning to Reason](https://mlsanity.com/paper/2604.18002/) (2026)
- [Zipping the Thought: When and How Compressed Reasoning Data Works in LLM Post-Training](https://mlsanity.com/paper/2605.28008/) (2026)
- [Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs](https://mlsanity.com/paper/2605.07806/) (2026)
- [CORE: Contrastive Reflection Enables Rapid Improvements in Reasoning](https://mlsanity.com/paper/2605.28742/) (2026)
- [Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning](https://mlsanity.com/paper/2606.13680/) (2026)
- [MILES: Modular Instruction Memory with Learnable Selection for Self-Improving LLM Reasoning](https://mlsanity.com/paper/2607.06974/) (2026)
- [Metacognition in LLMs: Foundations, Progress, and Opportunities](https://mlsanity.com/paper/2607.11881/) (2026)
- [Metacognitive Behavioral Tuning of Large Language Models for Multi-Hop Question Answering](https://mlsanity.com/paper/2602.22508/) (2026)
- [Know When to Trust the Skill: Delayed Appraisal and Epistemic Vigilance for Single-Agent LLMs](https://mlsanity.com/paper/2604.16753/) (2026)
- [Characterizing Model-Native Skills](https://mlsanity.com/paper/2604.17614/) (2026)
- [Notes to Self: Can LLMs Benefit from Experiential Abstractions?](https://mlsanity.com/paper/2607.20372/) (2026)
- [ThinkReset: Learnable Intermediate Interface Construction for Bounded-Context Long-Horizon Reasoning](https://mlsanity.com/paper/2607.28642/) (2026)
- [Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning](https://mlsanity.com/paper/2608.21265/) (2026)
- [Characterizing a Configuration Where Inference-Time PRM-Pruned Fragment Grafting Is Inert: Evidence from Three Reasoning LMs](https://mlsanity.com/paper/2610.00047/) (2026)
- [Test-Time Adaptation of Reasoning Strategies with Bayesian Nonparametric Memory](https://mlsanity.com/paper/2610.06516/) (2026)
- [Agent Plasticity: Measuring Self-Improvement Through Experience](https://mlsanity.com/paper/2610.08902/) (2026)
- [Which Skill to Distill? SGUID: Selecting a Compact Skill Bank for Model-Skill Co-Evolution](https://mlsanity.com/paper/2610.12367/) (2026)

## Abstract (arXiv)

Large language models (LLMs) now solve multi-step problems by emitting extended chains of thought. During the process, they often re-derive the same intermediate steps across problems, inflating token usage and latency. This saturation of the context window leaves less capacity for exploration. We study a simple mechanism that converts recurring reasoning fragments into concise, reusable "behaviors" (name + instruction) via the model's own metacognitive analysis of prior traces. These behaviors are stored in a "behavior handbook" which supplies them to the model in-context at inference or distills them into parameters via supervised fine-tuning. This approach achieves improved test-time reasoning across three different settings - 1) Behavior-conditioned inference: Providing the LLM relevant behaviors in-context during reasoning reduces number of reasoning tokens by up to 46% while matching or improving baseline accuracy; 2) Behavior-guided self-improvement: Without any parameter updates, the model improves its own future reasoning by leveraging behaviors from its own past problem solving attempts. This yields up to 10% higher accuracy than a naive critique-and-revise baseline; and 3) Behavior-conditioned SFT: SFT on behavior-conditioned reasoning traces is more effective at converting non-reasoning models into reasoning models as compared to vanilla SFT. Together, these results indicate that turning slow derivations into fast procedural hints enables LLMs to remember how to reason, not just what to conclude.