# NEAT: Neuron-Based Early Exit for Large Reasoning Models

arXiv: https://arxiv.org/abs/2602.02010 · HTML: https://mlsanity.com/paper/2602.02010/

## Key facts

- Cited by 3 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 35 references that are in the map.
- First posted on arXiv 2 Feb 2026.

## Placed in

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

## Builds on (most cited first, 21 of 21)

- [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)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models](https://mlsanity.com/paper/2503.16419/) (2025)
- [Dynamic Early Exit in Reasoning Models](https://mlsanity.com/paper/2504.15895/) (2025)
- [Reasoning Models Can Be Effective Without Thinking](https://mlsanity.com/paper/2504.09858/) (2025)
- [When More is Less: Understanding Chain-of-Thought Length in LLMs](https://mlsanity.com/paper/2502.07266/) (2025)
- [Reasoning Models Know When They're Right: Probing Hidden States for Self-Verification](https://mlsanity.com/paper/2504.05419/) (2025)
- [AdaptThink: Reasoning Models Can Learn When to Think](https://mlsanity.com/paper/2505.13417/) (2025)
- [AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learning](https://mlsanity.com/paper/2505.11896/) (2025)
- [Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency](https://mlsanity.com/paper/2506.08343/) (2025)
- [Answer Convergence as a Signal for Early Stopping in Reasoning](https://mlsanity.com/paper/2506.02536/) (2025)
- [Interpreting Arithmetic Mechanism in Large Language Models through Comparative Neuron Analysis](https://mlsanity.com/paper/2409.14144/) (2024)
- [Let LRMs Break Free from Overthinking via Self-Braking Tuning](https://mlsanity.com/paper/2505.14604/) (2025)
- [Efficient Reasoning for Large Reasoning Language Models via Certainty-Guided Reflection Suppression](https://mlsanity.com/paper/2508.05337/) (2025)
- [Back Attention: Understanding and Enhancing Multi-Hop Reasoning in Large Language Models](https://mlsanity.com/paper/2502.10835/) (2025)
- [Overclocking LLM Reasoning: Monitoring and Controlling Thinking Path Lengths in LLMs](https://mlsanity.com/paper/2506.07240/) (2025)
- [An Investigation of Neuron Activation as a Unified Lens to Explain Chain-of-Thought Eliciting Arithmetic Reasoning of LLMs](https://mlsanity.com/paper/2406.12288/) (2024)
- [Language Arithmetics: Towards Systematic Language Neuron Identification and Manipulation](https://mlsanity.com/paper/2507.22608/) (2025)
- [Activation Control for Efficiently Eliciting Long Chain-of-thought Ability of Language Models](https://mlsanity.com/paper/2505.17697/) (2025)
- [Locate-then-Merge: Neuron-Level Parameter Fusion for Mitigating Catastrophic Forgetting in Multimodal LLMs](https://mlsanity.com/paper/2505.16703/) (2025)

## Built on by (most cited first, 3 of 3)

- [Stop When Reasoning Converges: Semantic-Preserving Early Exit for Reasoning Models](https://mlsanity.com/paper/2605.17672/) (2026)
- [Two-dimensional early exit optimisation of LLM inference](https://mlsanity.com/paper/2604.18592/) (2026)
- [How Reasoning Shapes Social Bias in LLM-Generated Code?](https://mlsanity.com/paper/2608.06829/) (2026)

## Abstract (arXiv)

Large Reasoning Models (LRMs) often suffer from overthinking, a phenomenon in which redundant reasoning steps are generated after a correct solution has already been reached. Existing early reasoning exit methods primarily rely on output-level heuristics or trained probing models to skip redundant reasoning steps, thereby mitigating overthinking. However, these approaches typically require additional rollout computation or externally labeled datasets. In this paper, we propose NEAT, a Neuron-based Early reAsoning exiT framework that monitors neuron-level activation dynamics to enable training-free early exits, without introducing additional test-time computation. NEAT identifies exit-associated neurons and tracks their activation patterns during reasoning to dynamically trigger early exit or suppress reflection, thereby reducing unnecessary reasoning while preserving solution quality. Experiments on four reasoning benchmarks across six models with different scales and architectures show that, for each model, NEAT achieves an average token reduction of 22% to 28% when averaged over the four benchmarks, while maintaining accuracy.