# Search for Truth from Reasoning: A Dynamic Representation Editing Framework for Steering LLM Trajectories

arXiv: https://arxiv.org/abs/2606.28589 · HTML: https://mlsanity.com/paper/2606.28589/

## Key facts

- Cited by 1 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 24 of its 48 references that are in the map.
- First posted on arXiv 26 Jun 2026.

## 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, 24 of 24)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning](https://mlsanity.com/paper/2506.01939/) (2025)
- [GSM-Plus: A Comprehensive Benchmark for Evaluating the Robustness of LLMs as Mathematical Problem Solvers](https://mlsanity.com/paper/2402.19255/) (2024)
- [Multi-Step Reasoning with Large Language Models, a Survey](https://mlsanity.com/paper/2407.11511/) (2024)
- [Thought Anchors: Which LLM Reasoning Steps Matter?](https://mlsanity.com/paper/2506.19143/) (2025)
- [Rethinking Entropy Interventions in RLVR: An Entropy Change Perspective](https://mlsanity.com/paper/2510.10150/) (2025)
- [Understanding Aha Moments: from External Observations to Internal Mechanisms](https://mlsanity.com/paper/2504.02956/) (2025)
- [Entropy Law: The Story Behind Data Compression and LLM Performance](https://mlsanity.com/paper/2407.06645/) (2024)
- [Wait, Wait, Wait... Why Do Reasoning Models Loop?](https://mlsanity.com/paper/2512.12895/) (2025)
- [ProMed: Shapley Information Gain Guided Reinforcement Learning for Proactive Medical LLMs](https://mlsanity.com/paper/2508.13514/) (2025)
- [DRESSing Up LLM: Efficient Stylized Question-Answering via Style Subspace Editing](https://mlsanity.com/paper/2501.14371/) (2025)
- [Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editing](https://mlsanity.com/paper/2510.01243/) (2025)
- [LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models](https://mlsanity.com/paper/2504.02327/) (2025)
- [ADEPT: Continual Pretraining via Adaptive Expansion and Dynamic Decoupled Tuning](https://mlsanity.com/paper/2510.10071/) (2025)
- [Balancing Stylization and Truth via Disentangled Representation Steering](https://mlsanity.com/paper/2508.04530/) (2025)
- [GraphWalker: Patient Analogy Meets Information Gain for Clinical Reasoning with Large Language Models](https://mlsanity.com/paper/2604.06684/) (2026)

## Built on by (most cited first, 1 of 1)

- [The Imitation Game: When LLMs Learn to Reason Like Programs via Code-Centric Reasoning Data Synthesis](https://mlsanity.com/paper/2609.16076/) (2026)

## Abstract (arXiv)

Current approaches to enhance Large Language Model (LLM) reasoning, such as Chain-of-Thought and "Wait" prompts, primarily encourage models to think more, yet often fail to guide them toward Truth. While Representation Editing (RepE) offers a intrinsic control, its application to dynamic reasoning trajectories remains underexplored. In this work, we bridge this gap by investigating the geometry of truth within unfolding reasoning chains. We uncover three critical insights: (1) Truth is encoded at the sentence level and is entangled with latent reasoning patterns; (2) Effective intervention follows an Uncertainty Principle and a Decay Effect, requiring localization to early, high-entropy forks; (3) Naive steering vectors suffer from noise, risking collateral damage to correct trajectories. Based on these findings, we propose DynaSteer, a dynamic RepE framework. DynaSteer employs pattern clustering to disentangle reasoning manifolds and utilizes Fisher-LDA to project purified truth. By dynamically monitoring lookahead entropy, it selectively steers and rolls back trajectories only when necessary. Comprehensive experimental results on several MATH benchmark verify the effectiveness of DynaSteer, and experiments on out-of-domain coding tasks further confirm its generalization ability. Our code is publicly available at https://github.com/tianlwang/DynaSteer.