# Unveiling the Latent Directions of Reflection in Large Language Models

arXiv: https://arxiv.org/abs/2508.16989 · HTML: https://mlsanity.com/paper/2508.16989/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Activation steering methods, one of 13 directions in Mechanistic interpretability.
- Builds on 27 of its 40 references that are in the map.
- First posted on arXiv 23 Aug 2025.

## Placed in

- Mechanistic interpretability / [Activation steering methods](https://mlsanity.com/mechinterp/activation-steering-methods/)

## Builds on (most cited first, 27 of 27)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Distributed Representations of Words and Phrases and their Compositionality](https://mlsanity.com/paper/1310.4546/) (2013)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [The Linear Representation Hypothesis and the Geometry of Large Language Models](https://mlsanity.com/paper/2311.03658/) (2023)
- [Towards Reasoning in Large Language Models: A Survey](https://mlsanity.com/paper/2212.10403/) (2022)
- [Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models](https://mlsanity.com/paper/2503.09567/) (2025)
- [Demystifying Long Chain-of-Thought Reasoning in LLMs](https://mlsanity.com/paper/2502.03373/) (2025)
- [Not All Language Model Features Are One-Dimensionally Linear](https://mlsanity.com/paper/2405.14860/) (2024)
- [Fundamental Limitations of Alignment in Large Language Models](https://mlsanity.com/paper/2304.11082/) (2023)
- [On Prompt-Driven Safeguarding for Large Language Models](https://mlsanity.com/paper/2401.18018/) (2024)
- [On the Origins of Linear Representations in Large Language Models](https://mlsanity.com/paper/2403.03867/) (2024)
- [SEAL: Steerable Reasoning Calibration of Large Language Models for Free](https://mlsanity.com/paper/2504.07986/) (2025)
- [Concept Algebra for (Score-Based) Text-Controlled Generative Models](https://mlsanity.com/paper/2302.03693/) (2023)
- [Reinforce LLM Reasoning through Multi-Agent Reflection](https://mlsanity.com/paper/2506.08379/) (2025)
- [Improving Reasoning Performance in Large Language Models via Representation Engineering](https://mlsanity.com/paper/2504.19483/) (2025)
- [Large Language Models Have Intrinsic Meta-Cognition, but Need a Good Lens](https://mlsanity.com/paper/2506.08410/) (2025)
- [MyGO Multiplex CoT: A Method for Self-Reflection in Large Language Models via Double Chain of Thought Thinking](https://mlsanity.com/paper/2501.13117/) (2025)
- [Towards Understanding the Cognitive Habits of Large Reasoning Models](https://mlsanity.com/paper/2506.21571/) (2025)
- [Explaining Intrinsic Moral Self-Correction with Mechanistic Interpretability](https://mlsanity.com/paper/2505.11924/) (2025)

## Built on by (most cited first, 2 of 2)

- [Towards a Mechanistic Understanding of Large Reasoning Models: A Survey of Training, Inference, and Failures](https://mlsanity.com/paper/2601.19928/) (2026)
- [PathCal: State-Aware Reflection-Marker Calibration for Efficient Reasoning](https://mlsanity.com/paper/2605.23074/) (2026)

## Abstract (arXiv)

Reflection, the ability of large language models (LLMs) to evaluate and revise their own reasoning, has been widely used to improve performance on complex reasoning tasks. Yet, most prior works emphasizes designing reflective prompting strategies or reinforcement learning objectives, leaving the inner mechanisms of reflection underexplored. In this paper, we investigate reflection through the lens of latent directions in model activations. We propose a methodology based on activation steering to characterize how instructions with different reflective intentions: no reflection, intrinsic reflection, and triggered reflection. By constructing steering vectors between these reflection levels, we demonstrate that (1) new reflection-inducing instructions can be systematically identified, (2) reflective behavior can be directly enhanced or suppressed through activation interventions, and (3) suppressing reflection is considerably easier than stimulating it. Experiments on GSM8k-adv and Cruxeval-o-adv with Qwen2.5-3B and Gemma3-4B-IT reveal clear stratification across reflection levels, and steering interventions confirm the controllability of reflection. Our findings highlight both opportunities (e.g., reflection-enhancing defenses) and risks (e.g., adversarial inhibition of reflection in jailbreak attacks). This work opens a path toward mechanistic understanding of reflective reasoning in LLMs.