# Minimal and Mechanistic Conditions for Behavioral Self-Awareness in LLMs

arXiv: https://arxiv.org/abs/2511.04875 · HTML: https://mlsanity.com/paper/2511.04875/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 10 references that are in the map.
- First posted on arXiv 6 Nov 2025.

## Builds on (most cited first, 9 of 9)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [Looking Inward: Language Models Can Learn About Themselves by Introspection](https://mlsanity.com/paper/2410.13787/) (2024)
- [Tell me about yourself: LLMs are aware of their learned behaviors](https://mlsanity.com/paper/2501.11120/) (2025)
- [One-shot Optimized Steering Vectors Mediate Safety-relevant Behaviors in LLMs](https://mlsanity.com/paper/2502.18862/) (2025)
- [Does It Make Sense to Speak of Introspection in Large Language Models?](https://mlsanity.com/paper/2506.05068/) (2025)
- [Self-Cognition in Large Language Models: An Exploratory Study](https://mlsanity.com/paper/2407.01505/) (2024)
- [Simple Mechanistic Explanations for Out-Of-Context Reasoning](https://mlsanity.com/paper/2507.08218/) (2025)

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

- [Revealing Hidden Model Behaviors with Task-Specific Self-Reports](https://mlsanity.com/paper/2607.03640/) (2026)
- ["As a Language Model...": Chat Template Switches LLM Self-Referential Voice and Activation Steering Reproduces It](https://mlsanity.com/paper/2609.25021/) (2026)

## Abstract (arXiv)

Recent studies have revealed that LLMs can exhibit behavioral self-awareness: the ability to accurately describe or predict their own learned behaviors without explicit supervision. This capability raises safety concerns as it may, for example, allow models to better conceal their true abilities during evaluation. We attempt to characterize the minimal conditions under which such self-awareness emerges, and the mechanistic processes through which it manifests. Through controlled finetuning experiments on instruction-tuned LLMs with low-rank adapters (LoRA), we find: (1) that self-awareness can be reliably induced using a single rank-1 LoRA adapter; (2) that the learned self-aware behavior can be largely captured by a single steering vector in activation space, recovering nearly all of the fine-tune's behavioral effect; and (3) that self-awareness is non-universal and domain-localized, with independent representations across tasks. Together, these findings suggest that behavioral self-awareness emerges as a domain-specific, linear feature that can be easily induced and modulated.