# Emergent Introspective Awareness in Large Language Models

arXiv: https://arxiv.org/abs/2601.01828 · HTML: https://mlsanity.com/paper/2601.01828/

## Key facts

- Cited by 82 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jan 2026.

## Built on by (most cited first, 30 of 82)

- [Latent Introspection: Models Can Detect Prior Concept Injections](https://mlsanity.com/paper/2602.20031/) (2026)
- [A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior](https://mlsanity.com/paper/2602.02639/) (2026)
- [Can LLMs Introspect? A Reality Check](https://mlsanity.com/paper/2605.26242/) (2026)
- [Introspection Adapters: Training LLMs to Report Their Learned Behaviors](https://mlsanity.com/paper/2604.16812/) (2026)
- [Self-Compacting Language Model Agents](https://mlsanity.com/paper/2606.23525/) (2026)
- [Causal Evidence that Language Models use Confidence to Drive Behavior](https://mlsanity.com/paper/2603.22161/) (2026)
- [Agentic Uncertainty Reveals Agentic Overconfidence](https://mlsanity.com/paper/2602.06948/) (2026)
- [Loop as a Bridge: Can Looped Transformers Truly Link Representation Space and Natural Language Outputs?](https://mlsanity.com/paper/2601.10242/) (2026)
- [Reported Confidence in LLMs Tracks Commitment More Than Correctness](https://mlsanity.com/paper/2606.29490/) (2026)
- [Emergent Introspection in AI is Content-Agnostic](https://mlsanity.com/paper/2603.05414/) (2026)
- [Mechanisms of Introspective Awareness](https://mlsanity.com/paper/2603.21396/) (2026)
- [Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent Realignment](https://mlsanity.com/paper/2602.14777/) (2026)
- [RouteHijack: Routing-Aware Attack on Mixture-of-Experts LLMs](https://mlsanity.com/paper/2605.02946/) (2026)
- [From Simulation to Enaction: Post-trained language models recognize and react to their own generations](https://mlsanity.com/paper/2605.25459/) (2026)
- [Covert Influence Between Language Models](https://mlsanity.com/paper/2606.04071/) (2026)
- [Endogenous Resistance to Activation Steering in Language Models](https://mlsanity.com/paper/2602.06941/) (2026)
- [Steering at the Source: Style Modulation Heads for Robust Persona Control](https://mlsanity.com/paper/2603.13249/) (2026)
- [The Consciousness Cluster: Emergent preferences of Models that Claim to be Conscious](https://mlsanity.com/paper/2604.13051/) (2026)
- [Language models recognize dropout and Gaussian noise applied to their activations](https://mlsanity.com/paper/2604.17465/) (2026)
- [Hallucinations Undermine Trust; Metacognition is a Way Forward](https://mlsanity.com/paper/2605.01428/) (2026)
- [Positive Alignment: Artificial Intelligence for Human Flourishing](https://mlsanity.com/paper/2605.10310/) (2026)
- [ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions](https://mlsanity.com/paper/2605.24279/) (2026)
- [Revis: Sparse Latent Steering to Mitigate Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2602.11824/) (2026)
- [Prefill Awareness in Large Language Models](https://mlsanity.com/paper/2606.12747/) (2026)
- [HydraHead: From Head-Level Functional Heterogeneity to Specialized Attention Hybridization](https://mlsanity.com/paper/2606.20097/) (2026)
- [Introspective Coupling: Self-Explanation Training Tracks Behavioral Change Despite Fixed Supervision](https://mlsanity.com/paper/2606.32038/) (2026)
- [Representations of Text and Images Align From Layer One](https://mlsanity.com/paper/2601.08017/) (2026)
- [A Lightweight Modular Framework for Constructing Autonomous Agents Driven by Large Language Models: Design, Implementation, and Applications in AgentForge](https://mlsanity.com/paper/2601.13383/) (2026)
- [No Reliable Evidence of Self-Reported Sentience in Small Large Language Models](https://mlsanity.com/paper/2601.15334/) (2026)
- [Indications of Belief-Guided Agency and Meta-Cognitive Monitoring in Large Language Models](https://mlsanity.com/paper/2602.02467/) (2026)

## Abstract (arXiv)

We investigate whether large language models can introspect on their internal states. It is difficult to answer this question through conversation alone, as genuine introspection cannot be distinguished from confabulations. Here, we address this challenge by injecting representations of known concepts into a model's activations, and measuring the influence of these manipulations on the model's self-reported states. We find that models can, in certain scenarios, notice the presence of injected concepts and accurately identify them. Models demonstrate some ability to recall prior internal representations and distinguish them from raw text inputs. Strikingly, we find that some models can use their ability to recall prior intentions in order to distinguish their own outputs from artificial prefills. In all these experiments, Claude Opus 4 and 4.1, the most capable models we tested, generally demonstrate the greatest introspective awareness; however, trends across models are complex and sensitive to post-training strategies. Finally, we explore whether models can explicitly control their internal representations, finding that models can modulate their activations when instructed or incentivized to "think about" a concept. Overall, our results indicate that current language models possess some functional introspective awareness of their own internal states. We stress that in today's models, this capacity is highly unreliable and context-dependent; however, it may continue to develop with further improvements to model capabilities.