Evaluating true metacognition in Large Language Models (LLMs) is difficult due to biases and heuristics. This paper presents a framework to measure and enhance LLM metacognition while controlling for these biases. A measurement method using the dtype2′ metric is established to isolate metacognitive ability. The Evolution Strategy for Metacognitive Alignment (ESMA) is proposed, demonstrating robust generalization across unseen datasets, languages, and newly acquired knowledge. Finally, parameter analysis reveals that these improvements are driven by a sparse set of parameters, offering new pathways for targeted metacognitive optimization.
@article{arxiv.2602.02605,
title = {Fine-Tuning Language Models to Know What They Know},
author = {Sangjun Park and Elliot Meyerson and Xin Qiu and Risto Miikkulainen},
journal= {arXiv preprint arXiv:2602.02605},
year = {2026}
}