A Survey of Confidence Estimation and Calibration in Large Language Models
This is an incremental survey that organizes existing research to help researchers and practitioners mitigate risks and improve generations in LLMs.
The paper tackles the problem of unreliability in large language models due to factual errors by providing a comprehensive survey of confidence estimation and calibration methods, summarizing recent technical advancements and outlining challenges.
Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of tasks in various domains. Despite their impressive performance, they can be unreliable due to factual errors in their generations. Assessing their confidence and calibrating them across different tasks can help mitigate risks and enable LLMs to produce better generations. There has been a lot of recent research aiming to address this, but there has been no comprehensive overview to organize it and outline the main lessons learned. The present survey aims to bridge this gap. In particular, we outline the challenges and we summarize recent technical advancements for LLM confidence estimation and calibration. We further discuss their applications and suggest promising directions for future work.