A Survey of Confidence Estimation and Calibration in Large Language Models

TL;DR

This survey reviews recent advances in confidence estimation and calibration techniques for large language models, highlighting challenges, applications, and future research directions to improve their reliability.

Contribution

It provides a comprehensive overview and organization of recent research on LLM confidence estimation and calibration, which was lacking in prior work.

Findings

Summarizes key techniques for confidence estimation in LLMs

Highlights challenges and limitations in current calibration methods

Suggests promising future research directions

Abstract

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.