Uncertainty Estimation and Quantification for LLMs: A Simple Supervised Approach
This addresses uncertainty quantification for LLMs, an underexplored area, but the approach appears incremental as it adapts existing supervised methods to LLMs.
The paper tackles uncertainty estimation and calibration for large language models (LLMs) by proposing a supervised approach that uses labeled datasets and hidden activations to estimate uncertainty, showing improved performance across tasks and robust transferability in out-of-distribution settings.
In this paper, we study the problem of uncertainty estimation and calibration for LLMs. We begin by formulating the uncertainty estimation problem, a relevant yet underexplored area in existing literature. We then propose a supervised approach that leverages labeled datasets to estimate the uncertainty in LLMs' responses. Based on the formulation, we illustrate the difference between the uncertainty estimation for LLMs and that for standard ML models and explain why the hidden neurons of the LLMs may contain uncertainty information. Our designed approach demonstrates the benefits of utilizing hidden activations to enhance uncertainty estimation across various tasks and shows robust transferability in out-of-distribution settings. We distinguish the uncertainty estimation task from the uncertainty calibration task and show that better uncertainty estimation leads to better calibration performance. Furthermore, our method is easy to implement and adaptable to different levels of model accessibility including black box, grey box, and white box.