Just rephrase it! Uncertainty estimation in closed-source language models via multiple rephrased queries
This addresses the limitation of closed-source LLMs in critical applications where hallucination risks exist, though it is incremental as it builds on existing query-based uncertainty methods.
The paper tackles the problem of unreliable uncertainty estimates in closed-source large language models by using multiple rephrased queries to measure answer similarity, resulting in significant improvements in calibration compared to baselines.
State-of-the-art large language models are sometimes distributed as open-source software but are also increasingly provided as a closed-source service. These closed-source large-language models typically see the widest usage by the public, however, they often do not provide an estimate of their uncertainty when responding to queries. As even the best models are prone to ``hallucinating" false information with high confidence, a lack of a reliable estimate of uncertainty limits the applicability of these models in critical settings. We explore estimating the uncertainty of closed-source LLMs via multiple rephrasings of an original base query. Specifically, we ask the model, multiple rephrased questions, and use the similarity of the answers as an estimate of uncertainty. We diverge from previous work in i) providing rules for rephrasing that are simple to memorize and use in practice ii) proposing a theoretical framework for why multiple rephrased queries obtain calibrated uncertainty estimates. Our method demonstrates significant improvements in the calibration of uncertainty estimates compared to the baseline and provides intuition as to how query strategies should be designed for optimal test calibration.