Addressing Uncertainty in LLMs to Enhance Reliability in Generative AI

TL;DR

This paper introduces a novel uncertainty quantification method for Large Language Models using semantic clustering and conformal prediction, improving reliability and efficiency in question answering tasks.

Contribution

It presents a dynamic semantic clustering approach inspired by the Chinese Restaurant Process and integrates it with conformal prediction to better quantify and manage uncertainty in LLM outputs.

Findings

Achieves state-of-the-art uncertainty quantification performance on COQA and TriviaQA.

Produces smaller, more accurate prediction sets while maintaining probabilistic guarantees.

Demonstrates improved reliability and efficiency in LLM-based question answering.

Abstract

In this paper, we present a dynamic semantic clustering approach inspired by the Chinese Restaurant Process, aimed at addressing uncertainty in the inference of Large Language Models (LLMs). We quantify uncertainty of an LLM on a given query by calculating entropy of the generated semantic clusters. Further, we propose leveraging the (negative) likelihood of these clusters as the (non)conformity score within Conformal Prediction framework, allowing the model to predict a set of responses instead of a single output, thereby accounting for uncertainty in its predictions. We demonstrate the effectiveness of our uncertainty quantification (UQ) technique on two well known question answering benchmarks, COQA and TriviaQA, utilizing two LLMs, Llama2 and Mistral. Our approach achieves SOTA performance in UQ, as assessed by metrics such as AUROC, AUARC, and AURAC. The proposed conformal predictor is also shown to produce smaller prediction sets while maintaining the same probabilistic guarantee of including the correct response, in comparison to existing SOTA conformal prediction baseline.