Addressing Uncertainty in LLMs to Enhance Reliability in Generative AI
This addresses reliability issues in generative AI for users needing trustworthy outputs, though it is incremental as it builds on existing uncertainty quantification and conformal prediction methods.
The paper tackles uncertainty in Large Language Models (LLMs) by proposing a dynamic semantic clustering method based on the Chinese Restaurant Process, which quantifies uncertainty using entropy and integrates with Conformal Prediction to output sets of responses. It achieves state-of-the-art performance on COQA and TriviaQA benchmarks with LLMs like Llama2 and Mistral, producing smaller prediction sets while maintaining probabilistic guarantees.
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.