Estimating LLM Uncertainty with Logits

Ma, Huan, Chen, Jingdong, Wang, Guangyu, Zhang, Changqing

arXiv.org Artificial Intelligence 

In recent years, Large Language Models (LLMs) have seen remarkable advancements and have been extensively integrated across various fields. Despite their progress, LLMs are prone to hallucinations, producing responses that may not be dependable if the models lack sufficient grounding knowledge. To mitigate this issue, methods for estimating uncertainty have been adopted, with a focus on critical tokens as indicators of reliability. Nevertheless, probability-based approaches have shown limitations in assessing token-level reliability due to the erosion of evidence strength information acquired during training. In this paper, we introduce Logits-induced Token Uncertainty (LogU), a novel framework designed to estimate token-specific uncertainty in LLMs in real time, without the need for multiple sampling rounds. By leveraging evidence modeling for the implementation of LogU, we utilize the derived uncertainty measures to steer downstream tasks. Our experimental findings highlight the substantial effectiveness and potential of LogU, marking a significant advancement in addressing the challenge of model hallucinations.