SH2: Self-Highlighted Hesitation Helps You Decode More Truthfully

arXiv:2401.05930v2 Announce Type: replace
Abstract: Large language models (LLMs) demonstrate great performance in text generation. However, LLMs are still suffering from hallucinations. In this work, we propose an inference-time method, Self-Highlighted Hesitation (SH2), to help LLMs decode more truthfully. SH2 is based on a simple fact rooted in information theory that for an LLM, the tokens predicted with lower probabilities are prone to be more informative than others. Our analysis shows that the tokens assigned with lower probabilities by an …

abstract arxiv cs.ai cs.cl decode hallucinations inference information language language models large language large language models llms performance simple text text generation theory type work