PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics

arXiv:2404.04722v1 Announce Type: cross
Abstract: Despite tremendous advancements in large language models (LLMs) over recent years, a notably urgent challenge for their practical deployment is the phenomenon of hallucination, where the model fabricates facts and produces non-factual statements. In response, we propose PoLLMgraph, a Polygraph for LLMs, as an effective model-based white-box detection and forecasting approach. PoLLMgraph distinctly differs from the large body of existing research that concentrates on addressing such challenges through black-box evaluations. In particular, we demonstrate that …

arxiv challenge cs.cl cs.cr cs.se deployment facts hallucination hallucinations language language models large llms non polygraph response state transition urgent