Hal-Eval: A Universal and Fine-grained Hallucination Evaluation Framework for Large Vision Language Models

arXiv:2402.15721v1 Announce Type: cross
Abstract: Large Vision Language Models exhibit remarkable capabilities but struggle with hallucinations inconsistencies between images and their descriptions. Previous hallucination evaluation studies on LVLMs have identified hallucinations in terms of objects, attributes, and relations but overlooked complex hallucinations that create an entire narrative around a fictional entity. In this paper, we introduce a refined taxonomy of hallucinations, featuring a new category: Event Hallucination. We then utilize advanced LLMs to generate and filter fine grained hallucinatory data …

abstract arxiv capabilities cs.ai cs.cl evaluation fine-grained framework hal hallucination hallucinations images language language models narrative objects relations struggle studies terms type universal vision