THRONE: An Object-based Hallucination Benchmark for the Free-form Generations of Large Vision-Language Models

arXiv:2405.05256v1 Announce Type: cross
Abstract: Mitigating hallucinations in large vision-language models (LVLMs) remains an open problem. Recent benchmarks do not address hallucinations in open-ended free-form responses, which we term "Type I hallucinations". Instead, they focus on hallucinations responding to very specific question formats -- typically a multiple-choice response regarding a particular object or attribute -- which we term "Type II hallucinations". Additionally, such benchmarks often require external API calls to models which are subject to change. In practice, we observe …

abstract arxiv benchmark benchmarks cs.ai cs.cv cs.lg focus form free hallucination hallucinations language language models multiple object question responses type vision vision-language vision-language models