Yiheng Pan

EchoBat: Echo-Vision Enhancement and Echo-Layered Sampling for Video LLMs Hallucination Mitigation

Recent advancements in multimodal large language models (MLLMs) have shown remarkable progress in video understanding. However, video MLLMs (VideoMLLMs) still suffer from hallucinations, generating nonsensical or irrelevant content. This issue partly stems from over-reliance on pre-trained knowledge