VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations for Synthetic Videos
Published in NeurIPS 2025, 2025
VideoHallu is a benchmark for evaluating and mitigating hallucinations in MLLMs on synthetic videos from Sora, Veo2, and Kling.
| arXiv | Project Page | Code |
Recommended citation: Zongxia Li*, Xiyang Wu*, Yubin Qin, Hongyang Du, Guangyao Shi, Dinesh Manocha, Tianyi Zhou, Jordan Lee Boyd-Graber. (2025). "VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations for Synthetic Videos." NeurIPS.
Download Paper