Mitigating Video Hallucinations via Attention-Guided Representation Perturbation
Video large language models (Video LLMs) often generate responses unsupported by the input video. Contrastive decoding (CD) can mitigate these hallucinations, but its effectiveness depends on how the negative branch weakens visual evidence. Existing methods rely on predefined perturbations or masking guided by question...
Zhi-Xi Cai, Ruo-Qi Yuan, Xin-Cheng Yang et al.
· 0 citations