Visual Representation Matters: Exploiting Temporal Differences in Video-to-Audio Generation

@misc{chen2026visual,
  title = {Visual Representation Matters: Exploiting Temporal Differences in Video-to-Audio Generation},
  author = {Zehua Chen and Junyou Wang and Yuxuan Jiang and Zhenying Fang and Yusheng Dai and Jianfei Chen and Ziwei Liu and Jun Zhu},
  year = {2026},
  eprint = {2608.04902},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.04902},
}