Echoes Over Time: Unlocking Length Generalization in Video-to-Audio Generation ModelsDate: 2026-02-27Fetched: 2026-02-28T01:46:39.629701+00:00AuthorsChristian Simon, Masato Ishii, Wei-Yao Wang, Koichi Saito, Akio Hayakawa, Dongseok Shim, Zhi Zhong, Shuyang Cui, Shusuke Takahashi, Takashi Shibuya, Yuki MitsufujiLinksHFarXivPDF0Abstract中文摘要EnglishMMHNet通过整合分层方法和非因果Mamba实现从视频生成长音频，性能优于现有视频到音频方法。
