作者位次 #7/10NeurIPS 2026Poster
AVOC: Enhancing Hour-Level Audio-Video Understanding in Omni-Modal LLMs via Retrieval-Inspired Token Compression
Yijing Chen · Wenhui Tan · Xiaoyi Yu · Yuyue Wang · Xin Cheng · Kaisi Guan · Hao Jiang · Xiangyang Li · Guojie Zhu · Ruihua Song