作者位次 #12/15ACL 2026Poster· Natural Language Processing
MTAVG -Bench: A Diagnostic Benchmark for Multi-Talker Dialogue-Centric Audio-Video Generation
Yanghao Zhou · Haitian Li · Rexar Lin · Heyan Huang · Jinxing Zhou · Changsen Yuan · Tian Lan · Ziqin Zhou · Yudong Li · Jiajun Xu · Jingyun Liao · YiMing Cheng · Xuefeng Chen · Xian-Ling Mao · Yousheng Feng
摘要
Recent advances in text-to-audio-video (T2AV) generation have enabled models to synthesize audio-visual videos with multi-participant dialogues. However, existing evaluation benchmarks remain largely designed for human-recorded videos or single-speaker settings. As a result, structural failures i...