Audio-Visual Captioning on f-Lifescene (test)
0.529BLEUSSU (ours)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SSU (ours)2026.03 | 0.529 | 0.5388 | 0.505 | |
| SSU-Dmode=Sound-Source Understanding with Description Only2026.03 | 0.5169 | 0.5025 | 0.4815 | |
| AVC2026.03 | 0.359 | 0.3423 | 0.3618 | |
| PG-Video-LLaVA2026.03 | 0.19 | 0.2164 | 0.2374 |