Multimodal Reasoning on Real-world audio-visual video (evaluation)
18.55BLEU-1MAviS-Chat
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MAviS-Chat2026.03 | 18.55 | 19.42 | 19.04 | 85.02 | 51.18 | 39.3 | |
| GPT-4o-mini2026.03 | 10.27 | 11.53 | 14.49 | 83.24 | 51.06 | 31.06 |
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MAviS-Chat2026.03 | 18.55 | 19.42 | 19.04 | 85.02 | 51.18 | 39.3 | |
| GPT-4o-mini2026.03 | 10.27 | 11.53 | 14.49 | 83.24 | 51.06 | 31.06 |