Multimodal Intent Recognition on MIntRec 2.0
66.99AccuracyQwen-VL-72B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen-VL-72B2026.03 | 66.99 | 66.63 | 67.45 | |
| LLaVA-Video-72B2026.03 | 64.98 | 64.72 | 65.4 | |
| Qwen-2VL-7B2026.03 | 64.19 | 63.31 | 64.39 | |
| MiniCPM-V-2.6-8B2026.03 | 53.58 | 51.91 | 61.66 | |
| Nano-EmoX2026.03 | 47.32 | 47.27 | 51.1 | |
| Nano-EmoX‡2026.03 | 47.13 | 45.38 | 47.83 | |
| GPT-4shots=32026.03 | 42.1 | 43.92 | 53.18 | |
| AffectGPT (s)2026.03 | 38.79 | 37.18 | 41.91 | |
| MobileVLM V2-3B2026.03 | 23.98 | 24.16 | 26.45 | |
| MobileVLM V2-1.7B2026.03 | 21.34 | 22.64 | 25.59 |