Visual Question Answering on YesBut
62.9AccuracyOurs
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| OursBackbone=Qwen2.5-VL-3B2026.01 | 62.9 | 83.1 | 9.4 | |
| LADVenue=arXiv'25, Backbone=Qwen2.5-VL-3B2026.01 | 60.9 | 42.3 | 22.7 | |
| CCoTVenue=CVPR'24, Backbone=Qwen2.5-VL-3B2026.01 | 59.7 | 58.9 | 15.4 | |
| ToTVenue=NIPS'23, Backbone=Qwen2.5-VL-3B2026.01 | 59.3 | 82.4 | 29 | |
| GoTVenue=AAAI'24, Backbone=Qwen2.5-VL-3B2026.01 | 59.3 | 79.8 | 18.7 | |
| CoT-SCVenue=ICLR'23, Backbone=Qwen2.5-VL-3B2026.01 | 58.9 | 81.2 | 19.7 | |
| CoTVenue=NIPS'22, Backbone=Qwen2.5-VL-3B2026.01 | 57.7 | 53.5 | 25.8 | |
| DDCoTVenue=NIPS'23, Backbone=Qwen2.5-VL-3B2026.01 | 54.5 | 61 | 24.6 |