Multimodal Reasoning on SAT
59.49AccuracyOcean_R1_3B_Instruct
Evaluation Results
| Method | Links | |
|---|---|---|
| Ocean_R1_3B_Instruct2025.10 | 59.49 | |
| RECAPbackbone=Qwen2.5-VL-3B2025.10 | 55.19 | |
| MM-R1-MGT-PerceReason2025.10 | 50.83 | |
| vision-grpo-qwen-2.5-vl-3b2025.10 | 50.57 | |
| Qwen2.5-VL-3Bvariant=MoDoMoDo2025.10 | 49.95 | |
| VLAA-Thinker-3B2025.10 | 49.38 | |
| Qwen2.5-VL-3Bvariant=Uniform2025.10 | 44.55 | |
| Qwen2.5-VL-3Bmode=Base model2025.10 | 43.98 | |
| Qwen2.5-VL-3B-Instruct-GRPO-deepmath2025.10 | 34.7 | |
| Qwen2.5VL-3b-RLCS2025.10 | 24.12 |