Mathematical Reasoning on MATH-Vision mini (test)
42.11ALGVReST
Evaluation Results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VReSTBase Model=Qwen2.5-VL-3B-Instruct2025.06 | 42.11 | 15.79 | 21.05 | 31.58 | 15.79 | 15.79 | 31.58 | 26.32 | 31.58 | 57.89 | 15.79 | 31.58 | 31.58 | 26.32 | 15.79 | 36.84 | 27.96 | |
| Best-of-NBase Model=Qwen2.5-VL-3B-Instruct2025.06 | 31.58 | 10.53 | 21.05 | 26.32 | 5.26 | 10.53 | 36.84 | 10.53 | 26.32 | 42.11 | 15.79 | 26.32 | 15.79 | 26.32 | 10.53 | 26.32 | 21.38 | |
| ToTBase Model=Qwen2.5-VL-3B-Instruct2025.06 | 31.58 | 21.05 | 26.32 | 26.32 | 10.53 | 10.53 | 31.58 | 26.32 | 26.32 | 42.11 | 21.05 | 26.32 | 15.79 | 26.32 | 10.53 | 26.32 | 23.68 | |
| VReST-VoteBase Model=Qwen2.5-VL-3B-Instruct2025.06 | 31.58 | 15.79 | 31.58 | 31.58 | 15.79 | 31.58 | 47.37 | 26.32 | 26.32 | 57.89 | 26.32 | 36.84 | 26.32 | 31.58 | 15.79 | 26.32 | 29.93 | |
| QABase Model=Qwen2.5-VL-3B-Instruct2025.06 | 26.32 | 10.53 | 21.05 | 26.32 | 5.26 | 10.53 | 10.53 | 26.32 | 42.11 | 15.79 | 26.32 | 15.79 | 21.05 | 10.53 | 26.32 | 26.32 | 20.39 | |
| CoTBase Model=Qwen2.5-VL-3B-Instruct2025.06 | 26.32 | 10.53 | 15.79 | 26.32 | 0 | 5.26 | 10.53 | 26.32 | 31.58 | 10.53 | 21.05 | 15.79 | 15.79 | 5.26 | 26.32 | 26.32 | 17.11 | |
| CoT-VoteBase Model=Qwen2.5-VL-3B-Instruct2025.06 | 26.32 | 15.79 | 26.32 | 31.58 | 10.53 | 10.53 | 31.58 | 10.53 | 26.32 | 57.89 | 15.79 | 26.32 | 15.79 | 26.32 | 10.53 | 26.32 | 23.03 | |
| CantorBase Model=Qwen2.5-VL-3B-Instruct2025.06 | 21.05 | 5.26 | 21.05 | 15.79 | 5.26 | 10.53 | 26.32 | 5.26 | 15.79 | 31.58 | 15.79 | 15.79 | 5.26 | 10.53 | 10.53 | 26.32 | 15.13 |