Mathematical Reasoning on GSM-Symbolic Vary Both
90.22AccuracyAbstRaL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AbstRaLModel=Qwen2.5-Math-7B-Instruct2025.06 | 90.22 | 0.86 | |
| AbstRaLModel=Qwen2.5-7B-Instruct2025.06 | 88.34 | 0.74 | |
| CoT-RLModel=Qwen2.5-Math-7B-Instruct2025.06 | 88.12 | 8.21 | |
| CoT-8SModel=Qwen2.5-Math-7B-Instruct2025.06 | 87.66 | 7.73 | |
| CoT-8SModel=Qwen2.5-7B-Instruct2025.06 | 87.4 | 5 | |
| AbstRaLModel=Llama-3.1-8B-Instruct2025.06 | 86.2 | 0.92 | |
| CoT-8SModel=Llama-3.1-8B-Instruct2025.06 | 82.36 | 5.33 | |
| AbstRaLModel=Mathstral-7B-v0.12025.06 | 82.28 | -1.58 | |
| AbstRaLModel=Llama-3.2-3B-Instruct2025.06 | 79.46 | 0.68 | |
| CoT-RLModel=Qwen2.5-7B-Instruct2025.06 | 79.28 | 6.73 | |
| AbstRaLModel=Qwen2.5-3B-Instruct2025.06 | 78.34 | 0.84 | |
| CoT-RLModel=Mathstral-7B-v0.12025.06 | 76.88 | 1.44 | |
| CoT-8SModel=Mathstral-7B-v0.12025.06 | 76.04 | 4.95 | |
| CoT-RLModel=Llama-3.1-8B-Instruct2025.06 | 75.4 | 9.16 | |
| CoAModel=Qwen2.5-7B-Instruct2025.06 | 74.14 | 2.45 | |
| CoAModel=Mathstral-7B-v0.12025.06 | 74.02 | 1.31 | |
| CoT-8SModel=Qwen2.5-3B-Instruct2025.06 | 72.56 | 11.51 | |
| CoT-8SModel=Llama-3.2-3B-Instruct2025.06 | 71.14 | 15.31 | |
| CoAModel=Qwen2.5-Math-7B-Instruct2025.06 | 69.7 | 1.83 | |
| CoAModel=Llama-3.1-8B-Instruct2025.06 | 69.44 | 3.56 | |
| CoT-RLModel=Llama-3.2-3B-Instruct2025.06 | 68.98 | 13.78 | |
| CoT-RLModel=Qwen2.5-3B-Instruct2025.06 | 68.88 | 12.81 | |
| CoAModel=Llama-3.2-3B-Instruct2025.06 | 67.6 | 0.59 | |
| AbstRaLModel=Qwen2.5-1.5B-Instruct2025.06 | 64.16 | 1.29 | |
| CoAModel=Qwen2.5-3B-Instruct2025.06 | 62.34 | 4.09 | |
| AbstRaLModel=Llama-3.2-1B-Instruct2025.06 | 58.04 | 3.27 | |
| CoT-8SModel=Qwen2.5-1.5B-Instruct2025.06 | 57.52 | 12.85 | |
| CoT-RLModel=Qwen2.5-1.5B-Instruct2025.06 | 51.26 | 8.46 | |
| CoAModel=Qwen2.5-1.5B-Instruct2025.06 | 46.56 | 8.71 | |
| CoT-RLModel=Llama-3.2-1B-Instruct2025.06 | 44.88 | 21.26 | |
| AbstRaLModel=Qwen2.5-0.5B-Instruct2025.06 | 44.56 | -1.27 | |
| CoAModel=Llama-3.2-1B-Instruct2025.06 | 42.42 | 7.78 | |
| CoT-8SModel=Llama-3.2-1B-Instruct2025.06 | 38.9 | 18.96 | |
| CoT-8SModel=Qwen2.5-0.5B-Instruct2025.06 | 33.98 | 10.58 | |
| CoT-RLModel=Qwen2.5-0.5B-Instruct2025.06 | 32.28 | 7.77 | |
| CoAModel=Qwen2.5-0.5B-Instruct2025.06 | 28.72 | 0.97 |