Mathematical Reasoning on GSM8k (maj1@1)
73.84Accuracy (maj1@1)SIKeD (Adaptive)
Evaluation Results
| Method | Links | |
|---|---|---|
| SIKeD (Adaptive)Model=Gemma 7B2024.10 | 73.84 | |
| SIKeD (All)Model=Gemma 7B2024.10 | 71.42 | |
| PoTModel=Gemma 7B2024.10 | 71.34 | |
| CombinedModel=Gemma 7B2024.10 | 70.74 | |
| L2MModel=Gemma 7B2024.10 | 69.29 | |
| CoTModel=Gemma 7B2024.10 | 67.4 | |
| SIKeD (Adaptive)Model=Qwen 1.5B2024.10 | 64.97 | |
| CombinedModel=Qwen 1.5B2024.10 | 64.44 | |
| PoTModel=Qwen 1.5B2024.10 | 64.22 | |
| SIKeD (All)Model=Qwen 1.5B2024.10 | 64.14 | |
| LLAMA 2Size=70B2023.07 | 56.8 | |
| CoTModel=Qwen 1.5B2024.10 | 55.57 | |
| L2MModel=Qwen 1.5B2024.10 | 54.59 | |
| LLAMA 1Size=65B2023.07 | 50.9 | |
| SIKeD (Adaptive)Model=Gemma 2B2024.10 | 47.23 | |
| SIKeD (All)Model=Gemma 2B2024.10 | 45.26 | |
| PoTModel=Gemma 2B2024.10 | 44.05 | |
| CombinedModel=Gemma 2B2024.10 | 44.05 | |
| SIKeD (Adaptive)Model=Qwen 0.5B2024.10 | 43.14 | |
| SIKeD (All)Model=Qwen 0.5B2024.10 | 43.06 | |
| CombinedModel=Qwen 0.5B2024.10 | 42.38 | |
| LLAMA 2Size=34B2023.07 | 42.2 | |
| PoTModel=Qwen 0.5B2024.10 | 41.62 | |
| L2MModel=Gemma 2B2024.10 | 36.92 | |
| CoTModel=Gemma 2B2024.10 | 36.54 | |
| CoTModel=Qwen 0.5B2024.10 | 36.47 | |
| LLAMA 1Size=33B2023.07 | 35.6 | |
| L2MModel=Qwen 0.5B2024.10 | 33.59 | |
| LLAMA 2Size=13B2023.07 | 28.7 | |
| SIKeD (Adaptive)Model=SmolLM 1.7B2024.10 | 27.98 | |
| SIKeD (All)Model=SmolLM 1.7B2024.10 | 27.75 | |
| CombinedModel=SmolLM 1.7B2024.10 | 24.56 | |
| PoTModel=SmolLM 1.7B2024.10 | 23.73 | |
| FalconSize=40B2023.07 | 19.6 | |
| L2MModel=SmolLM 1.7B2024.10 | 18.73 | |
| LLAMA 1Size=13B2023.07 | 17.8 | |
| CoTModel=SmolLM 1.7B2024.10 | 16.38 | |
| MPTSize=30B2023.07 | 15.2 | |
| LLAMA 2Size=7B2023.07 | 14.6 | |
| LLAMA 1Size=7B2023.07 | 11 | |
| MPTSize=7B2023.07 | 6.8 | |
| FalconSize=7B2023.07 | 6.8 |