Knowledge Base Question Answering on CWQ
51.74Answer F1ChatGPT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ChatGPTSize=175B, Prompting Strategy=Fewshot CoT - Plan2024.02 | 51.74 | 38.6 | 13.14 | 56.61 | |
| ChatGPTSize=175B, Prompting Strategy=Fewshot CoT - SC2024.02 | 51.74 | 40.73 | 11.01 | 52.57 | |
| LLaMA2Size=70B, Prompting Strategy=Fewshot CoT - SC2024.02 | 50.42 | 37 | 13.42 | 58.55 | |
| LLaMA2Size=70B, Prompting Strategy=Fewshot CoT - Plan2024.02 | 50.26 | 37.08 | 13.18 | 57.81 | |
| ChatGPTSize=175B, Prompting Strategy=Fewshot CoT2024.02 | 49.85 | 37.13 | 12.72 | 57.94 | |
| LLaMA2Size=70B, Prompting Strategy=Fewshot CoT2024.02 | 49.8 | 33.98 | 15.82 | 62.23 | |
| VicunaSize=33B, Prompting Strategy=Fewshot CoT - Plan2024.02 | 48.8 | 20.24 | 28.56 | 63.93 | |
| VicunaSize=33B, Prompting Strategy=Fewshot CoT - SC2024.02 | 45.43 | 21.32 | 25.36 | 66.17 | |
| VicunaSize=33B, Prompting Strategy=Fewshot CoT2024.02 | 44.5 | 15.92 | 28.58 | 74.6 | |
| QwenSize=14B, Prompting Strategy=Fewshot CoT - SC2024.02 | 41.01 | 29.26 | 11.75 | 73.21 | |
| QwenSize=14B, Prompting Strategy=Fewshot CoT - Plan2024.02 | 40.86 | 27.97 | 12.02 | 73.68 | |
| MistralSize=7B, Prompting Strategy=Fewshot CoT - SC2024.02 | 40.86 | 30.38 | 10.48 | 65.21 | |
| QwenSize=14B, Prompting Strategy=Fewshot CoT2024.02 | 40.39 | 27.38 | 13.01 | 74.49 | |
| MistralSize=7B, Prompting Strategy=Fewshot CoT - Plan2024.02 | 37.14 | 25.69 | 11.45 | 70.01 | |
| MistralSize=7B, Prompting Strategy=Fewshot CoT2024.02 | 36.45 | 25.18 | 11.27 | 69.86 | |
| QwenSize=7B, Prompting Strategy=Fewshot CoT - Plan2024.02 | 35.35 | 21.57 | 13.86 | 74.74 | |
| QwenSize=7B, Prompting Strategy=Fewshot CoT - SC2024.02 | 34.75 | 23.21 | 15.39 | 74.24 | |
| QwenSize=7B, Prompting Strategy=Fewshot CoT2024.02 | 32.52 | 19.38 | 13.14 | 76.78 |