Multi-hop Reasoning on MuSiQue (F1)
73.3F1 ScoreTOTAL
Evaluation Results
| Method | Links | |
|---|---|---|
| TOTALModel=Claude, Prompting Method=TOTAL2025.10 | 73.3 | |
| TOTALModel=Gemini, Prompting Method=TOTAL2025.10 | 72.86 | |
| CIC + COTModel=Gemini, Prompting Method=CIC + COT2025.10 | 67.17 | |
| CICModel=Gemini, Prompting Method=CIC2025.10 | 66.54 | |
| TOTALModel=GPT, Prompting Method=TOTAL2025.10 | 66.38 | |
| CIC + COTModel=GPT, Prompting Method=CIC + COT2025.10 | 65.11 | |
| CIC + COTModel=Claude, Prompting Method=CIC + COT2025.10 | 65.07 | |
| CICModel=Claude, Prompting Method=CIC2025.10 | 63.87 | |
| CICModel=GPT, Prompting Method=CIC2025.10 | 63.79 | |
| NAÏVEModel=GPT, Prompting Method=NAÏVE2025.10 | 32.43 | |
| COTModel=GPT, Prompting Method=COT2025.10 | 32.39 | |
| COTModel=Claude, Prompting Method=COT2025.10 | 28.1 | |
| NAÏVEModel=Claude, Prompting Method=NAÏVE2025.10 | 27.57 | |
| NAÏVEModel=Gemini, Prompting Method=NAÏVE2025.10 | 25.48 | |
| COTModel=Gemini, Prompting Method=COT2025.10 | 23.03 |