Information Exchange on TriviaQA
78.4F1 ScoreDITS-iSFT-DPO
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DITS-iSFT-DPO2025.02 | 78.4 | — | |
| OPTIMA-iSFT-DPO SCoptimization_strategy=iSFT-DPO, self_consistency=true2024.10 | 77.5 | 494.1 | |
| OPTIMA-iSFT-DPOoptimization_strategy=iSFT-DPO2024.10 | 77.1 | 32.5 | |
| Optima-iSFT-DPO2025.02 | 77.1 | — | |
| OPTIMA-iSFT SCoptimization_strategy=iSFT, self_consistency=true2024.10 | 73.7 | 413.8 | |
| OPTIMA-iSFToptimization_strategy=iSFT2024.10 | 71.9 | 51.5 | |
| Optima-iSFT2025.02 | 71.9 | — | |
| OPTIMA-iDPO SCoptimization_strategy=iDPO, self_consistency=true2024.10 | 71.8 | 702.8 | |
| MAD2024.10 | 71 | 408.6 | |
| MAD2025.02 | 71 | — | |
| SCn=82024.10 | 70 | 891.4 | |
| OPTIMA-iDPOoptimization_strategy=iDPO2024.10 | 69.3 | 69.2 | |
| Optima-iDPO2025.02 | 69.3 | — | |
| AutoForm2024.10 | 60.9 | 74 | |
| AutoForm2025.02 | 60.9 | — | |
| CoT2024.10 | 59.8 | 110.3 | |
| CoT2025.02 | 59.8 | — |