Temporal Question Answering on TIQ 1.0 (test)
0.491P@1FAITH
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| FAITH2024.02 | 0.491 | 0.603 | 0.752 | |
| Un-FAITHtemporal_pruning=false2024.02 | 0.459 | 0.604 | 0.799 | |
| EXPLAIGNN2024.02 | 0.446 | 0.584 | 0.765 | |
| UNIK-QA2024.02 | 0.425 | 0.48 | 0.54 | |
| INSTRUCTGPTModel=text-davinci-0032024.02 | 0.237 | — | — | |
| GPT-4Model=gpt-42024.02 | 0.236 | — | — | |
| UNIQORN2024.02 | 0.236 | 0.255 | 0.277 | |
| EXAQT2024.02 | 0.232 | 0.378 | 0.587 | |
| TEMPOQRSetting=TempoQR-Hard2024.02 | 0.011 | 0.018 | 0.022 | |
| CRONKGQA2024.02 | 0.006 | 0.011 | 0.014 |