Temporal Question Answering on TIME QUESTIONS 1.0 (test)
75.4P@1QUASAR
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| QUASAR# Parameters=8,218 M2024.12 | 75.4 | — | — | |
| Un-FAITHtemporal_pruning=false2024.02 | 57.1 | 64 | 72.4 | |
| UN-FAITH2024.12 | 57.1 | — | — | |
| EXAQT2024.02 | 56.5 | 59.9 | 66.4 | |
| FAITH2024.02 | 53.5 | 58.2 | 63.5 | |
| EXPLAIGNN2024.02 | 52.5 | 58.7 | 67.3 | |
| EXPLAIGNN# Parameters=328 M2024.12 | 52.5 | — | — | |
| TEMPOQRSetting=TempoQR-Hard2024.02 | 43.8 | 46.5 | 48.8 | |
| UNIK-QA2024.02 | 42.4 | 45.3 | 48.6 | |
| UNIK-QA# Parameters=223 M2024.12 | 42.4 | — | — | |
| CONVINSE# Parameters=362 M2024.12 | 42.3 | — | — | |
| CRONKGQA2024.02 | 39.5 | 42.3 | 45 | |
| UNIQORN2024.02 | 33.1 | 40.9 | 48.8 | |
| GPT-4Model=gpt-42024.02 | 30.6 | — | — | |
| GPT-4# Parameters=1,760,000 M2024.12 | 30.6 | — | — | |
| INSTRUCTGPTModel=text-davinci-0032024.02 | 22.4 | — | — | |
| GPT-3# Parameters=175,000 M2024.12 | 22.4 | — | — | |
| LLAMA3 (8B-Instruct)# Parameters=8,030 M2024.12 | 17.8 | — | — |