Recall of Human QA on INFOLOSSQA (test)
0.281Recall (Full)GPT-4
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-42024.01 | 0.281 | 0.034 | 0.684 | |
| NLI Pipeline2024.01 | 0.229 | 0.2 | 0.572 | |
| Llama2-70B2024.01 | 0.162 | 0.044 | 0.793 | |
| Mistral-7B2024.01 | 0.142 | 0.041 | 0.818 |
Loading the SOTA2 catalog…