| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Long Dependency Question Answering | LooGLE | Retrieval40 | 21 | |
| Single-hop Question Answering | Loogle SD | Score45.1 | 17 | |
| Question Answering | LooGLE Long Dependency QA | BLEU-10.0942 | 12 | |
| Summarization | LooGLE ArXiv Paper Summarization | BLEU-129.15 | 11 | |
| Reasoning | LooGLE | Reasoning Accuracy57 | 10 | |
| Question Answering | LooGLE | QA Accuracy27 | 10 | |
| Long-Context Question Answering | LooGLE | EM66.3 | 6 | |
| Question Answering | LooGLE | Short QA Score86.02 | 5 | |
| Multi-hop Question Answering | LooGLE CR 16k | Score19.78 | 5 | |
| Multi-hop Question Answering | LooGLE-MR 16k | Score15.1 | 5 | |
| Single-hop Question Answering | LooGLE-SD 16k | Score45.1 | 5 | |
| Long-context question-answering | LooGLE (test) | ShortQA Score54.65 | 2 |