ELI5 is a dataset for long-form question answering. It contains 270K complex, diverse questions that require explanatory multi-sentence answers. Web search results are used as evidence documents to answer each question.
ELI5 is also a task in Dodecadialogue.
Source: ELI5
Image Source: https://arxiv.org/pdf/1907.09190v1.pdf
Variants: ELI5
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Open-Domain Question Answering | Fourier Transformer | Fourier Transformer: Fast Long Range … | 2023-05-24 |
| Open-Domain Question Answering | QG | Closed-book Question Generation via Contrastive … | 2022-10-13 |
| Open-Domain Question Answering | Multi-Inrerleave | Improving Conditioning in Context-Aware Sequence … | 2019-11-21 |
| Open-Domain Question Answering | BART | BART: Denoising Sequence-to-Sequence Pre-training for … | 2019-10-29 |
| Open-Domain Question Answering | E-MCA | Using Local Knowledge Graph Construction … | 2019-10-18 |
| Open-Domain Question Answering | Transformer Multitask + LayerDrop | Reducing Transformer Depth on Demand … | 2019-09-25 |
Recent papers with results on this dataset: