End-to-End NLG Challenge
End-to-End NLG Challenge (E2E) aims to assess whether recent end-to-end NLG systems can generate more complex output by learning from datasets containing higher lexical richness, syntactic complexity and diverse discourse phenomena.
Source: Evaluating the State-of-the-Art of End-to-End Natural Language Generation: The E2E NLG Challenge
Variants: E2E NLG Challenge, Cleaned E2E NLG Challenge, E2E
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Data-to-Text Generation | self-mem + new data (random) | Self-training from Self-memory in Data-to-text … | 2024-01-19 |
| Data-to-Text Generation | self-mem + new data (fixed) | Self-training from Self-memory in Data-to-text … | 2024-01-19 |
| Table-to-Text Generation | HTLM (fine-tuning) | HTLM: Hyper-Text Pre-Training and Prompting … | 2021-07-14 |
| Table-to-Text Generation | GPT-2-Large (fine-tuning) | HTLM: Hyper-Text Pre-Training and Prompting … | 2021-07-14 |
Recent papers with results on this dataset: