| Paper | Code | ROUGE-1 | ROUGE-2 | ROUGE-SU4 | ROUGE-L | ModelName | ReleaseDate |
|---|---|---|---|---|---|---|---|
| PRIMERA: Pyramid-based Masked Sentence Pre-training for Multi-document Summarization | ✓ Link | 49.9 | 21.1 | 25.9 | PRIMER | 2021-10-16 | |
| LongT5: Efficient Text-To-Text Transformer for Long Sequences | ✓ Link | 48.17 | 19.43 | 24.94 | LongT5 | 2021-12-15 | |
| Multi-Document Summarization withDeterminantal Point Process Attention | 45.84 | 15.94 | 19.19 | CTF+DPP | 2021-07-13 | ||
| Bottom-Up Abstractive Summarization | ✓ Link | 43.57 | 14.03 | 17.37 | CopyTransformer | 2018-08-31 | |
| Multi-News: a Large-Scale Multi-Document Summarization Dataset and Abstractive Hierarchical Model | ✓ Link | 43.47 | 14.89 | 17.41 | Hi-MAP | 2019-06-04 | |
| Bottom-Up Abstractive Summarization | ✓ Link | 42.80 | 14.19 | 16.75 | PG-BRNN | 2018-08-31 |