Optical Character Recognition2020en
howsumm-method
Dataset from Papers With Code
Metrics:accuracy, cer, wer, f1
rouge-1
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | LexRank (query: method + article + steps titles) | 53.5 | Oct 2021 | |
| 2 | CES (query: method + article + steps titles) | 52.2 | Oct 2021 | |
| 3 | GreedyRel (query: method + article + steps titles) | 48.6 | Oct 2021 | |
| 4 | CES (query: method title) | 48.4 | Oct 2021 | |
| 5 | CES (query: method + article titles) | 48.3 | Oct 2021 | |
| 6 | LexRank (query: method title) | 47.7 | Oct 2021 | |
| 7 | LexRank (query: method + article titles) | 47.1 | Oct 2021 | |
| 8 | GreedyRel (query: method title) | 43.4 | Oct 2021 | |
| 9 | GreedyRel (query: method + article titles) | 42.3 | Oct 2021 |
Related Papers1
HowSumm: A Multi-Document Summarization Dataset Derived from WikiHow Articles
Oct 2021Models: LexRank (query: method + article + steps titles), CES (query: method + article + steps titles), GreedyRel (query: method + article + steps titles) +6 more