Papers with information aggregation module
MLBiNet: A Cross-Sentence Collective Event Detection Network (2021.acl-long)
Copied to clipboard
| Challenge: | Detecting multiple events from natural language text is a challenge because of the following problems: a) Sentence-level contextual representation and document-level information aggregation are not enough to detect event triggers. |
| Approach: | They propose a multi-layer bidirectional network to capture document-level association of events and semantic information simultaneously. |
| Outcome: | The proposed approach improves performance over the current state-of-the-art approach. |