Multi-Sentence Argument Linking
This addresses the challenge of cross-sentence argument linking in NLP, which is incremental as it builds on sentence-level semantic role labeling and coreference resolution.
The authors tackled the problem of linking argument spans across sentences for event role filling by creating a new dataset, RAMS, with 9,124 annotated events, and demonstrated strong performance of their neural model on this and other datasets.
We present a novel document-level model for finding argument spans that fill an event's roles, connecting related ideas in sentence-level semantic role labeling and coreference resolution. Because existing datasets for cross-sentence linking are small, development of our neural model is supported through the creation of a new resource, Roles Across Multiple Sentences (RAMS), which contains 9,124 annotated events across 139 types. We demonstrate strong performance of our model on RAMS and other event-related datasets.