Knowledge-rich Bridging Anaphora Resolution

Bridging anaphora resolution, the task of identifying anaphoric noun phrases (i.e., bridging anaphors) and linking them to their antecedents in a document, is crucial for machine comprehension of the relations between discourse entities for various downstream applications, such as question answering and dialogue systems. Bridging resolution is arguably less studied but more challenging than entity coreference resolution, the task of determining which entity mentions refer to the same entity in the real world. Specifically, while linguistic constraints on coreference exist at the grammatical (e.g., gender and number agreement), syntactic (e.g., c-command), and semantic (e.g., semantic type agreement) levels that can be used to filter candidate antecedents, such constraints are largely absent for bridging resolution. For instance, a singular bridging anaphor (e.g., ”the book”) can refer to a plural antecedent (e.g., ”books”), and bridging relations can be formed from mentions with different entity types (e.g., ”the house” and ”the window”). In fact, while many coreference relations can be identified via string matching facilities, it is not uncommon for bridging relations to be identified using background knowledge and/or sophisticated inference mechanisms. The complexity of bridging resolution is further complicated by the lack of a large corpus annotated with bridging relations: while the most extensively-used coreference-annotated corpus, OntoNotes, contain more than 2000 documents, two of the most commonly-used corpora for bridging resolution, ISNotes and BASHI, each contains only 50 documents taken from OntoNotes. In this dissertation, we investigate knowledge-rich approaches in which we derive potentially useful knowledge for bridging anaphora resolution from a variety of sources, including tasks that we believe are closely related to bridging, manually defined rules based on various syntactic and semantic properties that are directly relevant to the prediction of bridging links, constraints that that encode commonsense knowledge of when two event mentions should or should not have a bridging link, as well as a pre-training objective that are relevant to bridging.

Author dc:creator

• Kobayashi, Hideo

Contributors dc:contributor

• Ng, Vincent
• Summers, Tyler
• Natarajan, Sriraam
• Ouyang, Jessica
• Iyer, Rishabh

• Natural language processing (Computer science)
• Machine learning
• Computer science
• Anaphora resolution

Language dc:language

English

Handle dc:identifier.uri

https://hdl.handle.net/10735.1/10164

OAI identifier oai:identifier

oai:tdl-ir.tdl.org:10735.1/10164