Bridging Resolution: Making Sense of the State of the Art

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1652–1659 June 6–11, 2021. ©2021 Association for Computational Linguistics 1652 Bridging Resolution: Making Sense of the State of the Art Hideo Kobayashi and Vincent Ng Human Language Technology Research Institute University of Texas at Dallas Richardson, TX 75083-0688 {hideo,vince}@hlt.utdallas.edu Abstract While Yu and Poesio (2020) have recently demonstrated the superiority of their neu- ral multi-task learning (MTL) model to rule- based approaches for bridging anaphora reso- lution, there is little understanding of (1) how it is better than the rule-based approaches (e.g., are the two approaches making similar or com- plementary mistakes?) and (2) what should be improved. To shed light on these issues, we (1) propose a hybrid rule-based and MTL ap- proach that would enable a better understand- ing of their comparative strengths and weak- nesses; and (2) perform a manual analysis of the errors made by the MTL model. 1 Introduction Bridging resolution is an anaphora resolution task that involves identifying and resolving bridg- ing/associative anaphors, which are anaphoric ref- erences to non-identical associated antecedents. To exemplify, consider the following sentences taken from the BASHI corpus (Rösiger, 2018a): Even if baseball triggers losses at CBS – and he doesn’t think it will – “I’d rather see the games on our air than on NBC and ABC,” he says . In this example, a bridging link exists between the anaphor the games and its antecedent baseball, as the definite description cannot be interpreted correctly unless it is associated with baseball. Bridging resolution is arguably more challenging than entity coreference resolution. The reason is that unlike in entity coreference, in bridging resolu- tion there are typically no clear syntactic or surface clues for identifying the antecedent of a bridging anaphor. In many cases, resolution requires the use of context as well as commonsense inference. Despite the difficulty of bridging resolution, the annotated corpora available for training bridging resolvers are much smaller than those for train- ing entity coreference resolvers (e.g., OntoNotes (Hovy et al., 2006)). As a result, early work has fo- cused on developing rule-based systems (e.g., Hou et al. (2014), Rösiger (2018b)). A key weakness of rule-based approaches is that the ruleset may have to be updated when it is applied to a new cor- pus (e.g., new rules may have to be added, and existing rules may have to be removed or modi- fied), as different bridging corpora are annotated with slightly different guidelines (to cover differ- ent kinds of bridging links, for instance). In light of this weakness, Yu and Poesio (2020) have re- cently proposed a neural bridging resolver based on multi-task learning (MTL). Despite being trained on the relatively small amount of labeled data that are currently available, their resolver has achieved state-of-the-art results on three evaluation corpora. In this paper, we seek to make sense of this state of the art by shedding light on two issues. First, how is the MTL …

Content truncated to 500 words