Multi-task Learning Based Neural Bridging Reference Resolution
This work solves the challenge of resolving bridging references in natural language processing, which is crucial for tasks like discourse understanding, but it is incremental as it builds on multi-task learning approaches.
The paper tackles the problem of bridging reference resolution by addressing the lack of large annotated corpora and inconsistent definitions across datasets, achieving state-of-the-art results with improvements of up to 8 p.p. on full bridging resolution and outperforming previous best results by up to 36.3 p.p. across multiple corpora.
We propose a multi task learning-based neural model for resolving bridging references tackling two key challenges. The first challenge is the lack of large corpora annotated with bridging references. To address this, we use multi-task learning to help bridging reference resolution with coreference resolution. We show that substantial improvements of up to 8 p.p. can be achieved on full bridging resolution with this architecture. The second challenge is the different definitions of bridging used in different corpora, meaning that hand-coded systems or systems using special features designed for one corpus do not work well with other corpora. Our neural model only uses a small number of corpus independent features, thus can be applied to different corpora. Evaluations with very different bridging corpora (ARRAU, ISNOTES, BASHI and SCICORP) suggest that our architecture works equally well on all corpora, and achieves the SoTA results on full bridging resolution for all corpora, outperforming the best reported results by up to 36.3 p.p..