Cross-document Coreference Resolution over Predicted Mentions

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 5100–5107 August 1–6, 2021. ©2021 Association for Computational Linguistics 5100 Cross-document Coreference Resolution over Predicted Mentions Arie Cattan1 Alon Eirew1,2 Gabriel Stanovsky3 Mandar Joshi4 Ido Dagan1 1Computer Science Department, Bar Ilan University 2Intel Labs, Israel 3The Hebrew University of Jerusalem 4Allen School of Computer Science & Engineering, University of Washington, Seattle, WA arie.cattan@gmail.com alon.eirew@intel.com gabis@cse.huji.ac.il mandar90@cs.washington.edu dagan@cs.biu.ac.il Abstract Coreference resolution has been mostly in- vestigated within a single document scope, showing impressive progress in recent years based on end-to-end models. However, the more challenging task of cross-document (CD) coreference resolution remained rela- tively under-explored, with the few recent models applied only to gold mentions. Here, we introduce the first end-to-end model for CD coreference resolution from raw text, which extends the prominent model for within- document coreference to the CD setting. Our model achieves competitive results for event and entity coreference resolution on gold men- tions. More importantly, we set first base- line results, on the standard ECB+ dataset, for CD coreference resolution over predicted men- tions. Further, our model is simpler and more efficient than recent CD coreference resolu- tion systems, while not using any external re- sources.1 1 Introduction Cross-document (CD) coreference resolution con- sists of identifying textual mentions across multiple documents that refer to the same concept. For ex- ample, consider the following sentences from the ECB+ dataset (Cybulska and Vossen, 2014), where colors represent coreference clusters (for brevity, we omit some clusters): 1. Thieves pulled off a two million euro jewellery heist in central Paris on Monday after smashing their car through the store’s front window. 2. Four men drove a 4x4 through the front win- dow of the store on Rue de Castiglione, before making off with the jewellery and watches. Despite its importance for downstream tasks, CD coreference resolution has been lagging behind the 1https://github.com/ariecattan/coref impressive strides made in the scope of a single doc- ument (Lee et al., 2017; Joshi et al., 2019, 2020; Wu et al., 2020). Further, state-of-the-art models exhibit several shortcomings, such as operating on gold mentions or relying on external resources such as SRL or a paraphrase dataset (Shwartz et al., 2017), preventing them from being applied on real- istic settings. To address these limitations, we develop the first end-to-end CD coreference model building upon a prominent within-document (WD) coreference model (Lee et al., 2017) which we extend with re- cent advances in transformer-based encoders. We address the inherently non-linear nature of the CD setting by combining the WD coreference model with agglomerative clustering that was shown use- ful in CD models. Our model achieves competitive results on ECB+ over gold mentions and sets base- line results over predicted mentions. Our model is also simpler and substantially more efficient than existing CD coreference systems. Taken together, our work seeks to bridge the gap between WD and CD coreference, driving further research of the lat- ter in realistic settings. 2 Background Cross-document coreference Previous works on CD coreference resolution learn …

Content truncated to 500 words