EMNLP Findings2020
Fast End-to-end Coreference Resolution for Korean
Cheoneum Park, Jamin Shin, Sungjoon Park, Joonho Lim, Changki Lee
Abstract
Recently, end-to-end neural network-based approaches have shown significant improvements over traditional pipeline-based models in English coreference resolution. However, such advancements came at a cost of computational complexity and recent works have not focused on tackling this problem. Hence, in this paper, to cope with this issue, we propose BERT-SRU-based Pointer Networks that leverages the linguistic property of head-final languages. Applying this model to the Korean coreference resolution, we significantly reduce the coreference linking search space. Combining this with Ensemble Knowledge Distillation, we maintain state-of-the-art performance 66.9% of CoNLL F1 on ETRI test set while achieving 2x speedup (30 doc/sec) in document processing time.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Related papers
- Efficient Constituency Parsing by PointingThanh-Tung Nguyen, Xuan-Phi Nguyen, Shafiq R. Joty, Xiaoli LiACL 2020 · 11 citations
- A Pointer Network Architecture for Joint Morphological Segmentation and TaggingAmit Seker, Reut TsarfatyEMNLP Findings 2020 · 9 citations
- Transition-based Semantic Dependency Parsing with Pointer NetworksDaniel Fernández-González, Carlos Gómez-RodríguezACL 2020 · 4 citations
- Encoder-Decoder Based Unified Semantic Role Labeling with Label-Aware SyntaxHao Fei, Fei Li, Bobo Li, Donghong JiAAAI 2021 · 60 citations
- Scaling Within Document Coreference to Long TextsRaghuveer Thirukovalluru, Nicholas Monath, Kumar Shridhar, Manzil Zaheer et al.ACL Findings 2021 · 11 citations
