TPLinker: Single-stage Joint Extraction of Entities and Relations Through Token Pair Linking
This addresses the problem of overlapping relation extraction in NLP, offering a more efficient solution for tasks like knowledge graph construction, but it is incremental as it builds on prior joint learning methods.
The paper tackles the challenge of extracting overlapping entities and relations from text by proposing TPLinker, a single-stage joint extraction model that avoids exposure bias and error accumulation. It achieves state-of-the-art performance on two public datasets, with significant improvements in handling overlapping and multiple relations.
Extracting entities and relations from unstructured text has attracted increasing attention in recent years but remains challenging, due to the intrinsic difficulty in identifying overlapping relations with shared entities. Prior works show that joint learning can result in a noticeable performance gain. However, they usually involve sequential interrelated steps and suffer from the problem of exposure bias. At training time, they predict with the ground truth conditions while at inference it has to make extraction from scratch. This discrepancy leads to error accumulation. To mitigate the issue, we propose in this paper a one-stage joint extraction model, namely, TPLinker, which is capable of discovering overlapping relations sharing one or both entities while immune from the exposure bias. TPLinker formulates joint extraction as a token pair linking problem and introduces a novel handshaking tagging scheme that aligns the boundary tokens of entity pairs under each relation type. Experiment results show that TPLinker performs significantly better on overlapping and multiple relation extraction, and achieves state-of-the-art performance on two public datasets.