Speaker-Oriented Latent Structures for Dialogue-Based Relation Extraction
This work solves the problem of extracting structured relations from conversational data for natural language processing applications, representing an incremental improvement over existing methods.
The paper tackles the problem of dialogue-based relation extraction (DiaRE) by addressing entangled logic and information sparsity in multi-speaker dialogues, introducing SOLS to induce speaker-oriented latent structures, which improves performance on three public datasets.
Dialogue-based relation extraction (DiaRE) aims to detect the structural information from unstructured utterances in dialogues. Existing relation extraction models may be unsatisfactory under such a conversational setting, due to the entangled logic and information sparsity issues in utterances involving multiple speakers. To this end, we introduce SOLS, a novel model which can explicitly induce speaker-oriented latent structures for better DiaRE. Specifically, we learn latent structures to capture the relationships among tokens beyond the utterance boundaries, alleviating the entangled logic issue. During the learning process, our speaker-specific regularization method progressively highlights speaker-related key clues and erases the irrelevant ones, alleviating the information sparsity issue. Experiments on three public datasets demonstrate the effectiveness of our proposed approach.