Coreference-aware Double-channel Attention Network for Multi-party Dialogue Reading Comprehension

• Published in: 2023 International Joint Conference on Neural Networks (IJCNN) pp. 1 - 8
• Main Authors: Li, Yanling, Zou, Bowei, Fan, Yifan, Dong, Mengxing, Hong, Yu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.06.2023
• Subjects: Benchmark testing; Context; Coreference-aware attention; Encoding; Feature extraction; Interaction modeling; Multi-party dialogue reading comprehension; Neural networks; Oral communication; Performance gain; Utterance profiling
• ISSN: 2161-4407
• DOI: 10.1109/IJCNN54540.2023.10191414

We tackle Multi-party Dialogue Reading Comprehension (abbr., MDRC). MDRC stands for an extractive reading comprehension task grounded on a batch of dialogues among multiple interlocutors. It is challenging due to the requirement of understanding cross-utterance contexts and relationships in a multi-turn multi-party conversation. Previous studies have made great efforts on the utterance profiling of a single interlocutor and graph-based interaction modeling. The corresponding solutions contribute to the answer-oriented reasoning on a series of well-organized and thread-aware conversational contexts. However, the current MDRC models still suffer from two bottlenecks. On the one hand, a pronoun like "it" most probably produces multi-skip reasoning throughout the utterances of different interlocutors. On the other hand, an MDRC encoder is potentially puzzled by fuzzy features, i.e., the mixture of inner linguistic features in utterances and external interactive features among utterances. To overcome the bottlenecks, we propose a coreference-aware attention modeling method to strengthen the reasoning ability. In addition, we construct a two-channel encoding network. It separately encodes utterance profiles and interactive relationships, so as to relieve the confusion among heterogeneous features. We experiment on the benchmark corpora Molweni and FriendsQA. Experimental results demonstrate that our approach yields substantial improvements on both corpora, compared to the fine-tuned BERT and ELECTRA baselines. The maximum performance gain is about 2.5% F\mathbf{1}-\mathbf{score} . Besides, our MDRC models outperform the state-of-the-art in most cases.