Structural Characterization for Dialogue Disentanglement
This addresses challenges in dialogue reading comprehension for both humans and machines, though it is incremental as it builds on prior work by focusing on structural features.
The paper tackles the problem of disentangling multi-party dialogue contexts, where multiple threads flow simultaneously, by modeling structural information like speaker properties and reference dependencies, achieving new state-of-the-art results on the Ubuntu IRC benchmark dataset.
Tangled multi-party dialogue contexts lead to challenges for dialogue reading comprehension, where multiple dialogue threads flow simultaneously within a common dialogue record, increasing difficulties in understanding the dialogue history for both human and machine. Previous studies mainly focus on utterance encoding methods with carefully designed features but pay inadequate attention to characteristic features of the structure of dialogues. We specially take structure factors into account and design a novel model for dialogue disentangling. Based on the fact that dialogues are constructed on successive participation and interactions between speakers, we model structural information of dialogues in two aspects: 1)speaker property that indicates whom a message is from, and 2) reference dependency that shows whom a message may refer to. The proposed method achieves new state-of-the-art on the Ubuntu IRC benchmark dataset and contributes to dialogue-related comprehension.