Conversational Semantic Role Labeling with Predicate-Oriented Latent Graph
This work addresses the CSRL task for dialogue analysis, focusing on structural information integration and near-neighbor influence, which is incremental as it builds on existing CSRL methods.
The authors tackled conversational semantic role labeling (CSRL) by proposing a predicate-oriented latent graph (POLar) with a predicate-centered Gaussian mechanism and a dialogue-level pre-trained language model, CoDiaBERT, achieving over 4% F1 score improvements on cross-utterance argument detection compared to best-performing baselines.
Conversational semantic role labeling (CSRL) is a newly proposed task that uncovers the shallow semantic structures in a dialogue text. Unfortunately several important characteristics of the CSRL task have been overlooked by the existing works, such as the structural information integration, near-neighbor influence. In this work, we investigate the integration of a latent graph for CSRL. We propose to automatically induce a predicate-oriented latent graph (POLar) with a predicate-centered Gaussian mechanism, by which the nearer and informative words to the predicate will be allocated with more attention. The POLar structure is then dynamically pruned and refined so as to best fit the task need. We additionally introduce an effective dialogue-level pre-trained language model, CoDiaBERT, for better supporting multiple utterance sentences and handling the speaker coreference issue in CSRL. Our system outperforms best-performing baselines on three benchmark CSRL datasets with big margins, especially achieving over 4% F1 score improvements on the cross-utterance argument detection. Further analyses are presented to better understand the effectiveness of our proposed methods.