Semantic Role Labeling Guided Multi-turn Dialogue ReWriter
This work addresses noise reduction in dialogue systems for applications like chatbots, but it is incremental as it builds on existing attentive models with an additional guidance mechanism.
The paper tackles the problem of inaccurate attention in multi-turn dialogue rewriting by using semantic role labeling to guide the model, resulting in significant performance improvements over a RoBERTa-based model that already surpasses previous state-of-the-art systems.
For multi-turn dialogue rewriting, the capacity of effectively modeling the linguistic knowledge in dialog context and getting rid of the noises is essential to improve its performance. Existing attentive models attend to all words without prior focus, which results in inaccurate concentration on some dispensable words. In this paper, we propose to use semantic role labeling (SRL), which highlights the core semantic information of who did what to whom, to provide additional guidance for the rewriter model. Experiments show that this information significantly improves a RoBERTa-based model that already outperforms previous state-of-the-art systems.