Robust Dialogue Utterance Rewriting as Sequence Tagging

TL;DR

This paper introduces a sequence-tagging-based model for dialogue rewriting that enhances robustness across domains by reducing search space and employing reinforcement learning with BLEU or GPT-2 signals to improve fluency.

Contribution

The paper proposes a novel sequence-tagging approach combined with reinforcement learning to improve domain robustness and fluency in dialogue utterance rewriting.

Findings

Significant performance improvements on domain transfer tasks.

Enhanced robustness compared to existing models.

Improved fluency through reinforcement learning signals.

Abstract

The task of dialogue rewriting aims to reconstruct the latest dialogue utterance by copying the missing content from the dialogue context. Until now, the existing models for this task suffer from the robustness issue, i.e., performances drop dramatically when testing on a different domain. We address this robustness issue by proposing a novel sequence-tagging-based model so that the search space is significantly reduced, yet the core of this task is still well covered. As a common issue of most tagging models for text generation, the model's outputs may lack fluency. To alleviate this issue, we inject the loss signal from BLEU or GPT-2 under a REINFORCE framework. Experiments show huge improvements of our model over the current state-of-the-art systems on domain transfer.