Enhancing Goal-oriented Proactive Dialogue Systems via Dynamic Multi-dimensional Consistency Optimization

AbstractPrevious work on goal-oriented proactive dialogue systems frequently failed to address the multi-dimensional consistency issue between generated responses and key contextual elements (e.g., user profile, dialogue history, domain knowledge, and subgoal). To address this issue, we propose a novel Dynamic Multi-dimensional Consistency Reinforcement Learning (DMCRL) framework, which adaptively measures the impact of each consistency dimension on overall dialogue quality and provides targeted feedback to improve response quality. Experimental results on two datasets demonstrate that our DMCRL significantly improves the consistency of generated responses.