Is MultiWOZ a Solved Task? An Interactive TOD Evaluation Framework with User Simulator
This addresses the evaluation bottleneck for TOD researchers by enabling more realistic assessments, though it is incremental as it builds on existing simulators and metrics.
The authors tackled the policy mismatch problem in evaluating Task-Oriented Dialogue (TOD) systems by proposing an interactive evaluation framework with a user simulator, achieving nearly 98% inform and success rates on the MultiWOZ dataset.
Task-Oriented Dialogue (TOD) systems are drawing more and more attention in recent studies. Current methods focus on constructing pre-trained models or fine-tuning strategies while the evaluation of TOD is limited by a policy mismatch problem. That is, during evaluation, the user utterances are from the annotated dataset while these utterances should interact with previous responses which can have many alternatives besides annotated texts. Therefore, in this work, we propose an interactive evaluation framework for TOD. We first build a goal-oriented user simulator based on pre-trained models and then use the user simulator to interact with the dialogue system to generate dialogues. Besides, we introduce a sentence-level and a session-level score to measure the sentence fluency and session coherence in the interactive evaluation. Experimental results show that RL-based TOD systems trained by our proposed user simulator can achieve nearly 98% inform and success rates in the interactive evaluation of MultiWOZ dataset and the proposed scores measure the response quality besides the inform and success rates. We are hoping that our work will encourage simulator-based interactive evaluations in the TOD task.