Unlocking the Potential of User Feedback: Leveraging Large Language Model as User Simulator to Enhance Dialogue System
This work addresses the problem of improving dialogue systems for users by combining LLMs with task-specific models, though it is incremental as it builds on existing methods.
The paper tackles the underperformance of large language models (LLMs) in task-oriented dialogue (TOD) by proposing User-Guided Response Optimization (UGRO), which uses an LLM as a user simulator to provide feedback and optimize a smaller TOD model, achieving state-of-the-art results on two TOD benchmarks.
Dialogue systems and large language models (LLMs) have gained considerable attention. However, the direct utilization of LLMs as task-oriented dialogue (TOD) models has been found to underperform compared to smaller task-specific models. Nonetheless, it is crucial to acknowledge the significant potential of LLMs and explore improved approaches for leveraging their impressive abilities. Motivated by the goal of leveraging LLMs, we propose an alternative approach called User-Guided Response Optimization (UGRO) to combine it with a smaller TOD model. This approach uses LLM as annotation-free user simulator to assess dialogue responses, combining them with smaller fine-tuned end-to-end TOD models. By utilizing the satisfaction feedback generated by LLMs, UGRO further optimizes the supervised fine-tuned TOD model. Specifically, the TOD model takes the dialogue history as input and, with the assistance of the user simulator's feedback, generates high-satisfaction responses that meet the user's requirements. Through empirical experiments on two TOD benchmarks, we validate the effectiveness of our method. The results demonstrate that our approach outperforms previous state-of-the-art (SOTA) results.