RecMind: Japanese Movie Recommendation Dialogue with Seeker's Internal State
This work addresses the challenge of improving recommendation dialogue systems by incorporating internal state modeling, though it is incremental as it builds on existing dialogue and recommendation techniques.
The authors tackled the problem of modeling a seeker's internal state in recommendation dialogues by constructing RecMind, a Japanese movie recommendation dataset with entity-level annotations, and found that entities unknown but interesting to the seeker boost recommendation success, with their proposed method outperforming baselines in human evaluations.
Humans pay careful attention to the interlocutor's internal state in dialogues. For example, in recommendation dialogues, we make recommendations while estimating the seeker's internal state, such as his/her level of knowledge and interest. Since there are no existing annotated resources for the analysis, we constructed RecMind, a Japanese movie recommendation dialogue dataset with annotations of the seeker's internal state at the entity level. Each entity has a subjective label annotated by the seeker and an objective label annotated by the recommender. RecMind also features engaging dialogues with long seeker's utterances, enabling a detailed analysis of the seeker's internal state. Our analysis based on RecMind reveals that entities that the seeker has no knowledge about but has an interest in contribute to recommendation success. We also propose a response generation framework that explicitly considers the seeker's internal state, utilizing the chain-of-thought prompting. The human evaluation results show that our proposed method outperforms the baseline method in both consistency and the success of recommendations.