A Neuro-Symbolic Approach to Monitoring Salt Content in Food
This addresses a health monitoring problem for heart failure patients, but it is incremental as it applies a known neuro-symbolic approach to a specific domain.
The paper tackles the problem of monitoring salt content in food for heart failure patients by developing a dialogue system, and finds that integrating neuro-symbolic rules improves joint goal accuracy by over 20% compared to fine-tuning transformer models alone.
We propose a dialogue system that enables heart failure patients to inquire about salt content in foods and help them monitor and reduce salt intake. Addressing the lack of specific datasets for food-based salt content inquiries, we develop a template-based conversational dataset. The dataset is structured to ask clarification questions to identify food items and their salt content. Our findings indicate that while fine-tuning transformer-based models on the dataset yields limited performance, the integration of Neuro-Symbolic Rules significantly enhances the system's performance. Our experiments show that by integrating neuro-symbolic rules, our system achieves an improvement in joint goal accuracy of over 20% across different data sizes compared to naively fine-tuning transformer-based models.