Nutri-ATLAS: Embodied Agent for Tabulated Lookup and Assistance for Smarter nutrition

2026-09-26

AI Summary

This study addresses the reliability limitations of large language model-based nutritional assistants when processing ambiguous food descriptions or lacking empirical evidence. To overcome these challenges, we propose an embodied agent framework that introduces a unified knowledge graph coupled with a hybrid scoring mechanism. The framework integrates GATv2 graph reasoning, a hardware-aware quantized model (Qwen3.5), and a robotic navigation interface for evidence acquisition, thereby enabling a paradigm shift from passive question answering to active real-world verification. Experimental results demonstrate that the proposed approach significantly improves retrieval accuracy across multiple benchmarks while supporting edge deployment and strictly adhering to patient-specific dietary constraints.

Abstract

Generative and Agentic IoT systems offer a promising foundation for digital healthcare applications that combine sensing, personalized reasoning, and autonomous interaction in real-world environments. Nutrition assistance is a natural use case, but existing Large Language Model (LLM)-based systems are often limited to passive text interaction and static context, making them unreliable when food descriptions are ambiguous or nutritional evidence is missing. We propose Nutri-ATLAS, an Embodied Agent for Tabulated Lookup and Assistance for smarter nutrition in the real world. It integrates graph-grounded nutrition reasoning, hardware-aware LLM selection, and robot-based evidence acquisition. Nutri-ATLAS builds a unified Food-Nutrient knowledge graph from USDA FoodData Central and FoodKG and learns 64-dimensional GATv2 food and recipe embeddings. A shared hybrid graph-text scoring mechanism supports food nutrition extraction, nutritional gap filling, substitute retrieval, and recipe-level meal composition, while an LLM-guided skill interface navigates landmarks, updates dietary-context and food-accessibility memory, and grounds recommendations in observed food availability. We evaluate Nutri-ATLAS across nutrient estimation, substitution retrieval, recipe recommendation, patient-profile adherence, edge deployment, and real-world embodied execution. On HealthyFoodSubs, the hybrid retriever achieves 37.9% MAP, 80.7% RR@5, and 90.1% RR@10. On NutriBench v2, Dense+GAT retrieval grounds nutrient estimation across nine quantized Qwen3.5-9B configurations. On PFoodReQ, Nutri-ATLAS reaches 78.8% MAP, 83.0% MAR, and 77.5% F1. A patient-profile study shows adherence to allergy and healthy-target constraints for all selected cases.

Problem

• Nutrition Assistance
• Embodied Agent
• Large Language Model
• Food Ambiguity
• Digital Healthcare

Innovation

• Embodied Agent
• Knowledge Graph
• Graph Attention Network
• Hybrid Retrieval
• Edge Deployment