AI 中文总结
Nutri-ATLAS是一个具身智能体，通过知识图谱、混合检索和机器人证据获取，实现现实世界中的营养估计、替代品检索和餐食推荐，并在多个基准上取得显著性能。
AI 中文摘要
生成式与智能体物联网系统为数字医疗应用提供了有前景的基础，这些应用结合了现实环境中的感知、个性化推理和自主交互。营养辅助是一个自然的用例，但现有的基于大型语言模型（LLM）的系统通常局限于被动的文本交互和静态上下文，当食物描述模糊或营养证据缺失时，它们变得不可靠。我们提出了Nutri-ATLAS，一个面向现实世界更智能营养的具身智能体，用于表格查询与辅助。它整合了基于图结构的营养推理、硬件感知的LLM选择以及基于机器人的证据获取。Nutri-ATLAS从USDA FoodData Central和FoodKG构建了统一的食物-营养知识图谱，并学习了64维的GATv2食物和食谱嵌入。一个共享的混合图-文本评分机制支持食物营养提取、营养缺口填补、替代品检索和食谱级餐食组合，而一个LLM引导的技能接口导航地标、更新饮食上下文和食物可及性记忆，并将推荐基于观察到的食物可用性。我们在营养估计、替代品检索、食谱推荐、患者画像依从性、边缘部署和现实世界具身执行方面评估了Nutri-ATLAS。在HealthyFoodSubs上，混合检索器达到了37.9%的MAP、80.7%的RR@5和90.1%的RR@10。在NutriBench v2上，Dense+GAT检索在九种量化Qwen3.5-9B配置中为营养估计提供了基础。在PFoodReQ上，Nutri-ATLAS达到了78.8%的MAP、83.0%的MAR和77.5%的F1。一项患者画像研究显示，所有选定案例均符合过敏和健康目标约束。
英文摘要
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.
