FAM-Bench: A Multimodal Benchmark for Condition-Aware Food-as-Medicine Reasoning
This benchmark addresses a critical gap in food AI by providing a standardized testbed for health-aware food reasoning, which is essential for developing practical Food-as-Medicine applications for patients and healthcare providers.
This paper introduces FAM-Bench, a multimodal benchmark designed to evaluate AI models' ability to determine the suitability of food choices for specific health conditions. It comprises 2500 expert-verified instances across 13 diet-related conditions, testing models on dish suitability assessment and comparative dish analysis.
Food-as-Medicine requires models to reason beyond what a dish is or what nutrition it contains: they must decide whether a concrete food choice is appropriate for a specific health condition. Existing food AI benchmarks primarily evaluate dish recognition, recipe understanding, nutrient estimation, or general nutrition question answering, leaving this health-aware decision layer largely untested. We introduce FAM-Bench, a multi-modal Food-as-Medicine benchmark with 2500 nutrition-expert-verified instances across 13 diet-related health conditions. The benchmark contains two complementary tasks: dish-level suitability assessment, where models judge whether a dish is suitable for a condition from its image and ingredient list, and comparative dish analysis, where models rank four candidate dishes by condition-specific suitability. Both tasks require integrating ingredient evidence, visual preparation cues, and clinical nutrition constraints, providing a standardized testbed for grounded health-aware reasoning in language and vision-language models.