FAM-Bench: A Multimodal Benchmark for Condition-Aware Food-as-Medicine Reasoning

2026-05-29

AI Summary

This work addresses the lack of evaluation benchmarks capable of assessing food suitability under specific health conditions in existing food AI systems. To bridge this gap, the study proposes the first multimodal “Food-as-Medicine” reasoning benchmark tailored to 13 diet-related health conditions. The framework integrates visual recognition, ingredient-level semantic understanding, and clinical nutritional constraints to enable condition-aware food suitability assessment. It supports a four-dish comparative ranking task and introduces a standardized test set of 2,500 expert-validated instances. This benchmark provides a unified platform for evaluating health-aware reasoning capabilities in both language and vision-language models, thereby filling a critical void in health-oriented food reasoning evaluation.

Abstract

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.

Problem

• Food-as-Medicine
• health-aware reasoning
• diet-related conditions
• multimodal benchmark
• condition-specific suitability

Innovation

• Food-as-Medicine
• multimodal benchmark
• condition-aware reasoning
• health-aware AI
• nutrition-sensitive evaluation