OmniFood-Bench: Evaluating VLMs for Nutrient Reasoning and Personalized Health Advice

2026-07-09

AI Summary

This work addresses the limitations of vision-language models (VLMs) in delivering accurate nutritional reasoning and personalized health recommendations within food systems, primarily due to information asymmetry. To this end, the authors introduce OmniFood-Bench, a novel benchmark built upon the MM-Food-100K dataset, featuring a three-tiered evaluation framework that encompasses ingredient recognition, portion estimation, and disease-specific dietary advice. This is the first systematic assessment of VLMs’ capabilities spanning from basic visual perception to safety-critical health decision-making. Through large-scale automated evaluation of state-of-the-art VLMs—including GPT-5.1, Gemini-3-Flash, and Qwen3-VL-8B—augmented with fine-grained visual and nutrition-medical knowledge, the study reveals that while current models achieve near-human performance in dish recognition, they exhibit significant deficiencies in portion estimation and high-risk diabetic recommendations, highlighting critical bottlenecks for trustworthy deployment in health-related applications.

Abstract

The rapid integration of Large Vision-Language Models (VLMs) into critical infrastructure promises to revolutionize personalized healthcare and dietary management. However, in the domain of food systems, autonomous agents face a unique and persistent challenge: the "Systemic Information Asymmetry" between visual appearance and intrinsic nutritional composition. Existing benchmarks primarily focus on coarse-grained classification tasks, such as food category recognition, which fail to evaluate the intricate reasoning chain required for real-world dietary management -- specifically, the ability to traverse from identifying hidden ingredients to estimating physical mass, and finally synthesizing safety-critical medical advice. In this paper, we introduce OmniFood-Bench, a comprehensive benchmark constructed from the MM-Food-100K dataset. Unlike previous works, OmniFood-Bench evaluates VLMs across three progressive capabilities: Basic Perception (Ingredients & Cooking Methods), Quantitative Reasoning (Portion Size & Nutritional Profiling), and Safety-Critical Advisory (Disease-Specific Recommendations). We evaluate six state-of-the-art VLMs, including gpt-5.1, gemini-3-flash, and qwen3-vl-8B. Our extensive experiments reveal a startling "Semantic-Physical Gap": while models achieve near-human accuracy in naming dishes, they exhibit catastrophic failure in mass estimation and frequently hallucinate benign advice for high-risk diabetic profiles. This work establishes a rigorous standard for trustworthiness in autonomous agents deployed for public health. The code and datasets are available in: https://anonymous.4open.science/r/OmniFood-Bench-7D0B

Problem

• Systemic Information Asymmetry
• Nutrient Reasoning
• Personalized Health Advice
• Visual-Language Models
• Food Systems

Innovation

• Vision-Language Models
• Nutrient Reasoning
• Personalized Health Advice
• Benchmark Evaluation
• Systemic Information Asymmetry