Comprehensive Evaluation of Large Multimodal Models for Nutrition Analysis: A New Benchmark Enriched with Contextual Metadata

TL;DR

This paper evaluates large multimodal models for nutrition analysis, demonstrating that integrating contextual metadata like location and time significantly improves prediction accuracy, and introduces a new dataset for benchmarking.

Contribution

It introduces ACETADA, a new publicly available dataset, and systematically assesses how contextual metadata enhances LMM performance in nutrition estimation tasks.

Findings

Metadata integration reduces MAE and MAPE in nutritional predictions.

Context-aware prompting strategies outperform image-only prompts.

Incorporating reasoning modifiers further improves model accuracy.

Abstract

Large Multimodal Models (LMMs) are increasingly applied to meal images for nutrition analysis. However, existing work primarily evaluates proprietary models, such as GPT-4. This leaves the broad range of LLMs underexplored. Additionally, the influence of integrating contextual metadata and its interaction with various reasoning modifiers remains largely uncharted. This work investigates how interpreting contextual metadata derived from GPS coordinates (converted to location/venue type), timestamps (transformed into meal/day type), and the food items present can enhance LMM performance in estimating key nutritional values. These values include calories, macronutrients (protein, carbohydrates, fat), and portion sizes. We also introduce \textbf{ACETADA}, a new food-image dataset slated for public release. This open dataset provides nutrition information verified by the dietitian and serves as the foundation for our analysis. Our evaluation across eight LMMs (four open-weight and four closed-weight) first establishes the benefit of contextual metadata integration over straightforward prompting with images alone. We then demonstrate how this incorporation of contextual information enhances the efficacy of reasoning modifiers, such as Chain-of-Thought, Multimodal Chain-of-Thought, Scale Hint, Few-Shot, and Expert Persona. Empirical results show that integrating metadata intelligently, when applied through straightforward prompting strategies, can significantly reduce the Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) in predicted nutritional values. This work highlights the potential of context-aware LMMs for improved nutrition analysis.