CGU-ILALab at FoodBench-QA 2026: Comparing Traditional and LLM-based Approaches for Recipe Nutrient Estimation

TL;DR

This paper systematically compares traditional lexical, deep semantic, and LLM-based models for recipe nutrient estimation, revealing trade-offs between accuracy and efficiency, with LLMs offering superior accuracy but higher latency.

Contribution

It provides a comprehensive evaluation of multiple modeling approaches for nutrient estimation, highlighting the benefits and limitations of LLMs in this domain.

Findings

LLMs outperform lexical and deep encoders in accuracy.

Lexical methods are fastest but less accurate.

LLMs have higher inference latency, affecting real-time use.

Abstract

Accurate nutrient estimation from unstructured recipe text is an important yet challenging problem in dietary monitoring, due to ambiguous ingredient terminology and highly variable quantity expressions. We systematically evaluate models spanning a wide range of representational capacity, from lexical matching methods (TF-IDF with Ridge Regression), to deep semantic encoders (DeBERTa-v3), to generative reasoning with large language models (LLMs). Under the strict tolerance criteria defined by EU Regulation 1169/2011, our empirical results reveal a clear trade-off between predictive accuracy and computational efficiency. The TF-IDF baseline achieves moderate nutrient estimation performance with near-instantaneous inference, whereas the DeBERTa-v3 encoder performs poorly under task-specific data scarcity. In contrast, few-shot LLM inference (e.g., Gemini 2.5 Flash) and a hybrid LLM refinement pipeline (TF-IDF combined with Gemini 2.5 Flash) deliver the highest validation accuracy across all nutrient categories. These improvements likely arise from the ability of LLMs to leverage pre-trained world knowledge to resolve ambiguous terminology and normalize non-standard units, which remain difficult for purely lexical approaches. However, these gains come at the cost of substantially higher inference latency, highlighting a practical deployment trade-off between real-time efficiency and nutritional precision in dietary monitoring systems.