What can large language models do for sustainable food?

2025-02-02

arXiv.org

AI Summary

This study investigates how large language models (LLMs) can advance sustainable food systems and reduce carbon emissions from food production. Addressing the challenge of jointly optimizing environmental impact and human preferences in multi-objective food decision-making, we propose the first general-purpose framework integrating LLMs with combinatorial optimization. We establish a taxonomy of sustainable food tasks and conduct empirical evaluations on protein design and menu optimization. Results show that, in a simulated restaurant setting, our method reduces dietary carbon emissions by 79% while preserving user satisfaction. In protein design, it cuts expert time by 45% on average—substantially outperforming human–human collaboration (22%). The study elucidates LLMs’ efficacy-enhancing mechanisms in expert co-development and exposes their limitations in multi-objective trade-offs, thereby providing a scalable methodology and empirical benchmark for AI-driven sustainable food decision-making.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Food systems are responsible for a third of human-caused greenhouse gas emissions. We investigate what Large Language Models (LLMs) can contribute to reducing the environmental impacts of food production. We define a typology of design and prediction tasks based on the sustainable food literature and collaboration with domain experts, and evaluate six LLMs on four tasks in our typology. For example, for a sustainable protein design task, food science experts estimated that collaboration with an LLM can reduce time spent by 45% on average, compared to 22% for collaboration with another expert human food scientist. However, for a sustainable menu design task, LLMs produce suboptimal solutions when instructed to consider both human satisfaction and climate impacts. We propose a general framework for integrating LLMs with combinatorial optimization to improve reasoning capabilities. Our approach decreases emissions of food choices by 79% in a hypothetical restaurant while maintaining participants' satisfaction with their set of choices. Our results demonstrate LLMs' potential, supported by optimization techniques, to accelerate sustainable food development and adoption.

Problem

• Assess LLMs' role in reducing food system emissions
• Evaluate LLMs on sustainable food design tasks
• Integrate LLMs with optimization to improve sustainability

Innovation

• LLMs reduce design time by 45%
• LLMs integrated with optimization techniques
• Framework cuts emissions by 79%