This paper presents a useful tool leveraging Large Language Models (LLM) to resolve discrepancies between corporate and national standards. Firstly, we introduce a Food Standard Dataset comprising 1420 Chinese national standards and 100 corporate standards. Secondly, we offer a user-friendly web application to show human-readable modification suggestions. Thirdly, we propose a technique for standard information extraction, which efficiently retrieves relevant information from complete national standards.
Li et al. (2026) studied this question.