Standardize: Aligning Language Models with Expert-Defined Standards for Content Generation

TL;DR

This paper introduces Standardize, a framework that uses expert-defined standards to guide large language models in generating content aligned with domain-specific quality criteria, improving accuracy significantly.

Contribution

It presents a retrieval-style in-context learning approach to incorporate standards into language model content generation, a novel method in controllable text generation.

Findings

Models improved 45% to 100% in accuracy when guided by standards.

The approach is effective across different large language models.

Standards-based guidance enhances content quality and consistency.

Abstract

Domain experts across engineering, healthcare, and education follow strict standards for producing quality content such as technical manuals, medication instructions, and children's reading materials. However, current works in controllable text generation have yet to explore using these standards as references for control. Towards this end, we introduce Standardize, a retrieval-style in-context learning-based framework to guide large language models to align with expert-defined standards. Focusing on English language standards in the education domain as a use case, we consider the Common European Framework of Reference for Languages (CEFR) and Common Core Standards (CCS) for the task of open-ended content generation. Our findings show that models can gain a 45% to 100% increase in precise accuracy across open and commercial LLMs evaluated, demonstrating that the use of knowledge artifacts extracted from standards and integrating them in the generation process can effectively guide models to produce better standard-aligned content.