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

This paper introduces STANDARDIZE, a retrieval-style in-context learning framework that aligns large language models with expert-defined educational standards (CEFR and Common Core Standards) for automated content generation. The framework extracts knowledge artifacts from standards to guide LLMs in producing text that meets specific grade-level and proficiency requirements for K-12 learners.

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. Focusi

Study Type

Source

Framework Categories

Tool Types

Tags