Research project

LLM Generation and Evaluation

Research project associated with ACM Journal of Data and Information Quality (JDIQ).

Core approach

Uses selected teaching examples and prompting to improve classroom dialogue completion and expertise transfer, with automatic and human evaluation.

The framework studies dialogue completion and transfer of instructional expertise to new teaching topics. Evaluation on two collected datasets combines language metrics, perplexity, and human questionnaires.

Associated publication

ELF: Educational LLM Framework of Improving and Evaluating AI-generated Content for Classroom Teaching