Research on Personalized Popular Science Q&A Generation Method Integrating Difficulty-Aware Retrieval-Augmented Generation
Cognitive Strategies in Study · 2026-08-16 · 2 (3) · 1-18
DOI: 10.63808/css.v2i3.426 ↗At a glance
Journal article in Cognitive Strategies in Study, published online on 2026-08-16. Written by Yuewen Cao and 3 co-authors. The full text is freely available (open access, CC BY 4.0).
Abstract (original language)
General large language models suffer from factual hallucinations, mismatched expression difficulty and insufficient individual adaptation in science education question-and-answer (Q&A) tasks. To address these issues, this paper proposes a personalized popular science Q&A generation method integrating difficulty-aware retrieval-augmented generation. Firstly, a scientific knowledge base with academic stage difficulty labels is constructed in this method. Secondly, learner portrait matching items and knowledge difficulty deviation penalty items are introduced into the traditional semantic retrieval score to realize evidence screening for students in different academic stages. In the generation stage, structured prompts are organized with retrieved evidence, student portraits and answer constraints, and parameter-efficient fine-tuning is adopted to enhance the model’s adaptability to popular science Q&A style and knowledge boundaries. Taking a self-built primary and secondary school science Q&A dataset as the research object, experiments are conducted from the dimensions of scientific accuracy, age appropriateness, completeness and hallucination rate. The experimental results show that the proposed method outperforms conventional retrieval-augmented methods in terms of scientific accuracy and age appropriateness, and can improve the reliability and teaching applicability of popular science Q&A with low training costs.
Cite
APA
Cao, Y., Li, X., Lei, C., & Zhang, L. (2026). Research on Personalized Popular Science Q&A Generation Method Integrating Difficulty-Aware Retrieval-Augmented Generation. Cognitive Strategies in Study, 2(3), 1-18. https://doi.org/10.63808/css.v2i3.426
Vancouver
Cao Y, Li X, Lei C, Zhang L. Research on Personalized Popular Science Q&A Generation Method Integrating Difficulty-Aware Retrieval-Augmented Generation. Cognitive Strategies in Study. 2026;2(3):1-18. doi:10.63808/css.v2i3.426
BibTeX
@article{Cao2026,
author = {Cao, Yuewen and Li, Xinyang and Lei, Chunyu and Zhang, Lin},
title = {Research on Personalized Popular Science Q&A Generation Method Integrating Difficulty-Aware Retrieval-Augmented Generation},
journal = {Cognitive Strategies in Study},
year = {2026},
volume = {2},
number = {3},
pages = {1--18},
doi = {10.63808/css.v2i3.426}
}
EUVIMED