AI-Empowered Teaching of Chemistry Literature Searching: Paradigm Restructuring, Practical Pathways, and Core Competency Cultivation

Abstract

With the comprehensive establishment of the "AI for Science" (AI4S) paradigm, Artificial Intelligence Generated Content (AIGC) and Large Language Models (LLMs) have become fundamental infrastructures for scientific research. Chemistry literature retrieval and reading, as core components of chemical education, are undergoing a profound transformation from "keyword matching" to "semantic interaction and intelligent generation." This paper systematically analyzes the pain points in traditional chemistry literature retrieval teaching, such as high cognitive load, information overload, and low efficiency in cross-lingual reading. It deeply explores the application mechanisms of semantic search, multimodal chemical information extraction, and AI-assisted reading technologies in acquiring chemical information. Based on this, the paper proposes a restructuring pathway for teaching paradigms empowered by AI, emphasizing that the focus of teaching should shift toward "Prompt Engineering, human-machine collaborative retrieval, result verification, and critical thinking cultivation." Furthermore, it provides specific Project-Based Learning (PBL) cases and evaluation system reform schemes, aiming to offer theoretical reference and practical guidance for the information literacy education of outstanding chemical innovators in the new era.