Reproduce “ClickSight: An LLM Framework for Interpreting Student Clickstreams Through Learning Strategies” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Bahar Radmehr, Lan Tran, Ekaterina Shved, Fatma Bet{ü}l G{ü}reş, Adish Singla, Tanja K{ä}ser
ClickSight connects student clickstreams with literature-defined learning strategies, using relevant actions to support explanations and optional trace-derived counts. In Beer’s Law Lab, Qwen3-80B-Thinking’s overall score rose from 0.451 without guidance to 0.703 with count guidance. No guidance condition significantly improved either tested model in PharmaSim; results concern rubric quality, not learning gains.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.