Reproduce “Human Thinking under Plural LLM Assistance: Mathematical Problem Solving and Open-Ended Writing” 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
Harsh Kumar, Jace Mu, Jonathan Vincentius, Ashton Anderson
The study compares arrangements of AI assistance in math problem solving and essay writing. In math, participants receiving tutor and peer support had the highest observed accuracy on a later unassisted test, but did not significantly outperform tutor-only support. In writing, two role-specialized assistants kept idea similarity near the no-AI baseline; this difference from one assistant was not significant.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.