Reproduce “ProofPaths: An AI Agent Framework for Assessing Student Reasoning” 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
Mohsen Hariri, Vipin Chaudhary
ProofPaths is a framework for grading student work and conducting individualized written interviews about students’ reasoning. Three agents review and revise rubric-based assessments, while an interview agent uses answers to select follow-up questions tied to retrieved submission passages. In nine deterministic probes with synthetic inputs and scripted responses, checks passed; this verifies controller behavior, not educational effectiveness or grading accuracy.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.