Reproduce “TanCRAFT: Context-Reasoned Anticipatory Feedback for Assembly Tasks” 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
Sassan Mokhtar, Lars Doorenbos, Marius Bock, Sina Mokhtar, Katharina Beckh, Dominik Bach, Juergen Gall
TanCRAFT predicts a manipulated tangram piece’s destination and checks whether the current partial assembly has capacity for it, enabling warnings without prescribing a fixed next step. In a study of 20 participants, predictive feedback reduced mistakes to 0.90 ± 1.4, versus 4.90 ± 5.4 without feedback and 1.95 ± 2.4 with reactive feedback. Findings concern the tested tangram setting.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.