Reproduce “Visual Task Modeling and Automated Planning for Humanoid Robots in Education of Individuals with Autism Spectrum Disorder” 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
Lada Kudláčková
The authors present a vision-language system that turns images of Shoebox Tasks into task descriptions and action plans, then compares successive images to update the state and remaining plan. On colors-3, ChatGPT-4o averaged 10.9 seconds and ChatGPT-5 75.8 seconds, with zero semantic errors each across ten trials. Tests involved solver responses and manual moves, not robot-assisted teaching.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.