Supporting Data and Full Run Artifacts for "Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?"

This dataset contains the complete result data and full run artifacts supporting the manuscript “Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?” The artifacts include recorded agent–environment interactions, agent workspaces, and the executable world models generated by the agents during the runs. Files: - article2_main_runs.tar.gz — Runs from the main controlled comparison of four agent variants using GPT-5.4 and GPT-5.5 at high and xhigh reasoning effort - article2_main_runs_additional.tar.gz — Exploratory follow-up runs using GPT-5.6-sol at xhigh and max reasoning effort.