37% on ARC-AGI 2 without an LLM
Comments
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Summary
A solo researcher's mdlARC, a 100M-parameter autoregressive transformer trained from scratch only on ARC puzzles with no LLM, reports 37.08% on the Kaggle public score for ARC-AGI-2, against 27.64% for last year's LLM-based winner.