An open-source LLM pipeline with augmentation-aware training, probability-thresholded depth-first search, and product-of-experts scoring reaches 71.6% on the public ARC-AGI evaluation set.
H ow I came in first on A R C - A G I - P ub using S onnet 3.5 with E volutionary T est-time C ompute --- jeremyberman.substack.com
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Product of Experts with LLMs: Boosting Performance on ARC Is a Matter of Perspective
An open-source LLM pipeline with augmentation-aware training, probability-thresholded depth-first search, and product-of-experts scoring reaches 71.6% on the public ARC-AGI evaluation set.