Research
Training and Calibration Data for Open Models on ARC-AGI-2
This paper investigates which training and calibration data help three open models solve ARC-AGI-2 tasks offline. The authors built a corpus of executable rules and synthetic tasks, but found that fine-tuning on real task records outperformed synthetic data. For a reasoning model, an inference procedure contributed most of the score gain, while expert pruning and routing statistics improved a mixture-of-experts model's performance.
ARC-AGI-2open modelstraining data