ARC-AGI-2 is a benchmark of human-intuitive visual puzzles that measures a machine's ability to generalize from limited examples, interpret symbolic meaning, and flexibly apply rules in varying contexts. In this paper, we discuss our approach to solving the ARC-AGI-2 puzzles with TinyLM, with additional fine-tuning at test time, including Test-Time-Training (TTT) and Products of Experts (POE). Our model achieves 96.1% accuracy on the training set and 21.7% accuracy on the evaluation set.
Cite
@article{arxiv.2605.01154,
title = {Multi-Perspective Transformers in ARC-AGI-2 Challenge},
author = {Caleb Talley and Vedant Tibrewal and Seun Adekunle and Weiwen Dong and Xinyu Wu and Fariha Sheikh},
journal= {arXiv preprint arXiv:2605.01154},
year = {2026}
}