We run a controlled compositional generalization experiment in the ARC-AGI domain: an open-world problem domain in which the ability to generalize out-of-distribution is, by design, an essential characteristic for success. We compare neural program synthesis and test-time fine-tuning approaches on this experiment. We find that execution-guided neural program synthesis outperforms all reference algorithms in its ability to compose novel solutions. Our empirical findings also suggest that the success of TTFT on ARC-AGI lies mainly in eliciting in-distribution knowledge that the LLM otherwise fails to rely on directly.
@article{arxiv.2507.15877,
title = {Out-of-Distribution Generalization in the ARC-AGI Domain: Comparing Execution-Guided Neural Program Synthesis and Test-Time Fine-Tuning},
author = {Simon Ouellette},
journal= {arXiv preprint arXiv:2507.15877},
year = {2025}
}
Comments
this version fixes errors in AlphaEvolve total % calculation, Table 3 DSL description, and adds clarifications in response to review criticisms