DIO-Agent frames IO2Code as LLM-driven evolutionary search over programs with a Transformation Priority Premise to favor simple hypotheses, outperforming baselines on a new IO2CodeBench.
Once the evolved code successfully passes all existing examples, the model is permitted to generate additional I/O pairs, increasing the sample count by 2 in each subsequent step
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From I/O to Code with Discovery Agent
DIO-Agent frames IO2Code as LLM-driven evolutionary search over programs with a Transformation Priority Premise to favor simple hypotheses, outperforming baselines on a new IO2CodeBench.