A neuro-symbolic pipeline combining object extraction, neural transformation proposals from a DSL, and symbolic consistency filtering raises LLM accuracy on ARC-AGI-2 from 16% to 24.4%.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Compositional Neuro-Symbolic Reasoning
A neuro-symbolic pipeline combining object extraction, neural transformation proposals from a DSL, and symbolic consistency filtering raises LLM accuracy on ARC-AGI-2 from 16% to 24.4%.