Arrow is a pretrained transformer that discovers causal graphs from observational data by factorizing them into skeletons and orders, trained end-to-end on diverse synthetic examples to match or exceed prior methods at lower inference cost.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
CoOD decomposes inputs into components and applies Component Shift Score plus Compositional Consistency Score to improve detection of both standard and compositional out-of-distribution data.
citing papers explorer
-
Arrow: A Foundation Model for Causal Discovery
Arrow is a pretrained transformer that discovers causal graphs from observational data by factorizing them into skeletons and orders, trained end-to-end on diverse synthetic examples to match or exceed prior methods at lower inference cost.
-
Component-Based Out-of-Distribution Detection
CoOD decomposes inputs into components and applies Component Shift Score plus Compositional Consistency Score to improve detection of both standard and compositional out-of-distribution data.