Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

ACRE

Benchmarks

Task NameDataset NameSOTA ResultTrend
Abstract Causal ReasoningACRE Systematicity
Query Accuracy67.44
10
Abstract Causal ReasoningACRE Compositionality
Query Accuracy69.04
10
Abstract Causal ReasoningACRE (I.I.D.)
Query Accuracy66.29
10
Visual ReasoningACRE 1.0 (systematicity)
Accuracy93.9
8
Visual ReasoningACRE compositionality 1.0
Accuracy93.27
8
Visual ReasoningACRE I.I.D. 1.0
Accuracy92.98
7
Success RateACRE
Success Rate93
4
Causal InductionACRE (Systematic)
Overall Score93.9
3
Causal InductionACRE Compositional
All91.76
3
Showing 9 of 9 rows