List your company

ResearchFormerly WizWand

CAA

Benchmarks

TaskDatasetSOTA resultTrendResults

Behavior Selection

CAA (50 random samples)

98Accuracy (coordinate-ais, pair)

22

Hallucination Steering

CAA

19.5Runtime

13

Concept Alignment

CAA (test)

75.18AIC

12

Open-ended behavior generation

CAA

5.33CoAIS Score

10

CAA

88Accuracy (pair)

8