Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ReactorX arm

Benchmarks

Task NameDataset NameSOTA ResultTrend
Rollout-level failure detectionReactorX arm SmolVLA policy rollouts
ROC-AUC0.82
7
Rollout-level failure detectionReactorX arm π0.5 policy rollouts
ROC-AUC0.87
7
Rollout-level failure detectionReactorX arm ACT policy rollouts
ROC-AUC0.93
7
Showing 3 of 3 rows