List your company

Synthseq on BabyAI

32.1Average Pass RatePoly-PPO w/ UCB
10.67616.23821.827.362Sep 29, 2025

Evaluation results

MethodLinks
Poly-PPO w/ UCBevaluation_protocol=pass@1, initial-state perturbations=large2025.0932.1
Polychromic PPOevaluation_protocol=pass@1, initial-state perturbations=large2025.0930.6
REINFORCE w/ UCBevaluation_protocol=pass@1, initial-state perturbations=large2025.0926.2
Pretrained policyevaluation_protocol=pass@1, initial-state perturbations=large2025.0920
REINFORCE w/ Baselineevaluation_protocol=pass@1, initial-state perturbations=large2025.0919.3
PPOevaluation_protocol=pass@1, initial-state perturbations=large2025.0916.6
PPO w/ UCBevaluation_protocol=pass@1, initial-state perturbations=large2025.0911.5