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

RoboFAC

Benchmarks

Task NameDataset NameSOTA ResultTrend
Robotic Task PerceptionRoboFAC real-robot
VOC Success Rate93.01
8
Robot Failure Analysis (MCQ)RoboFAC (Real-world)
FD96
7
Robot Failure Analysis (MCQ)RoboFAC Simulation
FD Score93
7
Robotic Failure AnalysisRoboFAC 1.0 (mixed simulated and real-world)
Task Success Rate (Short Horizon)82.74
6
Free-language reasoningRoboFAC (Real-world)
ROUGE-L (TI)33.8
4
Free-language reasoningRoboFAC Simulation
ROUGE-L (TI)32.6
4
Showing 6 of 6 rows