Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Coffee Task on MimicGen
Loading...
99.2
Success Rate
Ours IL
36.592
52.846
69.1
85.354
Sep 24, 2025
Success Rate
Updated 1d ago
Evaluation Results
Method
Method
Links
Success Rate
Ours IL
Number of demos=1000
2025.09
99.2
Ours Ens
Number of demos=1000
2025.09
98.8
Ours Ens
Number of demos=200
2025.09
96.4
Ours Ens
Number of demos=500
2025.09
96.4
MimicGen
Number of demos=1000
2025.09
96.4
Ours IL
Number of demos=200
2025.09
96
MimicGen
Number of demos=500
2025.09
95.6
Ours IL
Number of demos=500
2025.09
95.2
Ours Ens
Number of demos=50
2025.09
93.6
Ours Ens
Number of demos=100
2025.09
93.2
MimicGen
Number of demos=100
2025.09
92.4
MimicGen
Number of demos=200
2025.09
92.4
Ours IL
Number of demos=100
2025.09
89.6
LLM Trainer
Optimization=Best Anno...
2025.09
89
Ours IL
Number of demos=50
2025.09
86.4
MimicGen
Number of demos=50
2025.09
84
LLM Trainer
Optimization=Total
2025.09
82.6
MimicGen
Config=Baseline
2025.09
78.2
LLM Trainer
Optimization=No RL
2025.09
39
Feedback
Search any
task
Search any
task