Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Embodied Task Execution on VLABench Commonsense
Loading...
57.3
Progress Score (PS)
π0.5
37.02
42.285
47.55
52.815
Jun 2, 2026
Progress Score (PS)
Success Rate (SR)
Instruction Score (IS)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Progress Score (PS)
Success Rate (SR)
Instruction Score (IS)
π0.5
mode=Zero-shot
2026.06
57.3
43.9
60
ERVLA
mode=Zero-shot
2026.06
55
44
57.2
Choice + No Knowledge Truncation
Ablation=Choice withou...
2026.06
52.6
43
56.4
No Choice + Knowledge Insulation
Ablation=No choice and...
2026.06
50
41.2
55
π0-FAST
mode=Zero-shot
2026.06
48.6
38
56.8
X-VLA
mode=Zero-shot
2026.06
48.2
-
-
No CoT
Ablation=No explicit c...
2026.06
47.8
38.8
54
No Choice (End-to-End)
Ablation=End-to-End No...
2026.06
45.4
36.4
52.8
π0
mode=Zero-shot
2026.06
43
29.1
54.9
ACoT-VLA
mode=Zero-shot
2026.06
37.8
-
52.3
Feedback
Search any
task
Search any
task