Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Embodied Task Execution on VLABench Cross Category
Loading...
61
Primitive Success (PS)
ERVLA
23.664
33.357
43.05
52.743
Jun 2, 2026
Primitive Success (PS)
Success Rate (SR)
Instance Success (IS)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Primitive Success (PS)
Success Rate (SR)
Instance Success (IS)
ERVLA
mode=Zero-shot
2026.06
61
47
66.4
Choice + No Knowledge Truncation
Ablation=Choice withou...
2026.06
55.2
42.4
59.4
π0.5
mode=Zero-shot
2026.06
49.7
38.2
52
No Choice + Knowledge Insulation
Ablation=No choice and...
2026.06
47.6
36
53.8
No CoT
Ablation=No explicit c...
2026.06
44.8
34.8
51.6
No Choice (End-to-End)
Ablation=End-to-End No...
2026.06
42
33.6
49.6
ACoT-VLA
mode=Zero-shot
2026.06
38.9
-
54.1
π0-FAST
mode=Zero-shot
2026.06
38.2
31
47.8
π0
mode=Zero-shot
2026.06
33.6
21.2
44
X-VLA
mode=Zero-shot
2026.06
25.1
-
-
Feedback
Search any
task
Search any
task