Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Representational Alignment on BabyAI instruction set
Loading...
46.65
P@10
Qwen
6.1108
16.6354
27.16
37.6846
Jan 30, 2026
P@10
P@1
P@5
P@15
Updated 5mo ago
Evaluation Results
Method
Method
Links
P@10
P@1
P@5
P@15
Qwen
Alignment Anchor=ALM
2026.01
46.65
-
-
67.93
LLaMA
Alignment Anchor=ALM
2026.01
46.65
-
-
69.73
BLIP
Alignment Anchor=ALM
2026.01
46.61
-
-
65.98
DeepSeek
Alignment Anchor=ALM
2026.01
46.2
-
-
70.22
BERT
Alignment Anchor=ALM
2026.01
27.04
-
-
35.36
CLIP
Alignment Anchor=ALM
2026.01
22.7
-
-
30.58
RANDOM
Alignment Anchor=ALM
2026.01
7.67
-
-
13.12
Feedback
Search any
task
Search any
task