Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Logical Reasoning on BBH (Accuracy, Loss)
Loading...
52.5
Exact Match Accuracy (BBH Logical Reasoning)
Dream-7B-Base + DyStruct
43.348
45.724
48.1
50.476
May 10, 2026
Exact Match Accuracy (BBH Logical Reasoning)
Accuracy (BBH Logical Reasoning)
Loss (BBH Logical Reasoning)
Updated 22d ago
Evaluation Results
Method
Method
Links
Exact Match Accuracy (BBH Logical Reasoning)
Accuracy (BBH Logical Reasoning)
Loss (BBH Logical Reasoning)
Dream-7B-Base + DyStruct
Backbone=Dream-7B, Dec...
2026.05
52.5
-
-
Dream-7B-Base
Backbone=Dream-7B, Dec...
2026.05
51.7
-
-
LLaDA-8B-Base + DyStruct
Backbone=LLaDA-8B, Dec...
2026.05
49.3
-
-
LLaDA-8B-Base
Backbone=LLaDA-8B, Dec...
2026.05
44.9
-
-
Dream-7B-Base + DAEDAL
Backbone=Dream-7B, Dec...
2026.05
44.8
-
-
LLaDA-8B-Base + DAEDAL
Backbone=LLaDA-8B, Dec...
2026.05
43.7
-
-
Adam
Model scale=1B
2026.04
-
24.8
1.7
Nexus
Model scale=1B
2026.04
-
23
1.689
Adam
Model scale=3B
2026.04
-
47.4
1.529
Nexus
Model scale=3B
2026.04
-
44.4
1.54
Feedback
Search any
task
Search any
task