Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Language Modeling on PubMed OOD from The Pile
Loading...
12.61
Perplexity
ASEntmax
12.3852
13.9026
15.42
16.9374
Jun 19, 2025
Perplexity
Updated 1mo ago
Evaluation Results
Method
Method
Links
Perplexity
ASEntmax
Context Length=8K
2025.06
12.61
Entmax
Context Length=8K
2025.06
12.86
ASEntmax
Context Length=16K
2025.06
12.9
Entmax
Context Length=16K
2025.06
13.02
SSMax
Context Length=8K
2025.06
13.75
SSMax
Context Length=16K
2025.06
14.72
ASEntmax
Context Length=4K
2025.06
14.76
Entmax
Context Length=4K
2025.06
14.79
SSMax
Context Length=4K
2025.06
15.14
Softmax
Context Length=8K
2025.06
15.31
Softmax
Context Length=4K
2025.06
15.59
Softmax
Context Length=16K
2025.06
18.23
Feedback
Search any
task
Search any
task