Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Language Modeling on Pubmed (val)
Loading...
36.5
Perplexity
LoopMDM
35.3492
43.1171
50.885
58.6529
Jun 2, 2025
Jul 31, 2025
Sep 29, 2025
Nov 27, 2025
Jan 26, 2026
Mar 26, 2026
May 25, 2026
Perplexity
Updated 8d ago
Evaluation Results
Method
Method
Links
Perplexity
LoopMDM
S (inference-time loop...
2026.05
36.5
LoopMDM
S (inference-time loop...
2026.05
37.2
DCDM
Param=0.1B, Training T...
2026.05
40.17
BDLM
Param=0.1B, Training T...
2026.05
42.52
MDM
Zero-shot protocol=tru...
2026.05
43
MDLM
Param=0.1B, Training T...
2026.05
43.13
BD3-LM
Training steps=250K, T...
2025.06
43.41
MDLM
Training steps=250K, T...
2025.06
46.51
SEDD Absorb
Training steps=250K, T...
2025.06
47.07
LoopMDM
S (inference-time loop...
2026.05
47.9
AR
Param=0.1B, Evaluation...
2026.05
48.59
AR Transformer
Training steps=250K, T...
2025.06
49.49
Eso-LMs
Training steps=250K, T...
2025.06
60.2
Eso-LMs
Training steps=250K, T...
2025.06
62.37
Eso-LMs
Training steps=250K, T...
2025.06
62.45
Eso-LMs
Training steps=250K, T...
2025.06
65.27
Feedback
Search any
task
Search any
task
