Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Personalized news headline generation on LaMP-4 (test)
Loading...
40
Alignment Win (%)
DPS
33.76
35.38
37
38.62
Apr 24, 2026
Alignment Win (%)
Relevance
Fluency
Style
Alignment
Factuality
Updated 1mo ago
Evaluation Results
Method
Method
Links
Alignment Win (%)
Relevance
Fluency
Style
Alignment
Factuality
DPS
Evaluator=Human
2026.04
40
-
-
-
-
-
CAD
Evaluator=Human
2026.04
34
-
-
-
-
-
CAD
Evaluator=GPT-5.2
2026.04
-
3.97
4.51
3.62
3.63
4.08
DPS
Evaluator=GPT-5.2
2026.04
-
4.45
4.83
3.91
3.93
4.21
Feedback
Search any
task
Search any
task