Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Headline Generation on Tibetan Headline Generation (HG)
Loading...
0.3095
General Metric Value
Strong SFT
0.25074
0.265995
0.28125
0.296505
May 14, 2026
General Metric Value
Similarity Score
CMRC Average Score
CMRC F1 Score
LLM-Judge Win Rate
Updated 2mo ago
Evaluation Results
Method
Method
Links
General Metric Value
Similarity Score
CMRC Average Score
CMRC F1 Score
LLM-Judge Win Rate
Strong SFT
2026.05
0.3095
64.99
44.2
65.3
35.1
RL (Ours)
2026.05
0.253
64.04
45.1
65.2
51.2
Feedback
Search any
task
Search any
task