Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on TruthfulQA (BLEU and ROUGE Scores)
Loading...
53.37
BLEU Score
Base
33.246
38.4705
43.695
48.9195
Feb 9, 2026
BLEU Score
ROUGE-1 Score
ROUGE-2 Score
ROUGE-L Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
BLEU Score
ROUGE-1 Score
ROUGE-2 Score
ROUGE-L Score
Base
Model=Qwen2.5-14B
2026.02
53.37
54.35
44.68
51.41
CoIPO
Model=Qwen2.5-14B
2026.02
51.41
53.98
43.57
51.16
Base
Model=Qwen2.5-72B
2026.02
50.18
51.29
45.41
49.69
CoIPO
Model=Qwen2.5-72B
2026.02
50.18
51.65
44.31
48.59
CoIPO
Model=Qwen2.5-7B
2026.02
47.61
48.59
41.13
46.51
Base
Model=Qwen2.5-7B
2026.02
46.76
51.41
42.23
46.63
Base
Model=Llama-7B
2026.02
34.14
32.43
25.82
30.96
CoIPO
Model=Llama-7B
2026.02
34.02
32.92
28.02
31.57
Feedback
Search any
task
Search any
task