Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Abstention on AbstentionBench Clean-merged (test)
Loading...
92.7
Precision
SFT+RL
88.124
89.312
90.5
91.688
May 30, 2026
Precision
Recall
F1 Score
Updated 1d ago
Evaluation Results
Method
Method
Links
Precision
Recall
F1 Score
SFT+RL
Model Size=14B, Prompt...
2026.05
92.7
51.3
66
DPO+RL
Model Size=14B, Prompt...
2026.05
92.2
54.9
68.8
Abs-RL
Model Size=14B, Prompt...
2026.05
92.1
53.5
67.7
CPT+RL
Model Size=14B, Prompt...
2026.05
92.1
58.6
71.6
SFT-80K+RL
Model Size=14B, Prompt...
2026.05
92
56.1
69.7
SFT+RL
Model Size=14B, Prompt...
2026.05
91.9
56.4
69.9
DPO+RL
Model Size=14B, Prompt...
2026.05
91.6
59.2
71.9
Abs-RL
Model Size=14B, Prompt...
2026.05
90.2
61.7
73.3
SFT-80K+RL
Model Size=14B, Prompt...
2026.05
88.5
68.4
77.2
CPT+RL
Model Size=14B, Prompt...
2026.05
88.3
70.1
78.1
Feedback
Search any
task
Search any
task