Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Stereotypical Bias Mitigation on Unqover
Loading...
99.9
Accuracy
GRPO
20.236
40.918
61.6
82.282
Dec 29, 2025
Accuracy
Bias Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Bias Score
GRPO
Backbone=DeepSeek-R1-D...
2025.12
99.9
0.2
C2PO
Backbone=DeepSeek-R1-D...
2025.12
99.9
0.2
CPO
Backbone=DeepSeek-R1-D...
2025.12
99.7
1.9
IPO
Backbone=DeepSeek-R1-D...
2025.12
99.6
0.7
BCO
Backbone=DeepSeek-R1-D...
2025.12
96.3
5.1
FR
Backbone=DeepSeek-R1-D...
2025.12
94.9
7
DPO
Backbone=DeepSeek-R1-D...
2025.12
93.3
11.6
C2PO
Backbone=LLaMA-2-13B-Chat
2025.12
91.8
3
CPO
Backbone=LLaMA-2-13B-Chat
2025.12
85.9
8.8
BCO
Backbone=LLaMA-2-13B-Chat
2025.12
85.3
9.4
IPO
Backbone=LLaMA-2-13B-Chat
2025.12
83.3
10.9
GRPO
Backbone=LLaMA-2-13B-Chat
2025.12
36.3
17.4
FR
Backbone=LLaMA-2-13B-Chat
2025.12
26.4
7.1
DPO
Backbone=LLaMA-2-13B-Chat
2025.12
23.3
7.4
Feedback
Search any
task
Search any
task
