Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Preference Bias Mitigation on Preference Bias
Loading...
85.83
RR
SAC Single-task
-3.4332
19.7409
42.915
66.0891
Nov 4, 2024
RR
Updated 4d ago
Evaluation Results
Method
Method
Links
RR
SAC Single-task
Backbone=Qwen2-72B-Ins...
2024.11
85.83
SAC Multi-task
Backbone=Qwen2-72B-Ins...
2024.11
80.83
SAC
Control Dim=Single
2024.11
62.5
SAC Multi-task
Backbone=Qwen2-7B-Inst...
2024.11
58.33
SAC
Control Dim=Multiple
2024.11
53.75
SAC Single-task
Backbone=Qwen2-7B-Inst...
2024.11
50.83
RepE
Control Dim=Single
2024.11
39.17
No Control
Control Dim=Single
2024.11
10.83
No Control
Control Dim=Multiple
2024.11
10.83
RepE-Merge
Control Dim=Multiple
2024.11
10
RepE-Mean
Control Dim=Multiple
2024.11
5
No Control
Backbone=Qwen2-7B-Inst...
2024.11
0
No Control
Backbone=Qwen2-72B-Ins...
2024.11
0
Feedback
Search any
task
Search any
task