List your company
ResearchFormerly WizWand

Evaluation-based Bias Reduction on Bias Reduction Benchmark (Evaluation)

99.8Bias Reduction PerformanceCoT
-10.75217.94946.6575.351Nov 28, 2025
Updated 8mo ago

Evaluation results

MethodLinks
CoTModel=deepseek-r12025.1199.8
Counterfact.Model=deepseek-r12025.1199.8
ExplicitModel=deepseek-r12025.1199.6
CoTModel=qwen3-235b-a22b2025.1197.8
RDGModel=GPT-4o2025.1196.2
RDGModel=deepseek-r12025.1194.8
RDGModel=GPT-4o-mini2025.1194.7
RDGModel=Average2025.1194.3
RDGModel=qwen3-235b-a22b2025.1191.3
DDDModel=GPT-4o-mini2025.1190
DDDModel=qwen3-235b-a22b2025.1189.1
DDDModel=GPT-4o2025.1188.7
DDDModel=Average2025.1188.2
CoTModel=Average2025.1188.2
ExplicitModel=GPT-4o-mini2025.1185.3
DDDModel=deepseek-r12025.1185.1
CoTModel=GPT-4o-mini2025.1182.1
ExplicitModel=GPT-4o2025.1181.1
Counterfact.Model=Average2025.1174
Counterfact.Model=GPT-4o-mini2025.1173.7
CoTModel=GPT-4o2025.1173
Self-debiasModel=GPT-4o-mini2025.1169.8
Counterfact.Model=GPT-4o2025.1168
ExplicitModel=Average2025.1164.9
Self-debiasModel=GPT-4o2025.1164.7
Self-debiasModel=deepseek-r12025.1158.7
Self-debiasModel=Average2025.1155.4
Counterfact.Model=qwen3-235b-a22b2025.1154.3
CDDModel=GPT-4o-mini2025.1148.8
CDDModel=GPT-4o2025.1143.4
CDDModel=Average2025.1138.8
CDDModel=deepseek-r12025.1134.8
CDDModel=qwen3-235b-a22b2025.1128.3
Self-debiasModel=qwen3-235b-a22b2025.1128.3
ExplicitModel=qwen3-235b-a22b2025.11-6.5