Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Multi-turn conversation on MT-Bench (Accuracy)

82.7Accuracy

C2PO

Updated 1mo ago

Evaluation Results

Method	Links
C2PO 2025.12		82.7
GPT-4 2025.12		65.3

SOTA Paper

C2PO

C2PO: Diagnosing and Disentangling Bias Shortcuts in LLMs

Dataset

MT-Bench

Follow for update

@wizwand_team Discord

Related Benchmarks

Stereotypical Bias Mitigation on Unqover Structural Bias Evaluation on MNLI Structural Bias Evaluation on HANS General Utility Evaluation on Chatbot General Utility Evaluation on MT_Bench

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs