List your company

Logical Coherence on CIAR

96AccuracyMetaCrit + GPT-4o
16.9637.485878.52Jul 20, 2025

Evaluation results

MethodLinks
MetaCrit + GPT-4oGroup=Ours, Base Model=GPT-4o, Monitoring agent (phi_up)=included, Control agent (phi_down)=included2025.0796
MetaCrit + GPT-3.5-TurboGroup=Ours, Base Model=GPT-3.5-Turbo, Monitoring agent (phi_up)=included, Control agent (phi_down)=included2025.0784
MEPGroup=Baselines, Base Model=GPT-3.5-Turbo2025.0782
MetaCrit + Claude-3.5-SonnetGroup=Ours, Base Model=Claude-3.5-Sonnet, Monitoring agent (phi_up)=included, Control agent (phi_down)=included2025.0774
MetaCrit + DeepSeek-v3Group=Ours, Base Model=DeepSeek-v3, Monitoring agent (phi_up)=included, Control agent (phi_down)=included2025.0770
MetaCrit w/o ϕ↑Group=Ablation, Base Model=GPT-3.5-Turbo, Monitoring agent (phi_up)=excluded, Control agent (phi_down)=included2025.0748
MetaCrit w/o ϕ↓Group=Ablation, Base Model=GPT-3.5-Turbo, Monitoring agent (phi_up)=included, Control agent (phi_down)=excluded2025.0742
ExpertPromptingGroup=Baselines, Base Model=GPT-3.5-Turbo2025.0738
MADGroup=Baselines, Base Model=GPT-3.5-Turbo2025.0736
Universal Self-consistencyGroup=Baselines, Base Model=GPT-3.5-Turbo2025.0730
Zero-shot-CoTGroup=Baselines, Base Model=GPT-3.5-Turbo2025.0724
Self-refineGroup=Baselines, Base Model=GPT-3.5-Turbo2025.0720