List your company

Critical Error Detection on WMT 22 (test)

0.92MCCGPT-4o-mini
0.36880.51190.6550.7981Feb 11, 2026

Evaluation results

MethodLinks
GPT-4o-miniFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.020.929199
LLaMA-3.1-8BFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.020.878799
GPT-OSS-20BFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.020.393991