List your company

Critical Error Detection on WMT 21 (test)

0.51MCCGPT-4o-mini
0.3540.39450.4350.4755Feb 11, 2026

Evaluation results

MethodLinks
GPT-4o-miniFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.020.516585
LLaMA-3.1-8BFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.020.435885
GPT-OSS-20BFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.020.365674