List your company

Machine Translation Meta-evaluation on WMT En-De Metrics Shared Task Segment-Level 2023 (test)

57.4Accuracy (Test)COMET
46.68849.46952.2555.031Jan 12, 2026

Evaluation results

MethodLinks
COMETType=Reference-based Metrics2026.0157.40.432
M-MADType=LLM-as-a-Judge2026.0155.50.552
RATEType=LLM-as-a-Judge2026.0152.30.389
BLEUType=Reference-based Metrics2026.01520.192
GEMBA-MQMType=LLM-as-a-Judge2026.0147.40.429
EAPromptType=LLM-as-a-Judge2026.0147.10.52