List your company

System-level Speech Translation Quality Estimation on IWSLT Metrics Shared Task 2026 (test)

95.1Soft Pairwise Accuracy (en-de)HydraQE (all heads avg.)
85.32487.86290.492.938Jun 7, 2026

Evaluation results

MethodLinks
HydraQE (all heads avg.)2026.0695.198.496.8
HydraQE (xCOMET head)2026.0694.898.796.7
HydraQE (DA head)2026.0694.798.196.4
HydraQE (primary)2026.0694.597.996.2
HydraQE (MetricX head)2026.0694.297.195.7
CometKiwisource=organizers' results2026.0693.197.295.2
SpeechQE2026.0689.896.192.9
BLASER 2.0 QE2026.0688.888.288.5
COMET (partial)2026.0685.778.281.9