List your company

Question Answering on MuSiQue (EM, F1, Trigger rate (T))

21.8Exact Match (EM)Verbal-Calibrate
-0.2485.47611.216.924Apr 7, 2026

Evaluation results

MethodLinks
Verbal-CalibrateScope=global2026.0421.828.876.8
Uncertain-CalibrateScope=local2026.0417.624.194.2
DRAGIN2026.041727.187
Base-Verbal2026.0410.817.513.6
Ret-All2026.049.817.4100
ADARAGUE2026.049.615.953.8
Base-UncTok2026.048.816.24.3
FLARE2026.046.413.899.8
No-Ret2026.045.610.3—
SR-13BBackbone=13B2026.040.86.139.8
SR-7BBackbone=7B2026.040.65.256