Open access
Sep 2026
TrustMamba: calibrated uncertainty quantification for reliable and label-efficient bearing fault diagnosis
TrustMamba is proposed, a framework that reports calibrated uncertainty as part of the diagnostic output, and retains the lowest calibration error among competing methods under low signal-to-noise measurement conditions down to 0 dB.
Jun-Chi Xu, Lu Liu, Guo-Dong You et al.
· Measurement science and tech... · 0 citations