FISHER: A Foundation Model for Multimodal Industrial Signal Comprehensive Representation

Industrial signal analysis is hindered by severe data heterogeneity, which we characterize as the M5 problem. Existing solutions rely on specialized models that lack robustness and scalability, while large-scale pretraining has rarely been investigated in this area. In this work, we derive a prioritized roadmap for the M5 problem and propose FISHER, aFoundation model for multimodalIndustrialSignal compreHEnsiveRepresentation. To address the foremost multisampling-rate problem, FISHER utilizes a novel sub-band modeling approach that treats sampling rate increments as concatenated sub-band information, enabling the adaptive usage of full signal bandwidth without resampling. FISHER is pretrained by teacher–student self-distillation over external audio and music data. We also establish the RMIS benchmark, comprising 19 datasets across four modalities. In the experiment, FISHER outperforms 24 state-of-the-art series encoders (up to 2B) with much smaller sizes (up to 16×), showcasing groundbreaking diagnostic accuracy and remarkable versatility. We further demonstrate that 1) seamless adaptation to variable sampling rates is the key to generalization 2) audio and music data provide better temporal variability, which is essential for pretraining. Both FISHER and RMIS are open-sourced.