Dywave applies wavelet-based hierarchical decomposition to build dynamic, event-aligned tokens for heterogeneous IoT signals, cutting token length by up to 75% while raising accuracy up to 12% on sequence models.
Frequency-domain mlps are more effective learners in time series forecasting
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Dywave: Event-Aligned Dynamic Tokenization for Heterogeneous IoT Sensing Signal
Dywave applies wavelet-based hierarchical decomposition to build dynamic, event-aligned tokens for heterogeneous IoT signals, cutting token length by up to 75% while raising accuracy up to 12% on sequence models.