BenchHAR finds that hybrid reconstruction-plus-contrastive SSL with CNN encoders generalizes best for sensor HAR but overall performance on unseen distributions remains unsatisfactory.
One fits all: Power general time series analysis by pretrained lm.Advances in neural information processing systems, 36:43322–43355
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
BenchHAR: Benchmarking Self-Supervised Learning for Generalizable Sensor-based Activity Recognition
BenchHAR finds that hybrid reconstruction-plus-contrastive SSL with CNN encoders generalizes best for sensor HAR but overall performance on unseen distributions remains unsatisfactory.