An n-dimensional numerical Transformer with linear embedding, bin-based discretization, and parallel output heads improves human activity recognition accuracy by 10-15% over a tokenized vanilla Transformer.
Leveraging large language models for activity recognition in smart environments
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.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Self-supervised Learning Method Using Transformer for Multi-dimensional Sensor Data Processing
An n-dimensional numerical Transformer with linear embedding, bin-based discretization, and parallel output heads improves human activity recognition accuracy by 10-15% over a tokenized vanilla Transformer.