SALT is a residual adapter placed between frozen head and tail networks in split computing that improves user-specific classification accuracy and packet-loss robustness without modifying the proprietary model.
Tinytl: Reduce memory, not parameters for efficient on-device learning,
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
support 1representative citing papers
citing papers explorer
-
SALT: A Lightweight Model Adaptation Method for Closed Split Computing Environments
SALT is a residual adapter placed between frozen head and tail networks in split computing that improves user-specific classification accuracy and packet-loss robustness without modifying the proprietary model.