SHIP, a semantic hierarchical prompt-tuning method, raises VTAB-1k average accuracy for ViT-B/16 from 72.0% (VPT) to 76.9% with 0.38M trainable parameters.
Learning transferable visual models from natural language supervi- sion,
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
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Semantic Hierarchical Prompt Tuning for Parameter-Efficient Fine-Tuning
SHIP, a semantic hierarchical prompt-tuning method, raises VTAB-1k average accuracy for ViT-B/16 from 72.0% (VPT) to 76.9% with 0.38M trainable parameters.