A frozen transformer is fine-tuned by adding a sparse dictionary of feature atoms to each layer's output, enabling atom-level control that improves image editing and concept customization.
Towards monosemanticity: De- composing language models with dictionary learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Sparse Fine-Tuning of Transformers for Generative Tasks
A frozen transformer is fine-tuned by adding a sparse dictionary of feature atoms to each layer's output, enabling atom-level control that improves image editing and concept customization.