POST tunes a soft prompt privately on a small distilled model and transfers it to the large model using a two-term public-data loss, with optional differential privacy.
αce αlm αcos Learning Rate Batch Size 5.0 2.0 1.0 0.00025 5 B.2
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Efficient and Privacy-Preserving Soft Prompt Transfer for LLMs
POST tunes a soft prompt privately on a small distilled model and transfers it to the large model using a two-term public-data loss, with optional differential privacy.