A local small language model can predict when a differentially private sanitized prompt will still yield useful LLM output, saving up to 20% of wasted API calls, and an exact-nearest-neighbor implementation of the dX-privacy mechanism mostly returns the original token.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Preempting Text Sanitization Utility in Resource-Constrained Privacy-Preserving LLM Interactions
A local small language model can predict when a differentially private sanitized prompt will still yield useful LLM output, saving up to 20% of wasted API calls, and an exact-nearest-neighbor implementation of the dX-privacy mechanism mostly returns the original token.