Differential Privacy, Linguistic Fairness, and Training Data Influence: Impossibility and Possibility Theorems for Multilingual Language Models.
Phillip RustAnders SøgaardPublished in: CoRR (2023)
Keyphrases
- language model
- differential privacy
- training data
- differentially private
- language modeling
- privacy preserving
- document retrieval
- information retrieval
- probabilistic model
- speech recognition
- cross lingual
- query expansion
- n gram
- privacy preservation
- retrieval model
- data sets
- test collection
- data sharing
- language models for information retrieval
- training set
- query terms
- natural language processing
- decision trees
- cross language
- digital libraries
- smoothing methods
- data privacy
- natural language