SeqPA TE: Differentially Private Text Generation via Knowledge Distillation

Aug-14-2025, 15:08:13 GMT–Neural Information Processing Systems

P A TE is a recent DP learning algorithm that achieves high utility with strong privacy protection on training samples. However, text generation models output tokens sequentially in a large output space; the classic P A TE algorithm is not customized for this setting.

artificial intelligence, machine learning, natural language, (20 more...)

Neural Information Processing Systems

Aug-14-2025, 15:08:13 GMT

Conferences PDF

Add feedback

Country:
- Asia
  - China > Hong Kong (0.04)
  - Japan > Honshū
    - Kantō > Tokyo Metropolis Prefecture > Tokyo (0.04)
- North America > United States
  - Hawaii (0.04)

Genre:
- Research Report > New Finding (0.68)

Industry:
- Information Technology > Security & Privacy (1.00)

Technology:
- Information Technology
  - Artificial Intelligence
    - Machine Learning
      - Neural Networks > Deep Learning (0.69)
      - Statistical Learning (0.67)
    - Natural Language (1.00)
  - Security & Privacy (1.00)