Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Prompt

Documenti analoghi

• CryptPEFT: Efficient and Private Neural Network Inference via Parameter-Efficient Fine-Tuning
  di: Xia, Saisai, et al.
  Pubblicazione: (2025)
• TernaryVote: Differentially Private, Communication Efficient, and Byzantine Resilient Distributed Optimization on Heterogeneous Data
  di: Jin, Richeng, et al.
  Pubblicazione: (2024)
• CryptoTensors: A Light-Weight Large Language Model File Format for Highly-Secure Model Distribution
  di: Zhu, Huifeng, et al.
  Pubblicazione: (2025)
• HEQuant: Marrying Homomorphic Encryption and Quantization for Communication-Efficient Private Inference
  di: Xu, Tianshi, et al.
  Pubblicazione: (2024)
• UFO: Unlocking Ultra-Efficient Quantized Private Inference with Protocol and Algorithm Co-Optimization
  di: Zeng, Wenxuan, et al.
  Pubblicazione: (2026)