Research map / Federated learning and privacy / Privacy attacks and secure learning
CrypTen: Secure Multi-Party Computation Meets Machine Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| run on GPU instead of CPU | CrypTen Wav2Letter inference · LibriSpeech 1s clips | inference time: 1-2 orders of magnitude faster | Helped |
| increase batch size | CrypTen text classification · Yelp | time and communication per sample | Helped |
| increase batch size | CrypTen ResNet-18 / ViT · ImageNet | bytes communicated per sample | No clear effect |
| use secure MPC instead of plaintext PyTorch | inference · text and Wav2Letter | inference time: about 2.5-3 orders of magnitude slower | Hurt |
| increase number of parties from two to three | CrypTen · text classification | communication rounds | Hurt |
Closest papers by meaning
- Secure Quantized Training for Deep Learning (ICML 2022 · similarity 0.86)
- United We Stand: Accelerating Privacy-Preserving Neural Inference by Conjunctive Optimization with Interleaved Nexus (AAAI 2024 · similarity 0.83)
- CryptoMoE: Privacy-Preserving and Scalable Mixture of Experts Inference via Balanced Expert Routing (NeurIPS 2025 · similarity 0.83)
- Ditto: Quantization-aware Secure Inference of Transformers upon MPC (ICML 2024 · similarity 0.83)
- Privacy Auditing of Large Language Models (ICLR 2025 · similarity 0.82)
- An Efficient Private GPT Never Autoregressively Decodes (ICML 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.