Research map / Federated learning and privacy / Privacy attacks and secure learning
Inference-Based Similarity Search in Randomized Montgomery Domains for Privacy-Preserving Biometric Identification
Yi Wang, Jianwu Wan, Jun Qing Guo, Yiu‐ming Cheung, Pong Chi Yuen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| use nested Montgomery form instead of Paillier encryption | MIMP-RM index encoding · UBIRIS encoding cost | encoding time (s): 168.98 -> 1.01 | Improved |
| use nested Montgomery form instead of Paillier encryption | MIMP-RM scanning · UBIRIS scanning cost | scanning time (s): 301.03 -> 0.0045 | Improved |
| use nested Montgomery form instead of SHA-256 | MIMP-RM index · UBIRIS ciphertext size | size (MB): 3 -> 0.155 | Improved |
| increase substring count L from 25 to 50 (ablation) | MIMP-RM · UBIRIS, LFW, FERET | hit rate | Improved |
| increase substring count L from 50 to 75 (ablation) | MIMP-RM · UBIRIS, LFW, FERET | hit rate | No clear effect |
| increase substring count L from 75 to 100 (ablation) | MIMP-RM · UBIRIS, LFW, FERET | hit rate | Hurt |
| spread mismatching bits more evenly (smaller i) (ablation) | MIMP-RM · Simulated 2800-bit strings | precision and recall | Improved |
| use MIMP-RM instead of LSH + partial distance | privacy-preserving search · UBIRIS, LFW, FERET | hit rate | Improved |
| use MIMP-RM instead of exhaustive Hamming search | privacy-preserving search · UBIRIS, LFW, FERET | hit rate | No clear effect |
Closest papers by meaning
- Efficiently Computing Similarities to Private Datasets (ICLR 2024 · similarity 0.84)
- Instance-Specific Asymmetric Sensitivity in Differential Privacy (NeurIPS 2024 · similarity 0.83)
- A Unified Framework for Adversary-Aware Differential Privacy Bounds (NeurIPS 2026 · similarity 0.82)
- Learning Robust and Privacy-Preserving Representations via Information Theory (AAAI 2025 · similarity 0.82)
- Privacy-Preserving Embedding via Look-up Table Evaluation with Fully Homomorphic Encryption (ICML 2024 · similarity 0.82)
- Near-Exact Privacy Amplification for Matrix Mechanisms (ICLR 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.
