What’s new
2024.8
- Paper “Coral: Maliciously Secure Computation Framework for Packed and Mixed Circuits” [paper] [code] accepted by ACM CCS 2024.
- Paper “Sublinear Distributed Product Checks on Replicated Secret-Shared Data over without Ring Extensions” [paper] [code] accepted by ACM CCS 2024.
- Attended USENIX Security 2024 in Philadelphia as a silver sponsor. Pictures of our colleagues at the sponsor’s desk: pic1.
2024.7
- Paper “BumbleBee: Secure Two-party Inference Framework for Large Transformers” [paper] [code] was accepted by NDSS 2025.
2024.5
2023.12
- Attended ACM CCS 2023 in Copenhagen as a platinum sponsor. Picture of our colleagues at the table of Ant Research.
2023.8
Attended USENIX Security 2023 in Anaheim. Pictures of our colleagues presenting: pic1 and pic2.
Paper Accelerating Secure Collaborative Machine Learning with Protocol-Aware RDMA was accepted by USENIX Security 2024.
2023.5
- Paper “Efficient 3PC for Binary Circuits with Application to Maliciously-Secure DNN Inference” [paper] [code] was accepted by USENIX Security 2023.
- Attended SP 2023 in San Francisco.
2023.4
Paper “CHAM：A Customized Homomorphic Encryption Accelerator for Fast Matrix-Vector Product”[paper][code] was accepted by DAC 2023.
An Attack on RIAC Homomorphic Encryption Algorithm was published in Journal of Cryptologic Research (in Chinese). It describes how we broke the RIAC (Random iterative affine cipher) in 2022.1.
2023.3
- Gave a talk at the 6th HomomorphicEncryption.org Standards Meeting in Seoul, South Korea.
2023.2
- Paper “Squirrel: A Scalable Secure Two-Party Computation Framework for Training Gradient Boosting Decision Tree” [paper] [code] was accepted by USENIX Security 2023.
2023.1
- We have moved to Ant Group!
Prior works at Alibaba Gemini Lab
2022.8
We just deliver a major update to TF Encrypted, including benchmarks on several PPML operations such as NN training and inference. Please check the new performance section.
2022.3
The Microsoft SEAL homomorphic encryption library released version 4.0, including the BGV implementation we contributed.
2022.2
Paper Cheetah:Lean and Fast Secure Two-Party Deep Neural Network Inference accepted by USENIX’Sec 2022.
2022.1
We found that the RIAC (Random iterative affine cipher) used in several federated learning frameworks is broken. Please immediately stop using RIAC and check previous RIAC ciphertexts if there’s any. We have contacted the affected community and will publish the exploit details once we are allowed.
2021.12
Paper More Efficient Secure Matrix Multiplication for Unbalanced Recommender Systems accepted by IEEE TDSC.
2021.11
Standard IEEE Recommended Practice for Secure Multi-Party Computation is published! Thanks for everyone contributed in the project.
2021.5
Paper When Homomorphic Encryption Marries Secret Sharing accepted by SIGKDD2021.
2020.12
Paper Pegasus: bridging polynomial and Non-polynomial evaluations in Homomorphic Encryption accepted by S&P2021.
We got the second place in Track I of iDASH2020 secure genome analysis competition.
2020.9
Paper Improving Utility and Security of the Shuffler-based Differential Privacy accepted by VLDB2021.
Paper Faster Secure Multiparty Computation of Adaptive Gradient Descent accepted by the PPMLP workshop in Conjunction with CCS2020.
Paper hPRESS: A Hardware-enhanced Proxy Re-encryption Scheme using Secure Enclave accepted by IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD).
2020.8
Giving a talk about Federated learning and Secure Multi-party Computation in ISC2020.
2020.2
Paper Privacy-preserving collaborative machine learning on genomic data using TensorFlow accepted to the Trustworthy ML Workshop co-located with ICLR 2020. Update: We received the best paper award !
2020.1
Paper “Secure Social Recommendation based on Secret Sharing” accepted by ECAI2020.
Paper “HomoPAI: A Secure Collaborative Machine Learning Platform based on Homomorphic Encryption” accepted by ICDE2020 demo track. Watch the demo video on 优酷 or YouTube.
2019.12
We are named the first place in Track IV of iDASH2019 secure genome analysis competition.
We have open-sourced our competition code to TF-Encrypted and received their official acknowledgements !
Attended NeurIPS2019.
2019.11
Invited as panelist to 2019 Westlake International Forum on Cyber Security Research.
Joined MPC Alliance.
2019.10
Short paper Quantification of the Leakage in Federated Learning accepted by the Workshop on Federated Learning for Data Privacy and Confidentiality in Conjunction with NeurIPS 2019 (FL-NeurIPS-19).
Paper DPSAaS: Multi-Dimensional Data Sharing and Analytics as Services under Local Differential Privacy accepted by VLDB2019 demo track.
2019.9
Established working group P2842 - Recommended Practice for Secure Multi-party Computation in IEEE Standards Association.
Established working group Technical framework and application for secure multi-party computation in ITU-T.
2019.8
Attended the 4th HomomorphicEncryption.org Standards Meeting.
Attended USENIX Sec’2019.
2019.3
Paper Answering multi-dimensional analytical queries under local differential privacy accepted by SIGMOD2019.
2019.1
Paper Covert Security with Public Verifiability: Faster, Leaner, and Simpler accepted by EUROCRYPT2019.
2018.11
Invited as panelist to 2018 Westlake International Forum on Cyber Security Research .
2018.10
Attended ACM CCS 2018.
2018.5
Attended TPMPC2018.
Witnessed the inaugural ceremony of the AZFT Cyberspace Security Lab.
2017.11
Attended ACM CCS 2017.