BIAN, Song 
I am a tenure-track associate professor at the School of Cyber Science and Technology of Beihang University. My current research interests include applied cryptography, domain-specific accelerators, and AI security.
I was an assistant professor at Kyoto University from 2019 to 2021. I received Ph.D. and M.Sc in Informatics from Kyoto University in 2019 and 2017, respectively. I graduated from the University of Wisconsin-Madison with a B.Sc in Computer Engineering in 2014.
Selected Publications
Libra: Pattern-Scheduling Co-Optimization for Cross-Scheme FHE Code Generation over GPGPU
High-Precision Functional Bootstrapping for CKKS from Fourier Extension
Kangaroo: A Private and Amortized Inference Framework over WAN for Large-Scale Decision Tree Evaluation
cwPSU: Efficient Unbalanced Private Set Union via Constant-weight Codes
TensorFHE+: Fully Homomorphic Encryption Acceleration based on Linear Algebra
Engorgio: An Arbitrary-Precision Unbounded-Size Hybrid Encrypted Database via Quantized Fully Homomorphic Encryption
Presto: A Unified RISC-V-Compatible SoC for Multi-Scheme FHE Acceleration over Module Lattice
Enabling Energy-Efficient Homomorphic Encryption Evaluation via eDRAM-Based In-Situ Computing in an Edge Processor
FHECAP: An Encrypted Control System With Piecewise Continuous Actuation
An eDRAM-Based In-Situ-Computing Processor for Homomorphic Encryption Evaluation on the Edge
PPGNN: Fast and Accurate Privacy-Preserving Graph Neural Network Inference via Parallel and Pipelined Arithmetic-and-Logic FHE Accelerator
Alchemist: A Unified Accelerator Architecture for Cross-Scheme Fully Homomorphic Encryption
ESC-NTT: An Elastic, Seamless and Compact Architecture for Multi-Parameter NTT Acceleration
THE-V: Verifiable Privacy-Preserving Neural Network via Trusted Homomorphic Execution
PIMA-LPN: Processing-in-memory Acceleration for Efficient LPN-based Post-Quantum Cryptography
VisualNet: An End-to-End Human Visual System Inspired Framework to Reduce Inference Latency of Deep Neural Networks
Efficient Analysis for Mitigation of Workload-dependent Aging Degradation
AxRLWE: A Multi-level Approximate Ring-LWE Co-processor for Lightweight IoT Applications
Privacy-Preserving Medical Image Segmentation via Hybrid Trusted Execution Environment
Clonable PUF: on the Design of PUFs That Share Equivalent Responses
AutoPrivacy: Automated Layer-wise Parameter Selection for Secure Neural Network Inference
BUNET: Blind Medical Image Segmentation Based on Secure UNET
NASS: Optimizing Secure Inference via Neural Architecture Search
ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition
AxMM: Area and Power Efficient Approximate Modulo Multiplier for R-LWE Cryptosystem
Filianore: Better Multiplier Architectures for LWE-based Post-Quantum Key Exchange
DArL: Dynamic Parameter Adjustment for LWE-based Secure Inference
Towards Practical Homomorphic Email Filtering: A Hardware-Accelerated Secure Naive Bayesian Filter
DWE: Decrypting Learning with Errors with Errors
Coin Flipping PUF: A Novel PUF with Improved Resistance against Machine Learning Attacks
LSTA: Learning-Based Static Timing Analysis for High-Dimensional Correlated On-Chip Variations
SCAM: Secured Content Addressable Memory Based on Homomorphic Encryption
Runtime NBTI Mitigation for Processor Lifespan Extension via Selective Node Control
Workload-Aware Worst Path Analysis of Processor-Scale NBTI Degradation
Nonlinear Delay-Table Approach for Full-Chip NBTI Degradation Prediction
Mitigation of NBTI-induced Timing Degradation in Processor
