Brandon Reagen

New York University — United States

Computer Science · Physical Sciences

21h-index3.7kcitations103works2.82yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Cryptography and Data Security(42), Privacy-Preserving Technologies in Data(26), Stochastic Gradient Optimization Techniques(17), Adversarial Robustness in Machine Learning(16), Advanced Neural Network Applications(12)

Publications103 total

Network and Compiler Optimizations for Efficient Linear Algebra Kernels in Private Transformer Inference (Invited Paper)
2025
MTU: The Multifunction Tree Unit for Accelerating Zero-Knowledge Proofs
2025Open Access
A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention
arXiv (Cornell University)·2025Open Access
EinHops: Einsum Notation for Expressive Homomorphic Operations on RNS-CKKS Tensors
arXiv (Cornell University)·2025Open Access
Need for zkSpeed: Accelerating HyperPlonk for Zero-Knowledge Proofs
2025· 4 citedOpen Access
Orion: A Fully Homomorphic Encryption Framework for Deep Learning
2025· 17 cited
Spectral Scaling Laws in Language Models: emphHow Effectively Do Feed-Forward Networks Use Their Latent Space?
2025Open Access
AERO: Softmax-Only LLMs for Efficient Private Inference
2024Open Access
TruncFormer: Private LLM Inference Using Only Truncations
arXiv (Cornell University)·2024Open Access
AERO: Softmax-Only LLMs for Efficient Private Inference
2024Open Access
AERO: Entropy-Guided Framework for Private LLM Inference
arXiv (Cornell University)·2024Open Access
ReLU's Revival: On the Entropic Overload in Normalization-Free Large Language Models
arXiv (Cornell University)·2024· 1 citedOpen Access
SZKP: A Scalable Accelerator Architecture for Zero-Knowledge Proofs
2024· 11 citedOpen Access
Osiris: A Systolic Approach to Accelerating Fully Homomorphic Encryption
arXiv (Cornell University)·2024Open Access
NTTSuite: Number Theoretic Transform Benchmarks for Accelerating Encrypted Computation
arXiv (Cornell University)·2024Open Access
CiFlow: Dataflow Analysis and Optimization of Key Switching for Homomorphic Encryption
2024· 11 cited
Orion: A Fully Homomorphic Encryption Framework for Deep Learning
arXiv (Cornell University)·2023· 3 citedOpen Access
CiFlow: Dataflow Analysis and Optimization of Key Switching for Homomorphic Encryption
arXiv (Cornell University)·2023Open Access
PriViT: Vision Transformers for Fast Private Inference
arXiv (Cornell University)·2023Open Access
Generating High-Performance Number Theoretic Transform Implementations for Vector Architectures
2023· 5 cited
page 1 of 6Next →

Frequent Co-authors

David Brooks(30), Gu-Yeon Wei(28), Udit Gupta(13), Robert Adolf(10), Hsien-Hsin S. Lee(8), Siddharth Garg(8), Carole-Jean Wu(7), Yakun Sophia Shao(7), Sae Kyu Lee(6), Xiaodong Wang(6), Xuan Zhang(6), Paul N. Whatmough(5), Kim Hazelwood(5), Bill Jia(5), Lillian Pentecost(5), Zahra Ghodsi(5), Saketh Rama(4), Maxim Naumov(4), Dheevatsa Mudigere(4), Mikhail Smelyanskiy(4)