arXiv Machine Learning July 7, 2026 By Ligong Han, Kai Xu, Hao Wang, Ruijiang Gao, Akash Srivastava

Layer-Parallel Inference Reduces Encrypted Nonlinear Depth in Transformers

Read the original on arXiv Machine Learning →

arXiv:2607. 04819v1 Announce Type: new Abstract: Fully homomorphic encryption (FHE) enables computation on encrypted data, but practical encrypted Transformer inference is bottlenecked by the sequential composition of many nonlinear blocks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

llms

arXiv AI

Jul 28

ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour

arXiv:2607. 23478v1 Announce Type: cross Abstract: Fully homomorphic encryption (FHE) provides strong cryptographic guarantees for private inference, but deploying transformer models under FHE remains prohibitively expensive.

By Jianhang Xie, Sicheng Tan, Vishnu Naresh Boddeti, Zhichao Lu

llms

arXiv AI

Sep 3

HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation

HEAT introduces a fine‑tuning method that treats the number of iterations used to approximate nonlinearities in fully homomorphic encryption (FHE) as learnable parameters, allowing them to co‑adapt with model weights. By optimizing iteration counts per nonlinearity, HEAT reduces the required iterations, bootstraps, and overall latency for encrypted GPT‑2 decoding while improving decode agreement. The approach achieves a 3.1× reduction in iterations, a 1.6× reduction in bootstraps, and a 1.4× speed‑up in end‑to‑end latency without changing the model architecture or requiring retraining from scratch.

By Alessandro Zirilli, Davide Marincione, Evgenios M. Kornaropoulos, Giuseppe Ateniese, Emanuele Rodol\`a

llms fine-tuning

arXiv Machine Learning

Jul 9

SNLP: Layer-Parallel Inference via Structured Newton Corrections

arXiv:2605. 17842v3 Announce Type: replace Abstract: Autoregressive language models execute Transformer layers sequentially, creating a latency bottleneck that is not removed by conventional tensor or pipeline parallelism.

By Ligong Han, Kai Xu, Hao Wang, Akash Srivastava

llms fine-tuning

arXiv Machine Learning

Jun 3

Why Are Linear RNNs More Parallelizable?

arXiv:2603. 03612v3 Announce Type: replace Abstract: The community is increasingly exploring linear RNNs (LRNNs) as language models, motivated by their expressive power and parallelizability.

By William Merrill, Hongjian Jiang, Yanhong Li, Anthony Lin, Ashish Sabharwal

llms

arXiv AI

Jul 22

PRISM: Sensitivity-Aware PolynoMial PRuning for EffIcient Neural Network Encryption

arXiv:2607. 18342v1 Announce Type: cross Abstract: Structured pruning is essential for making neural network inference feasible under homomorphic encryption (HE), yet its impact on model reliability has remained unexplored.

By Sahaj Majavdia, Mahdi Taheri

efficiency

arXiv Machine Learning

Sep 15

EI-DDLGN: Efficient Encrypted Inference with Deep Differentiable Logic Gate Networks under TFHE

arXiv:2609.13636v1 Announce Type: cross Abstract: Privacy-preserving inference via Torus Fully Homomorphic Encryption (TFHE) provides strong protection for sensitive data in outsourced deep learning...

By Mahmoud Y. M. Yassin, Mahmoud AbdelHafeez Sayed, Mostafa Taha

efficiency safety

One email a morning, machine-written

Related stories

ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour

HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation

SNLP: Layer-Parallel Inference via Structured Newton Corrections

Why Are Linear RNNs More Parallelizable?

PRISM: Sensitivity-Aware PolynoMial PRuning for EffIcient Neural Network Encryption

EI-DDLGN: Efficient Encrypted Inference with Deep Differentiable Logic Gate Networks under TFHE