arXiv AI August 10, 2026 By Chunhui Shi, Xinwen Fu

Cryptanalytic Extraction of Isolated Bias-Free GLU Feed-Forward Blocks by Antipodal Separation

arXiv:2608. 06631v1 Announce Type: cross Abstract: Cryptanalytic extraction has been demonstrated for ReLU networks, for networks using componentwise activations such as GELU or SiLU, and for a Transformer's final projection matrix.

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 AI.

arXiv AI

Sep 1

Curvature Cryptanalysis of Smooth Transformer Feed-Forward Networks

arXiv:2608.28843v1 Announce Type: cross Abstract: We show that smooth two-layer feed-forward networks (FFNs) expose an additional structural model extraction channel under a chosen-input raw-output o...

By Munawar Hasan, Apostol Vassilev

llms safety

arXiv Machine Learning

Aug 26

Is the Hard-Label Cryptanalytic Model Extraction Really Polynomial?

arXiv:2510.06692v3 Announce Type: replace Abstract: Deep Neural Networks (DNNs) have attracted significant attention, and their internal models are now considered valuable intellectual assets. Extrac...

By Akira Ito, Takayuki Miura, Yosuke Todo

arXiv Machine Learning

Sep 21

End-to-End Hard-Label Cryptanalytic Model Extraction Using Efficient Sign Recovery

arXiv:2609.21941v1 Announce Type: cross Abstract: The importance of deep neural networks (DNNs) is widely recognized, and the parameters obtained through training are regarded as valuable assets. Rec...

By Akira Ito, Takayuki Miura, Yosuke Todo

arXiv AI

Jun 9

Ghosted Layers: Unconstrained Activation Alignment for Recovering Layer-Pruned LLMs

arXiv:2605. 15491v2 Announce Type: replace-cross Abstract: Layer pruning removes entire Transformer decoder blocks from large language models, but introduces a mismatch between the hidden state received by the next surviving layer and the distribution it was trained to process, leading to significant performance degradation.

By Vincent-Daniel Yun, Junhyuk Jo, Sai Praneeth Karimireddy, Sunwoo Lee

llms efficiency safety

arXiv Machine Learning

Jul 16

Targeted Recovery of Weight-Space Mechanisms From Neural Networks

arXiv:2607. 13047v1 Announce Type: new Abstract: Parameter decomposition (PD) decomposes neural networks into interpretable computational components that faithfully reflect the original network's operations.

By Antoine Vigouroux, Lee Sharkey

llms

arXiv AI

Jun 11

ICA Lens: Interpreting Language Models Without Training Another Dictionary

arXiv:2606. 11722v1 Announce Type: cross Abstract: Finding interpretable directions in language-model representations is critical for understanding and controlling model behavior.

By Sida Liu, Feijiang Han

llms safety

One email a morning, machine-written

Related stories

Curvature Cryptanalysis of Smooth Transformer Feed-Forward Networks

Is the Hard-Label Cryptanalytic Model Extraction Really Polynomial?

End-to-End Hard-Label Cryptanalytic Model Extraction Using Efficient Sign Recovery

Ghosted Layers: Unconstrained Activation Alignment for Recovering Layer-Pruned LLMs

Targeted Recovery of Weight-Space Mechanisms From Neural Networks

ICA Lens: Interpreting Language Models Without Training Another Dictionary