Metraipolis › archive › dc_43530f492da8

Breaking $\mathsf{TraintoDefend}$ via Adaptive Neuron Clustering in Cryptanalytic Model Extraction — arXiv:2611.07370

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.07370
License (as archived)
CC0-1.0
Archived
2026-11-06T14:43:47.794Z
SHA-256 of the archived text
f609621057186070b66d8aab644d99d358fae24f375c5b82fccb1c82409596d4
Stable URL of this page
https://metraipolis.com/documento/dc_43530f492da8

How to cite

Breaking $\mathsf{TraintoDefend}$ via Adaptive Neuron Clustering in Cryptanalytic Model Extraction — arXiv:2611.07370 — arXiv (en), https://arxiv.org/abs/2611.07370 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_43530f492da8 (Metraipolis archive dc_43530f492da8, sha256 f609621057186070b66d8aab644d99d358fae24f375c5b82fccb1c82409596d4)
BibTeX
@misc{dc43530f492da8,
  title = {Breaking $mathsfTraintoDefend$ via Adaptive Neuron Clustering in Cryptanalytic Model Extraction — arXiv:2611.07370},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.07370},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_43530f492da8; sha256 f609621057186070b66d8aab644d99d358fae24f375c5b82fccb1c82409596d4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.07370. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Cryptanalytic model extraction poses a fundamental threat to deep learning model confidentiality by enabling precise parameter recovery with theoretical guarantees of functional equivalence. To counter this growing threat, $\mathsf{TraintoDefend}$ (NeurIPS 2025) was proposed as the first dedicated defense against cryptanalytic extraction and remains the only defense believed to be effective to date. It encourages neurons to learn similar weights so that they switch activation states together, making their individual contributions harder to distinguish. However, we notice that neurons with similar weights need not switch activation states together under small perturbations. Instead, weight similarity can cause the original clustering rule to merge signatures from distinct neurons when only a subset of their normalized weight coordinates agree. Motivated by this observation, we introduce $\mathsf{SimBreak}$, an adaptive clustering rule that breaks the weight-similarity defense of $\mathsf{TraintoDefend}$. $\mathsf{SimBreak}$ estimates a relative scale from all jointly observed weight coordinates and uses the same scale to test both weight direction and bias consistency. Across six protected fully connected networks trained on MNIST, $\mathsf{SimBreak}$ raises target layer signature recovery from $0\%$ to $100\%$, with a maximum relative recovery error of $2^{-22.27}$. Our work challenges the use of neuron weight similarity as a sufficient basis for protection and establishes adaptive evaluation as essential to assessing cryptanalytic extraction defenses.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2611.07370 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-06T14:21:12.528Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).