Metraipolis › archive › dc_770f09583d49
Why Backdooring Neural Networks is so Easy? — arXiv:2609.36117
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.36117
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T17:27:05.232Z
- SHA-256 of the archived text
a8e4c13d8aadf035e30613ffe194f0408c57cc0b0b41976e5f4f5c0fba92fd6f- Stable URL of this page
https://metraipolis.com/documento/dc_770f09583d49
How to cite
Why Backdooring Neural Networks is so Easy? — arXiv:2609.36117 — arXiv (en), https://arxiv.org/abs/2609.36117 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_770f09583d49 (Metraipolis archive dc_770f09583d49, sha256 a8e4c13d8aadf035e30613ffe194f0408c57cc0b0b41976e5f4f5c0fba92fd6f)
BibTeX
@misc{dc770f09583d49,
title = {Why Backdooring Neural Networks is so Easy? — arXiv:2609.36117},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.36117},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_770f09583d49; sha256 a8e4c13d8aadf035e30613ffe194f0408c57cc0b0b41976e5f4f5c0fba92fd6f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36117. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Securing modern AI systems against backdoor attacks remains an open challenge and requires fundamentally principled estimates of the adversary's budget -- the poison fraction $\pi$ and trigger strength $\alpha$ needed to construct successful yet stealthy attacks. Motivated by recent empirical evidence that poisoning large language models can require a nearly constant number of malicious samples even as clean datasets grow, we derive an exact closed-form analysis of a quadratic neuron trained on a poisoned Gaussian mixture. We show, perhaps counterintuitively, that the same feature-learning dynamics that make neural networks powerful can also make them more vulnerable to backdoors. Specifically, with clean accuracy preserved to first order, $O(\pi)$, we demonstrate that lazy learning imposes the inverse-square-root scaling $\alpha \propto \pi^{-1/2}$ for a successful attack, while feature learning induces a quadratic detector whose loss margin scales as $O(\alpha^4)$, improving the attack budget to $\alpha \propto \pi^{-1/4}$. Consequently, nonlinear feature learning substantially reduces the trigger strength required at small poison fractions, thereby in a sense making feature learners more backdoor vulnerable. These results provide a theoretical mechanism consistent with large-scale empirical observations and demonstrate that security audits based on linear heuristics can systematically underestimate backdoor vulnerability in the widely adopted feature-learning regimes.
Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-30 in cs.CR).
Categorie arXiv: Machine Learning (cs.LG) ; Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2609.36117 — 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 2026-10-07T16:58:17.374Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).