Metraipolis › archive › dc_732aefb32c5f

Deep Learning Latency Attacks and Defenses: A Cross-Domain Survey of Availability Threats — arXiv:2609.36732

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.36732
License (as archived)
CC0-1.0
Archived
2026-10-07T17:26:52.957Z
SHA-256 of the archived text
e297baf30631335d20668d9afa20a272bcac91169e01223bb5dcd79f91ee32a7
Stable URL of this page
https://metraipolis.com/documento/dc_732aefb32c5f

How to cite

Deep Learning Latency Attacks and Defenses: A Cross-Domain Survey of Availability Threats — arXiv:2609.36732 — arXiv (en), https://arxiv.org/abs/2609.36732 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_732aefb32c5f (Metraipolis archive dc_732aefb32c5f, sha256 e297baf30631335d20668d9afa20a272bcac91169e01223bb5dcd79f91ee32a7)
BibTeX
@misc{dc732aefb32c5f,
  title = {Deep Learning Latency Attacks and Defenses: A Cross-Domain Survey of Availability Threats — arXiv:2609.36732},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.36732},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_732aefb32c5f; sha256 e297baf30631335d20668d9afa20a272bcac91169e01223bb5dcd79f91ee32a7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36732. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Adversarial machine learning has focused mainly on integrity, but availability is an increasingly consequential complement. Latency attacks (also energy-latency attacks) increase inference-time work, energy, or response time, causing deadline misses, throughput collapse, or resource exhaustion in vehicle controllers, interactive services, or battery-powered sensors, sometimes while preserving the nominal prediction. This survey unifies a fragmented literature spanning perception pipelines (including physical attacks on autonomous-driving detection and tracking), input-adaptive neural inference (sponge examples, dynamic networks), and autoregressive and agentic systems (output-length, verbose-image, and reasoning denial-of-service attacks on LLMs, VLMs, mixture-of-experts models, and tool-using agents). We organize attacks by exploited computational bottleneck rather than formulation, separating what makes a computation expensive from how the attacker triggers it; the delivery channel (input, prompt or retrieved content, message, poisoning, or weight tampering) is an orthogonal attribute. Many attacks share one mechanism, intermediate-work amplification, motivating a work-budget defense abstraction; we distinguish caps on the work entering an expensive stage from caps on the results leaving it. We further analyze when a model-level cost increase becomes a system-level availability failure, which depends on critical-path share, slack, existing ceilings, accumulation, resource sharing, and fallback policy, not on the amplification factor alone. We also provide a threat-model taxonomy, consolidated quantitative comparisons, a defense review by control mechanism, and open challenges such as standardized evaluation, physical realizability, and whole-system availability. Companion website: this https URL .

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-30 in cs.CR).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2609.36732 — 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 2026-10-07T16:58:17.339Z 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).