ai_security▌

← feed

NOTABLE

arXiv 2611.14978: a disguised LoRA adapter passes audit on the unquantized base, then hijacks decisions once deployment quantizes it

An attacker-supplied adapter for decision tasks such as content moderation and function selection preserves benign behavior on the unquantized trusted base — passing audit — but hijacks decisions after the same base is quantized for deployment, without modifying the base and without knowing the victim's quantization configuration.

released 2026-11-13 · ingested 2026-11-14T00:45Z

Sources

Primary source

Cite this

APA

AI Security Index. (2026). arXiv 2611.14978: a disguised LoRA adapter passes audit on the unquantized base, then hijacks decisions once deployment quantizes it. AI Security Index. https://arxiv.org/abs/2611.14978

BibTeX

@misc{aisec-2026-11-13-arxiv-2611-14978-disguised-adapter-quantized-deployment-hijack,
  title  = {arXiv 2611.14978: a disguised LoRA adapter passes audit on the unquantized base, then hijacks decisions once deployment quantizes it},
  year   = {2026},
  howpublished = {\url{https://arxiv.org/abs/2611.14978}},
  note   = {AI Security Index item 2026-11-13-arxiv-2611-14978-disguised-adapter-quantized-deployment-hijack}
}