arXiv:2611.03752v1 Announce Type: new Abstract: Third-party LoRA adapters introduce serious backdoor risks: malicious adapters can maintain normal task performance while embedding hidden, trigger-activated behaviors, making them difficult to detect through standard utility evaluations. Existing removal-based defenses typically sanitize backdoored models by updating, pruning, distilling, or mixing model parameters.
LoRA-Patch: Defending Against Backdoored LoRA Adapters via Directional Steering
About this summary. This is a short excerpt from the description that arXiv cs.CR published with its article. Cyber Security News did not report or verify the underlying story. Read the original: https://arxiv.org/abs/2611.03752
Source attribution: headline and facts are from arXiv cs.CR (arxiv.org). Summary method: excerpt of the source description. See our source attribution policy.
Related coverage
Agentic Backdoors via Pretraining Poisoning
arXiv:2610.37741v1 Announce Type: new Abstract: Large language models are pretrained on vast, uncurated corpora scraped from the public web. This creates a security risk…
Hugging Face malware campaign returns to hijack GPUs for crypto mining
A crypto-mining campaign abusing Hugging Face models returned days after takedown, hijacking GPU servers and planting backdoors, with one model downloaded over 28,000…
SECURITY AFFAIRS MALWARE NEWSLETTER ROUND 120
Security Affairs Malware newsletter includes a collection of the best articles and research on malware in the international landscape Malware Newsletter Auditing the…
Re: Godot 4.x: heap out-of-bounds read in OptimizedTranslation message lookup
Posted by Tomasz Kaye on Oct 23 Follow-up: the fix for this issue has been merged upstream. PR #124701, "Fix out-of-bounds reads with corrupt OptimizedTranslation data"…
LiteLLM Suspects GitHub Token Leak in AI Gateway Logs Behind Malicious Package Release
LiteLLM on Wednesday disclosed that threat actors compromised "multiple independent security controls" to push a malicious version of the Python package containing a…
LiteLLM Hijacked Again: Malicious 1.107.0 Wheels Steal Credentials and Load Remote Code
The JFrog Security Research team identified malicious code in `litellm` 1.107.0, a widely used Python interface and proxy for large language model providers. The payload…