AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

LLA: Enhancing Security and Privacy for Generative Models with Logic-Locked Accelerators

You Li, Guannan Zhao, Yuhao Ju, Yunqi He, Jie Gu, Hai Zhou

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Membership inference and inversion · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use LLA (outlier-targeted permutation locking) instead of HPNN negation lockingMinitron/Llama/NeMo LLMs · pre-attack MMLU, 256 to 8192 protected neuronsMMLUImproved
select protected block by first emergent outliers (add step i) (ablation)LLA · Mistral NeMo Minitron 8B, pre-attackMMLU and perplexityImproved
select protected neurons by outlier-tracing score (add step ii) (ablation)LLA · Mistral NeMo Minitron 8B, pre-attackMMLU and perplexityImproved
add Hadamard rotation matrix to obfuscate keys (ablation)LLA · Mistral NeMo Minitron 8B, pre-attackMMLU and perplexityImproved
use permutation-based locking instead of negation-based locking (ablation)LLA · gradient-based OG attackkey recovery fidelity: adversary recovers 100% of key bits (negation) vs at most 86% (permutation, 256 neurons)Improved
increase permutation group size from 4 to 16 (ablation)LLA · OG gradient attack, 4 models averagedrobustness: key bits per neuron 1.5 -> 3.5Improved
increase number of protected neurons (key length) (ablation)LLA · gradient-based OG attackmax key fidelity: 86% -> 38%Improved
use oracle-less (OL) instead of oracle-guided (OG) attack (ablation)attack on LLA · group size 16, gradientattack effectivenessHurt
use genetic-based instead of gradient-based key attack (ablation)attack on LLA · OG, group size 16attack effectivenessHurt
add LLA locking module with 7,168 key bitsAI accelerator · computational overheadoverhead: less than 0.1% for 7,168 key bitsNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy