AI Paper Atlas

Research map / Federated learning and privacy / Privacy attacks and secure learning

INDEXGUARD: Index-only Backdoor Vetting for Secure Federated PEFT of Large Language Models

Javad Dogani, Devriş İşler, Nikolaos Laoutaris

ICML 2026 · Federated learning and privacy › Privacy attacks and secure learning · approach: Gradient inversion attacks

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
make label skew stronger (Dirichlet alpha 0.5 to 0.1) (ablation)IndexGuard · AG News RIPPLESdetection accuracy: 99.1% -> 97.2%Hurt
make label skew stronger (Dirichlet alpha 0.5 to 0.1) (ablation)IndexGuard · AG News Syntacticdetection accuracy: 99.6% -> 98.7%Hurt
make label skew stronger (Dirichlet alpha 0.3 to 0.1) (ablation)IndexGuard · AG News InsertSentdetection accuracy: 100.00% -> 99.67%No clear effect
use encoder-only BERT-base instead of Llama-2-7B (ablation)IndexGuard · AG News InsertSentAUC: 1.00 -> 0.996Hurt
use RoBERTa-base instead of Llama-2-7B (ablation)IndexGuard · SST-2 InsertSentAUC: 0.996 -> 0.991Hurt
swap LoRA for other PEFT methods (AdaLoRA, QLoRA, LoRA+, DoRA) (ablation)IndexGuard · InsertSent AG News/SST-2AUC: all >= 0.998No clear effect
apply IndexGuard filtering of malicious clients (ablation)federated PEFT · BERT-base SST-2 / Llama-2-7B AG NewsASRImproved
apply IndexGuard filtering of malicious clients (ablation)federated PEFT · SST-2clean accuracy: 0.921 -> 0.895Hurt
apply IndexGuard filtering of malicious clients (ablation)federated PEFT · AG Newsclean accuracy: 0.690 -> 0.651Hurt
attacker mixes more poisoned indices into reported support (b 0.1 to 0.5) (ablation)IndexGuard · Adaptive reported-support attackdetection accuracy: 71.2 -> 100.0Improved
attacker mixes more poisoned indices (b 0.1 to 0.5) (ablation)unmitigated attack · Adaptive reported-support attackASR@10: 55.1% -> 82.4%Improved
IndexGuard (unsupervised, index-only) vs PEFTGuardIndexGuard · SC averageAUC: 1.000 -> 0.998No 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