AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

AdvBDGen: A Robust Framework for Generating Adaptive and Stealthy Backdoors in LLM Alignment

Pankayaraj Pathmanathan, Udari Madhushani Sehwag, Michael-Andrei Panaitescu-Liess, Cho-Yu Jason Chiang, Furong Huang

AAAI 2026 · Large language models › Instruction and fine-tuning · approach: Safety alignment after fine-tuning · 2 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
train stylistic paraphraser with the AdvBDGen paradigm (ablation)stylistic backdoor · PKU Beavertailsattack success rateImproved
use AdvBDGen triggers instead of constant triggersbackdoor attack · PKU Beavertails (no defense)attack success rateNo clear effect
use AdvBDGen generator instead of stylistic paraphraser for trigger variantsbackdoor trigger generation · PKU Beavertails, 100 samples per test promptfraction of successful triggers: 10-20% -> 40-60%Improved
add strong and weak discriminator pair to the generator objective (ablation)AdvBDGen · PKU Beavertailstrigger complexityImproved
transfer triggers from Mistral 7B generator to Gemma 7B and Llama 3 8B victims (ablation)AdvBDGen · PKU Beavertailsattack success rateImproved
use AdvBDGen triggers instead of constant triggers under trigger-removal unlearning (ablation)backdoor attack · PKU Beavertails, trigger removalattack success rate after defenseImproved
apply latent adversarial training defenseAdvBDGen and constant triggers · PKU Beavertailsattack success rateHurt

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