AI Paper Atlas

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

Greedy Coordinate Diffusion: Effective and Semantically Coherent Adversarial Attacks via Diffusion Guidance

Bohdan Turbal, Blossom Metevier, Max Springer, Aleksandra Korolova

ICML 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial prompts for language models · 0 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
replace gradient token proposals with discrete diffusion model proposalsGCG · JailbreakBench subset, Llama-3-8BHarmBench ASR: 0.00% -> 85.42%Improved
use diffusion-guided proposals under perplexity filteringGCD vs GCG/GBDA · JailbreakBench subset, perplexity defenseHarmBench ASR: Qwen GCG 68.75% -> 0.00% under filter; GCD stays 97.92%Improved
add guard-model loss to the attack objectiveGCD · Llama Guard 3-1B defense, Llama-3-8BHarmBench ASR: 87.50% vs best baseline 12.50%Improved
use GCD instead of AutoDAN on Qwen-2.5-7B response qualityGCD · Qwen-2.5-7BStrongReject score: 0.8054 -> 0.7789Hurt
use GCD instead of AutoDAN on Qwen-2.5-7B ASRGCD · Qwen-2.5-7BHarmBench ASR: 91.67% -> 97.92%Improved
remove one-shot diffusion lookahead (GCD-no-OSD) (ablation)GCD · JailbreakBench subsetattack performanceImproved

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