AI Paper Atlas

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

Attack for Defense: Adversarial Agents for Point Prompt Optimization Empowering Segment Anything Model

Xueyu Liu, Xiaoyi Zhang, Meilin Liu, Guangze Shi, Jia Shen, Yujie Wang, Cai Zhao, Ziyuan He, Yongfei Wu, Mingqiang Wei, Yongle Chen

CVPR 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Attacks and defenses in sequential settings · 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
Apply trained attack agent to ideal prompts (ablation)SAM with ideal prompts · PASCAL VOCmDSC: 78.5 -> 34.2Hurt
Apply defense agent after the attack (ablation)SAM with attacked prompts · PASCAL VOCmDSC: 34.2 -> 83.5Improved
Apply defense agent after the attack (ablation)SAM with attacked prompts · KvasirmDSC: 59.7 -> 76.4Improved
Add PPD defense agent to feature-matching promptsSAM with feature matching · PASCAL VOCmDSC: 41.3 -> 73.0Improved
Add PPD defense agent to feature-matching promptsSAM with feature matching · ISICmDSC: 66.4 -> 76.3Improved
Add PPD defense agent to feature-matching promptsSAM with feature matching · KvasirmDSC: 31.4 -> 64.1Improved
Use adversarial PPD instead of PPO prompt optimizationfeature-matching prompts on SAM · KvasirmDSC: 42.9 -> 64.1Improved
Use adversarial PPD instead of PPO prompt optimizationfeature-matching prompts on SAM · PASCAL VOCmDSC: 61.7 -> 73.0Improved
Use learned prompt encoder trained on natural images (VRP-SAM) on medical dataVRP-SAM · ISIC / KvasirmDSC: 58.2 (VOC) -> 5.9 (ISIC), 0.0 (Kvasir)Hurt
Add PPD refinement on top of Matcher-style prompts vs running PPDruntime · One-shot workflow timeprompt engineering time (s): 2.768 (FM-PPO) -> 1.249 (FM-PPD)Improved

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