AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models

Zhongqi Wang, Jie Zhang, Shiguang Shan, Xilin Chen

TPAMI 2025 · Adversarial robustness and security › Backdoor and poisoning · approach: Trigger-based backdoor attacks · 8 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
add spatial correlation among attention maps (DAA-S over DAA-I) (ablation)DAA-I · average over 6 attacksF1: 73.43 -> 79.27Improved
add spatial correlation among attention maps (DAA-S over DAA-I) (ablation)DAA-I · average over 6 attacksAUC: 84.14 -> 86.27Improved
use EOS token instead of BOS token in the evolve-rate feature (ablation)DAA-S · avg AUC over 6 attacksAUC: 64.79 -> 86.27Improved
use EOS token instead of averaging all tokens (ablation)DAA-S · avg AUC over 6 attacksAUC: 70.15 -> 86.27Improved
use F-norm instead of SSIM as attention similarity (ablation)DAA-S · avg F1 over 6 attacksF1: 62.22 -> 79.27Improved
use F-norm instead of 1-norm as attention similarity (ablation)DAA-S · avg F1 over 6 attacksF1: 75.94 -> 79.27Improved
vary stability parameter gamma_L from -15 to -0.5 (ablation)DAA-S · avg AUC over 6 attacksAUC: 83.33 -> 87.09No clear effect
vary coupling coefficient c from 0.5 to 10 (ablation)DAA-S · avg AUCAUC: 85.40 -> 86.27No clear effect
start at later time step t=3 with span s=2 vs t=0 with s=1 (ablation)DAA-I · training F1F1: 69.72 -> 76.55Improved
start at t=3 with span s=2 instead of t=3, s=9 (ablation)DAA-I · training F1F1: 73.70 -> 76.55Improved
start at t=2 or 3 with long span for DAA-S (ablation)DAA-S · training F1F1: 79.72 -> 4.67Hurt
apply adaptive attack removing the EOS embedding to EvilEditDAA-S · EvilEditAUC: 85.95 -> 82.83Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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