AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

DIFT: Protecting Contrastive Learning Against Data Poisoning Backdoor Attacks

Jiang Zhu, Yulin Jin, Qingqing Ye, Zhibiao Guo, Kun Fang, Ruochen Du, Yingnan Zhao, Haibo Hu

AAAI 2026 · Adversarial robustness and security › Backdoor and poisoning · approach: Detecting poisoned samples · 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 DIFT in-training defenseSimCLR · CIFAR-10 (CTRL)ASR (%): 99.0 -> 2.0Improved
apply DIFT instead of post-training ensemble defense (B2)SimCLR · CIFAR-10 (CTRL)ASR w/ defense (%): 13.1 -> 2.0Improved
apply DIFT on ImageNet-100SimCLR · ImageNet-100 (CTRL)ASR w/ defense (%): 49.1 -> 5.3Improved
apply DIFT instead of retraining baseline on accuracySimCLR · CIFAR-10 (CTRL)accuracy drop (%): B2 1.4 -> DIFT 0.4Improved
add fine-tuning on detected poisoned samples with density-reduction loss (ablation)DIFT · CIFAR-10 (SimCLR)ASR w/ defense (%): 86.4% -> 2.0Improved
add label spreading after density-based detection (ablation)DIFT · CIFAR-10 and CIFAR-100recall of poisoned samplesImproved
add label spreading after density-based detection (ablation)DIFT · CIFAR-10 and CIFAR-100precision of poisoned sample detectionHurt
omit label spreading from DIFT (ablation)DIFT · final modelresidual ASRHurt

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