AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protection

Yiming Li, Yang Bai, Yong Jiang, Yong Yang, Shu-Tao Xia, Bo Li

NeurIPS 2022 · Adversarial robustness and security › Backdoor and poisoning · approach: Trigger-based backdoor attacks · 25 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
shuffle poisoned labels rather than assign one fixed target labelUBW-P · CIFAR-10D_p: 0.0000 -> 2.2548Improved
shuffle poisoned labels rather than assign one fixed target labelUBW-P · ImageNet (50-class subset)D_p: 0.0000 -> 2.7156Improved
use clean labels and optimize sample-specific triggers for dispersible predictionsUBW-C · CIFAR-10ASR-C (%): 54.46 -> 87.56Improved
use clean labels and optimize sample-specific triggers for dispersible predictionsUBW-C · ImageNet (50-class subset)ASR-C (%): 2.78 -> 60.00Improved
use source-class samples rather than all classes for gradient matching (ablation)UBW-C · CIFAR-10ASR-A (%): 58.83 -> 89.80Improved
use source-class samples rather than all classes for gradient matching (ablation)UBW-C · ImageNet (50-class subset)ASR-A (%): 42.03 -> 74.00Improved
increase trigger size from 2 to 12 (ablation)UBW-P · CIFAR-10ASR-A (%): 82.60 -> 88.30Improved
increase trigger size from 2 to 12 (ablation)UBW-C · CIFAR-10ASR-A (%): 4.38 -> 98.00Improved
increase poisoning rate (ablation)UBW-P and UBW-C · CIFAR-10 and ImageNetASRImproved
increase poisoning rate (ablation)UBW-P and UBW-C · CIFAR-10 and ImageNetBAHurt
increase UBW-C trade-off λ (ablation)UBW-C · CIFAR-10 and ImageNetD_pImproved
increase UBW-C trade-off λ (ablation)UBW-C · CIFAR-10 and ImageNetASRNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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