AI Paper Atlas

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

Exploiting Supervised Poison Vulnerability to Strengthen Self-Supervised Defense

Jeremy Styborski, Mingzhi Lyu, Yi Huang, Adams Kong

ECCV 2024 · Adversarial robustness and security › Attacks on text and graphs · approach: Attacks and defenses in sequential settings · 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
add SL-guided adversarial training to SSL+SL (VESPR) (ablation)SSL+SL · ImageNet-100 poisoned trainingPsn Min accuracy (%): 21.82 -> 45.20Improved
add SL-guided adversarial training to SSL+SL (VESPR) (ablation)SSL+SL · ImageNet-100 poisoned trainingPsn Avg accuracy (%): 51.17 -> 55.37Improved
use gradient-based adversarial perturbation instead of Gaussian noise augmentation (ablation)SSL+SL · ImageNet-100 poisoned trainingPsn Avg accuracy (%): 52.27 -> 55.37Improved
generate adversarial examples with SL cross-entropy loss instead of SSL loss (ablation)VESPR · ImageNet-100 poisoned trainingPsn Min accuracy (%): 36.64 -> 45.20Improved
generate adversarial examples with CE loss instead of combined loss (ablation)VESPR · ImageNet-100 poisoned trainingPsn Avg accuracy (%): 54.10 -> 55.37Improved
apply adversarial training to SSL (SSL+AT) instead of plain SSL (ablation)SSL · ImageNet-100 poisoned trainingPsn Avg accuracy (%): 46.05 -> 44.81Hurt
add adversarial training to supervised learning (SL+AT) (ablation)SL · ImageNet-100 poisoned trainingPsn Avg accuracy (%): 21.41 -> 44.29Improved
use SSL-style augmentations with supervised training (ablation)SL · ImageNet-100 poisoned trainingPsn Avg accuracy (%): 21.41 -> 41.54Improved
add supervised CE loss to SSL without adversarial training (SSL+SL) (ablation)SSL · ImageNet-100 poisoned trainingPsn Min accuracy (%): Psn Avg 46.05 -> 51.17, Psn Min 14.80 -> 21.82Mixed
use cutout, mixup or CutMix augmentation defensesSL · ImageNet-100 poisoned trainingPsn Min accuracy (%): Psn Min 6.10 -> 5.06 / 5.16 / 5.70No clear effect
use SSL (SimCLR) against the SSL-targeted CP8 poisonSSL · ImageNet-100 CP8clean test accuracy (%): 14.80 vs SL 57.46Hurt
add adversarial training in VESPR instead of SSL+SL, trained on clean data (ablation)SSL+SL · ImageNet-100 clean dataclean test accuracy (%): 79.66 -> 74.84Hurt

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