Research map / Adversarial robustness and security / Black-box and transfer attacks
Consensus-Robust Transfer Attacks via Parameter and Representation Perturbations
Shixin Li, Zewei Li, Xiaojing Ma, Xiaofan Bai, Pingyi Hu, Dongmei Zhang, Bin Benjamin Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a transfer attack for untargeted adversarial examples that frames transferability as consensus robustness, combining parameter and representation perturbations to handle decision-boundary variation and representation drift across models.
Problem
Adversarial examples crafted on one model often transfer poorly to other models in black-box settings, due to decision-boundary variation across models and representation drift in feature space.
Approach
Frames transferability of untargeted attacks as a consensus-robustness problem, perturbing model parameters and intermediate representations so that the adversarial example remains effective across the decision boundaries and feature representations of different models.
Closest papers by meaning
- Towards Transferable Adversarial Attacks with Centralized Perturbation (AAAI 2024 · similarity 0.88)
- Enhancing Adversarial Transferability with Adversarial Weight Tuning (AAAI 2025 · similarity 0.88)
- Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment (NeurIPS 2025 · similarity 0.88)
- Rethinking Model Ensemble in Transfer-based Adversarial Attacks (ICLR 2024 · similarity 0.87)
- Enhancing Adversarial Transferability with Checkpoints of a Single Model's Training (CVPR 2025 · similarity 0.87)
- PubDef: Defending Against Transfer Attacks From Public Models (ICLR 2024 · similarity 0.87)
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.
