AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

Vulnerabilities in Video Quality Assessment Models: The Challenge of Adversarial Attacks

Aoxiang Zhang, Yu Ran, Weixuan Tang, Yuan-Gen Wang

NeurIPS 2023 · Adversarial robustness and security › Black-box and transfer attacks · approach: Robustness benchmarks and defenses · 3 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
patch-based instead of pixel-based random search (SimBA) (ablation)black-box attack on VSFA · KoNViD-1kSRCC after attack: 0.7831 -> -0.0305Improved
patch-based instead of pixel-based random search (SimBA) (ablation)black-box attack on VSFA · LSVQ-TestSRCC after attack: 0.7747 -> 0.2386Improved
increase T from 1 to 8 (black-box) (ablation)black-box attack on VSFA · KoNViD-1kSRCC after attack: -0.0305 -> 0.2747Hurt
increase T from 1 to 8 (white-box) (ablation)white-box attack on VSFA · KoNViD-1kSRCC after attack: -0.7003 -> -0.6964No clear effect
add motion embeddings (ablation)BVQA-2022 · KoNViD-1kSRCC under black-box attack: -0.1779 -> 0.1978Improved
add motion embeddings (ablation)VSFA · KoNViD-1kSRCC under black-box attack: -0.0305 -> 0.1858Improved
add motion embeddings (ablation)VSFA · LSVQ-TestSRCC under black-box attack: 0.2386 -> 0.4358Improved
add motion embeddings (ablation)BVQA-2022 · LSVQ-TestSRCC under black-box attack: 0.1917 -> 0.3355Improved
perturb 1/10 of frames instead of all (white-box) (ablation)white-box attack on VSFA · KoNViD-1kSRCC after attack: -0.7198 -> -0.2692Mixed
perturb 1/10 of frames instead of all (black-box) (ablation)black-box attack on VSFA · KoNViD-1kSRCC after attack: -0.0305 -> 0.6836Hurt
Score-Reversed Boundary Loss instead of MSE loss (ablation)black-box and white-box attack on VSFA · KoNViD-1kquery efficiencyImproved
L_infinity instead of L2 constraint in white-box attack (ablation)white-box attack on NR-VQA models · five datasetsSRCC degradation: 194% vs 195%No clear effect

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