AI Paper Atlas

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

SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning

Tairan Huang, Yulin Jin, Junxu Liu, Qingqing Ye, Haibo Hu

CVPR 2026 · Adversarial robustness and security › Black-box and transfer attacks · approach: Query-efficient black-box attacks · 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 discriminator in Stage 2 (ablation)SEBA · Pixel MuJoCoFID: 97.18 -> 62.43Improved
add discriminator in Stage 2 (ablation)SEBA · Cheetah Run (pixel MuJoCo)reward under attack: 3.15 -> 1.61No clear effect
train shadow critic on perturbed states in Stage 1 (ablation)SEBA · Walker Walk (pixel MuJoCo)reward under attack: 118.11 -> 35.74Improved
train shadow critic on perturbed states in Stage 1 (ablation)SEBA · Cheetah Run (pixel MuJoCo)reward under attack: 19.25 -> 1.61Improved
add world model for synthetic rollouts (ablation)SEBA · Pixel MuJoCotraining env queries: 800K -> 160KImproved
add world model for synthetic rollouts (ablation)SEBA · Cheetah Run (pixel MuJoCo)reward under attack: 0.18 vs 1.61No clear effect
use SEBA instead of PGD (white-box)Pixel MuJoCo attack · Cheetah Run (pixel MuJoCo)reward under attack: 150.72 -> 1.61Improved
use SEBA instead of OPTIMAL (black-box RL attacker)Pixel MuJoCo attack · Reacher Hard (pixel MuJoCo)reward under attack: 592.64 -> 0.3Improved
use SEBA instead of PA-AD in training costPixel MuJoCo attack · Pixel MuJoCotraining env queries: 4M -> 160KImproved
use SEBA instead of PA-ADAtari attack · Pong (Atari)reward under attack: -21 -> 3Hurt
use SEBA instead of OPTIMALAtari attack · Alien (Atari)reward under attack: 7400 -> 982Improved
use SEBA objective instead of PGD for targeted attackTargeted attack · Walker Walk targetedsuccess rate (%): 37.2 -> 91.3Improved

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