AI Paper Atlas

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

Keep On Going: Learning Robust Humanoid Motion Skills via Selective Adversarial Training

Yang Zhang, Zhanxiang Cao, Buqing Nie, Haoyang Li, Zhong Jiangwei, Qiao Sun, Xiaoyi Hu, Xiaokang Yang, Yue Gao

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Attacks and defenses in sequential settings · 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
replace domain randomization with selective adversarial trainingperceptive locomotion policy · Perceptive locomotion, clean env (sim)velocity tracking error: 0.25 -> 0.19Improved
replace domain randomization with selective adversarial trainingperceptive locomotion policy · Perceptive locomotion, DR env (sim)velocity tracking error: 0.28 -> 0.21Improved
replace domain randomization with selective adversarial trainingperceptive locomotion policy · Perceptive locomotion, DR env (sim)gravity projection error: 0.42 -> 0.36Improved
replace domain randomization with selective adversarial trainingperceptive locomotion policy (real G1) · Real-robot traversal: Stairstraversal success rate (%): 47 -> 84Improved
replace domain randomization with selective adversarial trainingperceptive locomotion policy (real G1) · Real-robot traversal: Slopestraversal success rate (%): 54 -> 90Improved
replace domain randomization with selective adversarial trainingwhole-body control policy (real G1) · Real-robot dance imitationtrajectory tracking error reduction (%): 32%Improved
replace domain randomization with selective adversarial trainingwhole-body control policy (real G1) · Real-robot dance imitation (120 s)duration before termination (s): 65 -> 120Improved
use selective attack policy instead of always-on (full-time) attack policy (ablation)adversarial training · Perceptive locomotion (sim), perturbation levels 1-4normalized rewardImproved
use selective attack policy instead of always-on attack policy (ablation)adversarial training · Perceptive locomotion, unperturbed envnormalized rewardImproved
use full-time attack policy (PAP) instead of domain randomization (ablation)adversarial training · Perceptive locomotion, low-perturbation envnormalized rewardHurt
increase attack-budget penalty lambda from 10^-3 to 10^-1 (ablation)SAP (WBC) · Whole-body control (sim)rewardImproved
increase attack-budget penalty lambda from 10^-1 to 10^1 (ablation)SAP (WBC) · Whole-body control (sim)rewardHurt

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