Newest papers

PICA-WAM: Visual Perturbation Invariance via Cross-Appearance Self-Alignment in World Action Models

Yude Zou, Yifei Yao, Guanbo Jiao, Jianming Ma, Guanjie Zheng, Xing Gao · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26813 · cited by 0 works (OpenAlex) · open-access PDF

French

Les modèles d'action mondiale (WAM) prédisent conjointement les vidéos futures et génèrent des actions, mais restent sensibles aux perturbations visuelles non pertinentes à la tâche, c'est-à-dire aux modifications de l'apparence visuelle qui ne devraient pas affecter le comportement des robots. Nous identifions une lacune dans la formation WAM conventionnelle: la prédiction vidéo encourage les représentations pour préserver l'apparence des caractéristiques prétrainées externes et ne garantit pas l'invariance souhaitée. Pour résoudre cette lacune, nous proposons PICA-WAM, un cadre de formation auto-surveillant qui expose les WAM à de telles perturbations mais ne nécessite pas d'observations perturb

Machine translation (Meta M2M100); the English original follows.

Abstract

World Action Models (WAMs) jointly predict future video and generate actions, yet remain sensitive to task-irrelevant visual perturbations, i.e., changes in visual appearance that should not affect robot behavior. We identify a gap in conventional WAM training: video prediction encourages representations to preserve appearance information, while action generation does not explicitly enforce invariance to task-irrelevant appearance. Data augmentation exposes WAMs to such perturbations but does not require perturbed and nominal observations to produce consistent action representations, and alignment with external pretrained features does not guarantee the desired invariance. To address this gap, we propose PICA-WAM, a self-supervised training framework that casts perturbation invariance as an auxiliary cross-appearance self-alignment task for WAMs. An exponential-moving-average (EMA) teacher processes the nominal observation to provide a deeper-layer target, while the online WAM processes a perturbed observation and aligns its shallower-layer action representation to this target through a lightweight projection head. Across four strong WAMs evaluated on RoboTwin 2.0, LIBERO-Plus, EBench, and a real robot, PICA-WAM consistently improves robustness under visual perturbations, preserves nominal performance in simulation, and raises the real robot success rate. PICA-WAM further reduces the sensitivity of video features to task-irrelevant appearance and its propagation into actions, as shown by paired layer-wise diagnostics. Code: https://github.com/YudeZou/PICA-WAM-release

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Robustness (evolution), Artificial intelligence, Perturbation (astronomy), Computer science, Computer vision, Robot

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange