English

Attentive Feature Aggregation or: How Policies Learn to Stop Worrying about Robustness and Attend to Task-Relevant Visual Cues

Robotics 2026-03-02 v2 Computer Vision and Pattern Recognition

Abstract

The adoption of pre-trained visual representations (PVRs), leveraging features from large-scale vision models, has become a popular paradigm for training visuomotor policies. However, these powerful representations can encode a broad range of task-irrelevant scene information, making the resulting trained policies vulnerable to out-of-domain visual changes and distractors. In this work we address visuomotor policy feature pooling as a solution to the observed lack of robustness in perturbed scenes. We achieve this via Attentive Feature Aggregation (AFA), a lightweight, trainable pooling mechanism that learns to naturally attend to task-relevant visual cues, ignoring even semantically rich scene distractors. Through extensive experiments in both simulation and the real world, we demonstrate that policies trained with AFA significantly outperform standard pooling approaches in the presence of visual perturbations, without requiring expensive dataset augmentation or fine-tuning of the PVR. Our findings show that ignoring extraneous visual information is a crucial step towards deploying robust and generalisable visuomotor policies. Project Page: tsagkas.github.io/afa

Keywords

Cite

@article{arxiv.2511.10762,
  title  = {Attentive Feature Aggregation or: How Policies Learn to Stop Worrying about Robustness and Attend to Task-Relevant Visual Cues},
  author = {Nikolaos Tsagkas and Andreas Sochopoulos and Duolikun Danier and Sethu Vijayakumar and Alexandros Kouris and Oisin Mac Aodha and Chris Xiaoxuan Lu},
  journal= {arXiv preprint arXiv:2511.10762},
  year   = {2026}
}

Comments

This paper stems from a split of our earlier work "When Pre-trained Visual Representations Fall Short: Limitations in Visuo-Motor Robot Learning." While "The Temporal Trap" replaces the original and focuses on temporal entanglement, this companion study examines policy robustness and task-relevant visual cue selection. arXiv admin note: text overlap with arXiv:2502.03270