CLASS: Contrastive Learning via Action Sequence Supervision for Robot Manipulation

TL;DR

This paper introduces CLASS, a contrastive learning method that improves robot manipulation generalization by learning shared action sequence representations, especially under visual shifts, using weak supervision and a contrastive loss.

Contribution

The paper proposes a novel contrastive learning approach, CLASS, that leverages weak supervision from action sequences to enhance robotic manipulation generalization across visual variations.

Findings

CLASS achieves competitive results on simulation benchmarks.

Diffusion Policy with CLASS pre-training attains 75% success rate under visual shifts.

Baseline methods fail to perform well under significant visual variations.

Abstract

Recent advances in Behavior Cloning (BC) have led to strong performance in robotic manipulation, driven by expressive models, sequence modeling of actions, and large-scale demonstration data. However, BC faces significant challenges when applied to heterogeneous datasets, such as visual shift with different camera poses or object appearances, where performance degrades despite the benefits of learning at scale. This stems from BC's tendency to overfit individual demonstrations rather than capture shared structure, limiting generalization. To address this, we introduce Contrastive Learning via Action Sequence Supervision (CLASS), a method for learning behavioral representations from demonstrations using supervised contrastive learning. CLASS leverages weak supervision from similar action sequences identified via Dynamic Time Warping (DTW) and optimizes a soft InfoNCE loss with similarity-weighted positive pairs. We evaluate CLASS on 5 simulation benchmarks and 3 real-world tasks to achieve competitive results using retrieval-based control with representations only. Most notably, for downstream policy learning under significant visual shifts, Diffusion Policy with CLASS pre-training achieves an average success rate of 75%, while all other baseline methods fail to perform competitively. Project webpage: https://class-robot.github.io.