ConLA: Contrastive Latent Action Learning from Human Videos for Robotic Manipulation

Understand

Vision-Language-Action (VLA) models achieve preliminary generalization through pretraining on large scale robot teleoperation datasets.

• However, acquiring datasets that comprehensively cover diverse tasks and environments is extremely costly and difficult to scale.
• In contrast, human demonstration videos offer a rich and scalable source of diverse scenes and manipulation behaviors, yet their lack of explicit action supervision hinders direct utilization.
• Prior work leverages VQ-VAE based frameworks to learn latent actions from human videos in an unsupervised manner.