Discrete Token-Based Latent Actions Enhance Sample Efficiency in Visual Imitation Learning

This report synthesises findings from 16 peer-reviewed papers addressing the following research question: To what extent do discrete token-based latent action models improve sample efficiency and task success rates over continuous latent action models when training on unlabeled video demonstrations. 7 claims were extracted from source literature; 7 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 7.5/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: To what extent do discrete token-based latent action models improve sample efficiency and task success rates over continuous latent action models when training on unlabeled video demonstrations?

Autonomous literature synthesis. Automated review score: 7.5/10. Full text and citation available at Assignee Research.