Review possible WAM paper: RynnWorld-4D: 4D Embodied World Models for Robotic Manipulation

[279object]

Paper

• Title: RynnWorld-4D: 4D Embodied World Models for Robotic Manipulation
• Short name: RynnWorld-4D
• arXiv ID: 2607.06559v1
• Paper: https://arxiv.org/pdf/2607.06559
• Authors: Haoyu Zhao, Xingyue Zhao, Siteng Huang, Xin Li, Deli Zhao, Zhongyu Li
• Published: 2026-07-07
• arXiv categories: cs.RO
• Matched keywords: embodied world model, embodied world models

README Entry

- **RynnWorld-4D**: "RynnWorld-4D: 4D Embodied World Models for Robotic Manipulation", arXiv 2026. ![](https://img.shields.io/badge/Needs--Taxonomy--Review-64748b)
  [[📄 Paper](https://arxiv.org/pdf/2607.06559)]

Robot Decision

• Decision: pending
• Confidence: medium
• Taxonomy: None
• Taxonomy confidence: None
• README heading: ## World Action Model
• Badges: https://img.shields.io/badge/Needs--Taxonomy--Review-64748b

Reason

Embodied robotic world model used for manipulation policy, but the described dynamics generation is conditioned on RGB-D and language instruction rather than explicit robot actions.

Evidence

• generates future RGB, depth, and optical flow from RGB-D image and language instruction
• inverse dynamics head outputs robot actions for closed-loop control

Needs taxonomy review: fallback README heading/badge was used because taxonomy.toml has no README mapping for this classification.

Human Review Checklist

• This paper belongs in the WAM survey
• README section is correct
• Badges are correct
• Short name is correct
• Paper link is correct
• Merge this PR if accepted