Add WAM paper: Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on Heterogeneous Robots

[279object]

Paper

• Title: Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on Heterogeneous Robots
• Short name: Embodied.cpp
• arXiv ID: 2607.02501v1
• Paper: https://arxiv.org/pdf/2607.02501
• Authors: Ling Xu, Chuyu Han, Borui Li, Hao Wu, Shiqi Jiang, Ting Cao, Chuanyou Li, Sheng Zhong, Shuai Wang
• Published: 2026-07-02
• arXiv categories: cs.RO, cs.CV, cs.OS
• Matched keywords: WAM, world-action model, world-action models

README Entry

- **Embodied.cpp**: "Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on Heterogeneous Robots", arXiv 2026. ![](https://img.shields.io/badge/Needs--Taxonomy--Review-64748b)
  [[📄 Paper](https://arxiv.org/pdf/2607.02501)]

Robot Decision

• Decision: accept
• Confidence: medium
• Taxonomy: uncertain
• Taxonomy confidence: high
• README heading: ## World Action Model
• Badges: https://img.shields.io/badge/Needs--Taxonomy--Review-64748b

Reason

Paper presents an embodied inference runtime with explicit first-class support and benchmarking for world-action models, though it is infrastructure rather than a new WAM method.

Evidence

• abstract names world-action models as supported embodied model family
• evaluates a preliminary WAM benchmark using a LingBot-VA Transformer block
• related work analyzes WAM subtypes such as predict-then-act and unified AR-modeling WAMs

Taxonomy Evidence

• portable C++ inference runtime for embodied models
• evaluated on preliminary WAM benchmark using a LingBot-VA Transformer block

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