Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience

arXiv 2026 43.6 method

TLDR

Ropedia W0 is a cross-embodiment loco-manipulation world foundation model trained on human and robot data, enabling zero-shot to post-trained prediction, planning, and real-time VR rollouts.

Reasoning

The paper introduces a large-scale unified dataset and benchmark, with strong empirical evaluation across multiple embodiments. However, reliance on manual keypoint annotation and limited detail on zero-shot generalization constraints visible in the abstract temper the claims.

Read-first score

Read-first score 43.6, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 57.

Recency 5%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Topical relevance 25%

81.4

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Methodology quality 15%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=benchmark,dataset,evaluation

Reproducibility 15%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=code,dataset

Graph prestige 15%

5.7

PageRank-style prestige over the local citation graph built from available OpenAlex references. local_pagerank=0.0016

Citation impact 15%

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. cited_by_count=0

Citation velocity 10%

Citation velocity estimates citations per publication-year to reduce old-paper bias. velocity=0.00

Field roles

FrontierMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 635.

Keyword Scores

world model

interactive world model

world dynamics prediction

world simulator

generative world model

model-based reinforcement learning world model

video world model

Deep Analysis

Innovations

First egocentric loco-manipulation world foundation model spanning humans and diverse robot embodiments
Xp-Data: a curated dataset with 10k hours of egocentric human data and 3k hours of robot data, unified multimodal annotations, and systematic quality control
Unified kinematic representation encoding locomotion through camera poses and manipulation through projected skeletons, with robot skeletons combining embodiment-specific forward kinematics and extensive manual keypoint annotation and correction
Xp-Teleop: a long-horizon loco-manipulation benchmark based on high-quality dexterous humanoid teleoperation data
Distillation enabling real-time, minute-long rollouts for live VR interaction

Methodology

Ropedia W0 is a world foundation model trained on Xp-Data, which includes 10k hours of egocentric human data and 3k hours of robot data with unified multimodal annotations. It uses a unified kinematic representation where locomotion is encoded via camera poses and manipulation via projected skeletons, with robot skeletons combining embodiment-specific forward kinematics and manual keypoint annotation/correction. Evaluation is performed across five cross-embodiment benchmarks and held-out human egocentric video under zero-shot, few-shot, and full-data post-training settings, with distillation used for real-time deployment.

Key Results

Ropedia W0 achieves state-of-the-art performance across five cross-embodiment benchmarks (DreamDojo, Ψ_0, ω-HOME, Xp-Teleop, and AgiBot) and on held-out human egocentric video, from zero-shot inference to few-shot and full-data post-training. Scaling human pre-training data improves performance on both human and robot evaluations, and the distilled model supports real-time, minute-long rollouts enabling live VR interaction.

Resources

Paper