# PHUMA: Physically Reliable Humanoid Locomotion Dataset

arXiv: https://arxiv.org/abs/2510.26236 · HTML: https://mlsanity.com/paper/2510.26236/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Oct 2025.

## Built on by (most cited first, 5 of 5)

- [Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Tracking](https://mlsanity.com/paper/2606.03985/) (2026)
- [Towards Motion Turing Test: Evaluating Human-Likeness in Humanoid Robots](https://mlsanity.com/paper/2603.06181/) (2026)
- [PDF-HR: Pose Distance Fields for Humanoid Robots](https://mlsanity.com/paper/2602.04851/) (2026)
- [Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid Control](https://mlsanity.com/paper/2602.21599/) (2026)
- [Human-Centric Intelligence in the Era of Foundation Models: A Survey](https://mlsanity.com/paper/2608.18184/) (2026)

## Abstract (arXiv)

Motion imitation is a promising approach for humanoid locomotion, enabling agents to acquire humanlike behaviors. Existing methods typically rely on high-quality motion capture datasets such as AMASS, but these are scarce and expensive, limiting scalability and diversity. Recent studies attempt to scale data collection by converting large-scale internet videos, exemplified by Humanoid-X. However, they often suffer from physical artifacts such as floating, penetration, and foot skating, which hinder stable imitation. To address this, we introduce PHUMA, a Physically Reliable HUMAnoid locomotion dataset produced by a two-stage pipeline combining physics-aware curation and physics-constrained retargeting, aggregating both motion capture and internet video into a physically reliable, 73-hour corpus. On motion tracking benchmarks, PHUMA-trained policies achieve higher success rates than those trained on AMASS and Humanoid-X, and successfully transfer zero-shot to a real Unitree G1. The code is available at https://davian-robotics.github.io/PHUMA.