Humanoid robots performing manipulation tasks in simulation
// PAPERS · READING LIST

Research behind the build

The papers that make this possible, from PPO to TWIST2.

THEME
STATUS
Showing 19 of 19 papers

TWIST2

2025 · Teleop · ARXIV:2511.02832

Honest P10 option: reproduce in sim, deploy the manipulation half on the real SO-ARM101.

Read on arXiv ↗

TWIST

2025 · Teleop · ARXIV:2505.02833

The predecessor; full whole-body control via mocap. The “why whole-body” argument.

Read on arXiv ↗

OmniH2O

2024 · Teleop · ARXIV:2406.08858

Universal human-to-humanoid whole-body teleop and learning.

Read on arXiv ↗

HumanPlus

2024 · Teleop · ARXIV:2406.10454

Humanoid shadowing & imitation from human video.

Read on arXiv ↗

HOMIE

2025 · Teleop · ARXIV:2502.13013

Loco-manipulation via an isomorphic exoskeleton cockpit (decoupled-control baseline).

Read on arXiv ↗

CLONE

2025 · Teleop · ARXIV:2506.08931

Closed-loop whole-body teleop for long-horizon tasks (partial-control baseline).

Read on arXiv ↗

Open-TeleVision

2024 · Teleop · ARXIV:2407.01512

Teleoperation with immersive active visual feedback.

Read on arXiv ↗

Mobile-TeleVision

2024 · Teleop · ARXIV:2412.07773

Predictive motion priors for humanoid whole-body control.

Read on arXiv ↗

GMR: General Motion Retargeting

2025 · Retargeting · ARXIV:2510.02252

Optional P10 retargeting study: human→robot on CPU.

Read on arXiv ↗

PPO

2017 · RL · ARXIV:1707.06347

The core policy-gradient algorithm behind Isaac Lab / RSL-RL.

Read on arXiv ↗

Learning to Walk in Minutes

2021 · RL · ARXIV:2109.11978

Massively-parallel RL for legged locomotion (RSL-RL foundations).

Read on arXiv ↗

ACT / ALOHA

2023 · Imitation · ARXIV:2304.13705

Action-chunking transformer; LeRobot's recommended first policy for P4.

Read on arXiv ↗

Diffusion Policy

2023 · Imitation · ARXIV:2303.04137

Visuomotor policy via action diffusion; the P4 comparison policy.

Read on arXiv ↗

3D Diffusion Policy

2024 · Imitation · ARXIV:2403.03954

Generalizable visuomotor policy from simple 3D representations.

Read on arXiv ↗

iDP3 (Improved 3D Diffusion Policy)

2024 · Imitation · ARXIV:2410.10803

The lineage behind TWIST2's high-level policy.

Read on arXiv ↗

R3M

2022 · Imitation · ARXIV:2203.12601

Pretrained visual representations for manipulation (backbone).

Read on arXiv ↗

π0

2024 · VLA · ARXIV:2410.24164

Vision-language-action flow model; manipulation-at-scale reference for P10.

Read on arXiv ↗

π0.5

2025 · VLA · ARXIV:2504.16054

VLA with open-world generalization.

Read on arXiv ↗

VideoMimic

2025 · VLA · ARXIV:2505.03729

Real2sim2real; contextual humanoid control from video.

Read on arXiv ↗