Entity dossier

Bot Landscape / Ecosystem / Entity dossiers / LATENT

Model model--latent

LATENT

Research whole-body tennis controller evaluated on Unitree G1.

Reviewed relationships 3

  1. Verified

    LATENT evaluated on Unitree G1 Robot

    Scope humanoid tennis paper

    Boundary The paper and code target Unitree G1; ET1 is not retroactively inserted.

    First observed · reviewed · [M10] · Show in graph

  2. Boundary

    LATENT not proven on Galbot ET1 Robot

    Scope public evidence boundary

    Boundary The tennis paper evaluates Unitree G1; an ET1 tennis-training announcement is not that evaluation.

    First observed · reviewed · [R1] · Show in graph

  3. Verified

    LATENT uses simulator MuJoCo Simulator

    Scope repository simulator path

    Boundary The README names MuJoCo.

    First observed · reviewed · [M10] · Show in graph

Canonical evidence claims 10

Claims are projected from authored tables. A dash remains an explicit missing value; it is never filled from a neighboring product or name match.

Robot developer ecosystems, SDKs, and simulation Unitree G1 Research Code 5 rows · 1 declared gap

MetricValue and evidence
Control / tracking notedefault preprocess frequency 50 Hz [M10]
License Declared gap
PaperarXiv:2603.12686 [R2]
Simulator in READMEMuJoCo [M10]
Stated targetUnitree G1 (G1TrackingGeneral; motion path UnitreeG1) [M10]

Research — recent papers Paper Body Unitree G1 5 rows · 1 declared gap

MetricValue and evidence
LATENTarXiv:2603.12686 [R1]
LATENTpaper eval — tennis whole-body control; the announced Galbot biped’s tennis mention is not this paper’s eval [R1]
LATENT Declared gap
LATENT29-DoF Unitree G1 [R1]
LATENTofficial code on Galbot GitHub [R1]

Source ledger 3

  1. [R1]ResearchLATENT — “Learning Athletic Humanoid Tennis Skills…”. Type: research paper. ID: arXiv:2603.12686. Retrieved: 2026-08-25.Retrieved
  2. [M10]ManualGalaxyGeneralRobotics — LATENT README (default branch: master). Type: official research-code repository. Retrieved: 2026-08-27.Retrieved
  3. [R2]ResearchZhang et al. — LATENT: Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data. Type: research. Venue: IROS 2026. Retrieved: 2026-08-27.Retrieved

What is not inferred

No fuzzy entity matching, adjacent-column borrowing, universal compatibility, commercial maturity, or checkpoint-to-product identity. Absence of a relation means unreviewed or unsupported here—not proven incompatibility.