Robotics Research Hub
HoloBoxing: A Simple Recipe for Humanoid Boxing with 7 Minutes of Single-GPU Training

HoloBoxing: A Simple Recipe for Humanoid Boxing with 7 Minutes of Single-GPU Training

Xihan Ma, Bo Zhang, Boxiong Li, Yuhan Bao, Ziyue Liu, Yucheng Wang · N/A · 2026

reinforcement-learning control benchmark
PDF arXiv Code (GitHub)

Framework

N/A

License

N/A

Stars

N/A

Summary

HoloBoxing presents a simple recipe for learning humanoid boxing by building on a pretrained whole-body controller. The recipe separates reusable motor skills from task decisions: a compact planner requests future body-point targets, while the frozen controller handles joint-level execution.

Abstract Summary

HoloBoxing presents a simple recipe for learning humanoid boxing by building on a pretrained whole-body controller. The recipe separates reusable motor skills from task decisions: a compact planner requests future body-point targets, while the frozen controller handles joint-level execution. A learned decoder maps 32-dimensional actions to structured target windows, and measured-state imitation initializes the planner before reinforcement learning. Task learning then refines interaction behavior while reusing the decoder and controller, avoiding the need to relearn whole-body control. This combination keeps task optimization focused on a compact upper policy and supplies an initial behavior before reward-driven exploration. Starting from that available controller, the recipe accounts for about seven minutes of additional training, including reused representation fitting and a single-GPU task run. This excludes controller pretraining, data preparation, evaluation, and search. Simulated boxing experiments and component controls evaluate the resulting behavior and the roles of imitation initialization and the target representation.

Key Points

  • HoloBoxing presents a simple recipe for learning humanoid boxing by building on a pretrained whole-body controller.
  • The recipe separates reusable motor skills from task decisions: a compact planner requests future body-point targets, while the frozen controller handles joint-level execution.
  • A learned decoder maps 32-dimensional actions to structured target windows, and measured-state imitation initializes the planner before reinforcement learning.
  • Task learning then refines interaction behavior while reusing the decoder and controller, avoiding the need to relearn whole-body control.
  • This combination keeps task optimization focused on a compact upper policy and supplies an initial behavior before reward-driven exploration.

\calmregular 3 \calmregular HoloBoxing

The recipe learns task decisions above a fixed motor controller. It fits a target representation, initializes a feedback planner, and refines that planner through interaction while preserving whole-body execution. For masked observation $M(o_{t})$, the execution chain is Here, $\mathcal{N}(u)=u/\/u\/$ uses a fixed fallback at zero.

The planner $f_{\theta}$ requests targets $g_{t}$ through decoder $D_{\phi}$; the frozen controller $\pi_{\mathrm{HM}}$ combines them with robot feedback $c_{t}$.

\calmregular 3 \calmregular HoloBoxing — HoloBoxing: A Simple Recipe for Humanoid Boxing with 7 Minutes of Single-GPU Training
Figure 2: Elo trajectories and final ranking over 150 simulated bouts. Each pair plays ten bouts under the same robot parameters, arena, and referee.

\calmregular 4 \calmregular Evaluating short task adaptation

The evaluation asks whether the reused hierarchy produces interactive boxing after the recorded short training procedure. It measures training time, competitive behavior, and the effects of removing task-policy components. These measurements address different parts of the recipe: time quantifies adaptation cost, while behavior tests what that budget achieves.

All evaluated policies share the packaged robot, actuator settings, arena, simulator, and referee.

\calmregular 4 \calmregular Evaluating short task adaptation — HoloBoxing: A Simple Recipe for Humanoid Boxing with 7 Minutes of Single-GPU Training
Figure E.1: Recorded boxing by HoloBoxing (7-min version, red) against GGSD (top, match r02-m005) and RoboStriker (bottom, match r05-m015). The opponent is blue. Top frames occur at 3.02, 29.02, and 45.72 seconds; bottom frames at 3.02, 10.36, and 12.54 seconds. Hit labels follow recorded referee events; the visible poses are the corresponding control-frame states. Camera and lighting are adjusted; wall faces and corner posts are hidden for visibility.

\calmregular 5 \calmregular Experiments: training cost and resulting behavior

The selected HoloBoxing recipe uses the Unitree G1 and the frozen HoloMotion v2 controller ( Horizon Robotics, 2026 ) . The motion corpus combines paired InterHuman data ( Liang et al., 2024 ) and released RoboStriker trajectories ( RoboStriker Authors, 2026 ) . The 14 original training motions total 212.88 seconds after retargeting; both participants and mirrored versions remain grouped within each split.

Prior fitting uses 21148 training windows, 11074 validation windows, and a separate test population of 3246 windows.

Sources and Demonstrations

Figures are reproduced from Ma et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.

Share

Related Papers

A Sim-to-Real Integration Pipeline for Training and Deployment of Chunk-Based VLA Manipulation Policies
arXiv preprint

A Sim-to-Real Integration Pipeline for Training and Deployment of Chunk-Based VLA Manipulation Policies

Mathilde Kappel, Clémence Grislain, Mohamed Chetouani et al.

Vision-Language-Action (VLA) models have become a prominent paradigm for mapping multimodal inputs, including semantic instructions, visual observations of the scene, and proprioceptive observations, to robot actions.

manipulation vision vla reinforcement-learning sim-to-real control learning-from-demonstration benchmark
PDF Advanced