代码 · hold_missing_license
审核码：hold_missing_license; 官方关系：unverified_for_reproducible_open_use; 最近核验：2026-07-20代码：Public repository page and immutable head were inspected, but the promotion gate was not met.
权重：Not promoted or separately verified in this review.
数据：Not promoted or separately verified in this review.
许可证：not_disclosed；README identifies arXiv 2502.03206, but no root license file or GitHub-detected license was found.
依赖：未披露
硬件/传感器：未披露
实机证据：未披露
独立复现：No qualifying independent reproduction with a commit, disclosed hardware and quantitative metrics was established.
限制/反面证据：README identifies arXiv 2502.03206, but no root license file or GitHub-detected license was found.
当前建议：证据观察；README identifies arXiv 2502.03206, but no root license file or GitHub-detected license was found.
重点标签：promotion:withheld、license:missing-or-anomalous、runtime:not-reproduced
核验来源：来源 1
official_training_code_root_license_missing · hold_missing_root_license_and_deployment_release
审核码：hold_missing_root_license_and_deployment_release; 官方关系：verified_official_repository_but_root_license_missing; 最近核验：2026-07-20代码：官方训练代码存在。不可变提交 eb831af 包含 H1 资源、HugWBC 任务环境、RSL-RL、train.py 与 play.py；README 给出 Python 3.8、Isaac Gym Preview 4 和训练命令。
权重：未核验到官方策略检查点、模型发布页或 GitHub Release；play.py 需要用户自行训练产生策略。
数据：方法以仿真强化学习和命令空间训练为主，仓库未发布独立训练轨迹或数据集。
许可证：not_disclosed；提交根目录没有 LICENSE、LICENSE.md、COPYING 或 COPYING.md。legged_gym 与 rsl_rl 子目录各自保留上游 BSD-3-Clause 类许可，只覆盖相应上游组件，不能推定覆盖 HugWBC 新增代码、H1 资产或未来权重。
依赖：Python 3.8、PyTorch with CUDA 11.8 example、NVIDIA Isaac Gym Preview 4、bundled legged_gym and RSL-RL、external unitree_mujoco and unitree_sdk2_python for the described sim-to-sim / robot interface
硬件/传感器：CUDA-capable NVIDIA GPU; minimum VRAM, training steps and wall-clock cost are not disclosed、Unitree H1 for the author-side physical demonstrations、physical deployment requires an unreleased policy export and integration path
实机证据：The author project page shows outdoor and indoor Unitree H1 walking, jumping, standing, hopping, gait switching and upper-body intervention.；The public repository itself releases simulation training code and only points to external Unitree interfaces; it does not package the demonstrated physical deployment path.
独立复现：GitHub issue #9 contains multiple independent attempts to port the code to Unitree G1 and H1-2; participants repeatedly report unstable training or a disturbance curriculum that never advances. The reports identify concrete reward/curriculum symptoms but omit a fixed commit, full hardware/training budget and paper-comparable metrics, so they are counter-evidence rather than a qualifying reproduction.
限制/反面证据：缺少根许可证，不能进入已核验开放代码排序。；未发布检查点和完整 sim-to-sim / sim-to-real 代码，README 对外部 Unitree 工具的描述不能替代部署包。；方法官方只验证 Unitree H1；社区迁移到 G1/H1-2 的多个尝试暴露课程阈值和稳定性问题。；Isaac Gym Preview 4 已停止作为 NVIDIA 的主线仿真平台，依赖维护风险较高。
当前建议：主流全身控制技术 2.0 候选，许可证与跨本体迁移待解；统一多步态与上肢干预仍有较高工程价值，但当前最重要的新增证据是跨本体迁移失败而非性能宣传；等待根许可证、官方检查点、部署包和 G1/H1-2 修复后再晋级。
重点标签：recency:2025-release、关联:主流全身控制技术2.0、关联:近期热门、关联:技术改进版、whole-body-control:multi-gait-single-policy、community:cross-robot-failures、artifact:code-present-license-missing、deployment:official-path-unreleased
核验来源：来源 1 · 来源 2 · 来源 3 · 来源 4 · 来源 5 · 来源 6
