Skip to content
View rzc156's full-sized avatar
2 followers · 4 following

Achievements

Achievement: Starstruckx3

Achievements

Achievement: Starstruckx3

Block or report rzc156

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
rzc156/README.md 
rzc156@github:~$ whoami
<OuO>  ──  Embodied AI · Robot Learning researcher

rzc156@github:~$ cat ~/profile.json
{
  "affiliation" : "USTC · 电子信息 (M.Sc.)",
  "focus"       : ["World Action Model (WAM)", "Vision-Language-Action", "World Models", "Test-time Selection", "Manipulation"],
  "stack"       : ["PyTorch", "LeRobot", "LIBERO", "IsaacLab", "MuJoCo", "ROS2", "C++20"],
  "belief"      : "离线指标不代理决策质量 —— 先把指标对齐决策，再谈收益。",
  "status"      : "open to Embodied-AI / Robot-Learning internships 🟢"
}

rzc156@github:~$ ./run --now
[ ok ] loading world model ......... done
[ ok ] sampling candidate actions . done (K=8)
[ .. ] selecting the non-obvious ... this is the hard part

rzc156@github:~$ echo "$WAM"
World Action Model  →  在潜空间里「想象」，再把想象变成监督信号。
↓  scroll  ↓

~/stack

Python PyTorch CUDA C++ Linux ROS2 IsaacLab MuJoCo Git Docker

~/featured-repos

repo stars one-liner
dreamforge stars Turn world-model imagination into corrective training data for robot policies.
Embodied-Robot-Intelligence stars 研究级教材 · 覆盖操作 / VLA / 世界模型 / 人形 / Agentic Robotics。
near-zero-headroom stars Test-time action selection for VLA policies — 论文在投。

~/stats

metrics contribution calendar streak contribution snake 
  " 有得选才值钱 —— Action Vacancy, 4% of decisions.  "

⚡ built with a terminal, a green wall, and an unreasonable amount of paired replays

Popular repositories Loading

  1. dreamforge dreamforge Public

    Turn world-model imagination into corrective training data for robot policies — a single-GPU, falsifiable pipeline (SmolVLA · LIBERO · CEM · paired simulator replays).

    Python 130 3

  2. vla-diag vla-diag Public

    Check whether a VLA benchmark number is measuring what you think it measures — checkpoint integrity, failure taxonomy, headroom & power analysis (zero-dependency, framework-agnostic).

    Python 10

  3. Embodied-Robot-Intelligence Embodied-Robot-Intelligence Public

    A research-grade textbook and engineering handbook on embodied robot intelligence: robotics foundations, manipulation, locomotion, loco-manipulation, VLA, world models, humanoid control, agentic ro…

    4

  4. near-zero-headroom near-zero-headroom Public

    Research project on test-time action selection for VLA policies — manuscript under review; code, results & paper will be released after publication.

    Makefile 1

  5. rzc156 rzc156 Public

    Python