←
Tianxing Chen54 posts
Tianxing Chen avatar

Tianxing Chen

@MarioChan2002

PhD student @ HKU-MMLab · Founder of Lumina EAI Community · Work: RoboDojo RoboTwin

◷ Joined January 2023▧ 10 Photos and videos
Tianxing Chen avatar
Tianxing Chen@MarioChan2002·Nov 10

Honored to see RoboDojo included in Anthropic's official evaluation of Claude Fable 5.2! Is the Bitter Lesson coming for robotics? General-purpose foundation models are improving incredibly fast (or maybe our benchmark just isn't challenging enough 😅). For robotics companies, efficiency might be the real moat.

Fable 5.2 shows substantially improved skills in controlling physical robots, even on tasks it has never explicitly been trained to do. On RoboDojo, a benchmark with a simulated two-arm robot, it outperformed other models and most AI systems built to control robots.

0:23
Claude avatar
⟳ Tianxing Chen retweeted
Claude@claudeai·Nov 9

Fable 5.2 shows substantially improved skills in controlling physical robots, even on tasks it has never explicitly been trained to do. On RoboDojo, a benchmark with a simulated two-arm robot, it outperformed other models and most AI systems built to control robots.

0:23
NVIDIA Robotics avatar
⟳ Tianxing Chen retweeted
NVIDIA Robotics@NVIDIARobotics·Nov 6

Scale up testing for complex robot manipulation. 🤖 RoboDojo is now available on NVIDIA Isaac Lab-Arena, bringing challenging bimanual tasks like precision insertion, folding, tool use, and multi-step manipulation to scalable evaluation in simulation.

0:15
Fanqing avatar
⟳ Tianxing Chen retweeted
Fanqing@FanqingMengAI·Oct 8

💡Our view: RSI is a systems engineering problem, not just a model problem. Progress depends on the environment a research agent works in: what resources it can call, what it can change, and how it runs experiments. That environment should reflect real production workflows and be easy for agents to use. At @Evolvent_AI, we built RSIGym around Everything as a Service. Agents can call established research services and iterate on data, training settings, and harness code. We also introduce RSI-Index to measure how well frontier agents jointly improve a target model’s weights and harness. Across 6 research agents and 5 benchmarks, Opus 5 leads at 0.4809. 🧵 We’re releasing the code, experiment configurations, research trajectories, and evaluation logs. Try your own improvement methods with RSIGym. 🌐 Website: rsi-index.ai/ 💻 GitHub: github.com/evolvent-ai/RSIGy… 📄 arXiv: arxiv.org/abs/2610.10310

1:26
General Instinct avatar
⟳ Tianxing Chen retweeted
General Instinct@gen_instinct·Oct 8

What about dis-assembling? Excited to share some results (more coming next week!) deploying pi0.7 on a AI hardware dis-assembling task partnering with @physical_int. The robot autonomously removes retaining bars, riser boards, and fans with only 1.5 hours of post-train.

1:17

Can we build intelligent robots to assemble complex AI hardware? 🦾 At NVIDIA’s Seattle Robotics Lab, we’re tackling that question through GB300 tester-tray assembly, from fastening busbars to inserting delicate connectors. 👇

1:02
Songming Liu avatar
⟳ Tianxing Chen retweeted
Songming Liu@songming_liu·Sep 17

🔥 Liber-0 Preview is now #1 on RoboDojo, with a 30.74 overall score. It performs strongly across long-horizon execution, precision, memory, and generalization — and is the only one 👀 above GPT-6 Astra overall. But the part I find most interesting is this: RoboDojo is a gripper benchmark. Liber-0 is built on a bet that human experience is the largest source of embodied intelligence — and that the real challenge is transferring that intelligence efficiently across embodiments. A gripper is much more constrained than a human hand. Yet after adaptation, that human-derived intelligence still transfers well enough to lead the benchmark. That’s a pretty interesting signal. This gripper benchmark is simply an early test of whether the intelligence transfers. So far, it does. robodojo-benchmark.com/leade…

Post attachment
Tianxing Chen avatar
Tianxing Chen@MarioChan2002·Sep 16

Demo: Humanoid High-level Control and Dexterous Piano Playing

0:06
0:08
Tianxing Chen avatar
⟳ Tianxing Chen retweeted
Tianxing Chen@MarioChan2002·Sep 16

We put GPT-6 Astra in the RoboDojo. 🥋🤖 The RoboDojo Team conducted a comprehensive evaluation of GPT-6 Astra as an embodied agent, including: • RoboDojo Sim & Real, compared with GPT-5.5 and DeepSeek-Flash • Humanoid high-level control • Dexterous piano playing with RoboPianist 🎹 • A systematic study of in-context learning (ICL) Our key takeaway: GPT-6 Astra demonstrates remarkably strong semantic and spatial understanding, together with impressive in-context adaptation. At the same time, physical commonsense remains a clear bottleneck — revealing an important gap between understanding the world and truly reasoning about its physics. Full report & demos: robodojo-benchmark.com/repor… @_wenbozhang (project lead), @wenhaocha1, @frankzydou, @JinWeiyang18434, @YutaoOuyang, @minifullcapsule, @x_h_ucb, @YutaoOuyang, @YueChen614

Post attachment
Siqiao Huang avatar
⟳ Tianxing Chen retweeted
Siqiao Huang@KnightNemo_·Sep 9

🤖 World Action Models predict the future to act better. However, WAM research remains fragmented in code, ad hoc in design, and unprincipled at scale. 💡 Today, we release 𝗢𝗽𝗲𝗻𝗪𝗔𝗠: 𝗔𝗻 𝗢𝗽𝗲𝗻, 𝗠𝗼𝗱𝘂𝗹𝗮𝗿 𝗘𝘅𝗽𝗹𝗼𝗿𝗮𝘁𝗶𝗼𝗻 𝗧𝗼𝘄𝗮𝗿𝗱𝘀 𝗦𝘆𝘀𝘁𝗲𝗺𝗮𝘁𝗶𝗰 𝗪𝗼𝗿𝗹𝗱–𝗔𝗰𝘁𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹 𝗣𝗿𝗲𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴 to change this. 🌐 openwam-official.github.io/ 🧵1/14

3:00
Zhiyang (Frank) Dou avatar
⟳ Tianxing Chen retweeted
Zhiyang (Frank) Dou@frankzydou·Sep 11

I kept seeing #GPT-6 Astra modelling results on here and got curious enough to build the thing myself: an agentic pipeline that turns a plain RGB video into 3D assets. The living room for robots @MIT_CSAIL. Input: one handheld phone walkthrough of our lab kitchen (~20s). Nothing else — no depth sensor, no CAD, no asset library. ~1 day (I actually slept overnight), including human-in-the-loop. Live: frank-zy-dou.github.io/kitch… It is still not perfect — thin and shiny things are still weak, a few objects are drafts, the room shell needs another pass. The loop, roughly: - Monocular video → metric scan (ViPE): camera poses + depth. This is the measuring instrument, not the output. - GPT-6 lists what should exist as separate objects, then open-vocabulary detection + tracking gives per-object masks; each object is fused and measured in metres. - For every asset, GPT-6 works in its own sandbox with tools: it writes the object as a program in a small Blender DSL (closed solids, PBR colours, hinges/drawers), builds it, renders it over the original video frames, compares it with the scan points in 3D, and iterates. - A separate GPT-6 session is the verifier — it can render the model on any frame it wants and must justify every complaint with a frame. The modeller never grades its own work. - A completeness pass renders the whole modelled scene from the video's own cameras, puts it next to the real frames, and says what's still missing. What the detector keeps missing (a row of identical cabinets) gets placed geometrically instead. - Out comes MJCF/URDF with joints. Now, adding simulation to this env.

0:21
Yuxin Chen avatar
⟳ Tianxing Chen retweeted
Yuxin Chen@ThomasYuxinChen·Aug 3

🤖 One brain, for any robot. But how does a generalist robot achieve mastery on a specific task? 🚀 Introducing CLIFT (Closed-Loop Iterative Fine-Tuning), a framework that turns Gemini Robotics On-Device into a humanoid specialist using only a managed fine-tuning API. [1/6] 🌐 thomaschen98.github.io/clift… 📄 arxiv.org/abs/2607.29172

1:12
Tianxing Chen avatar
Tianxing Chen@MarioChan2002·Jul 10

There seems to be an issue with the video and audio. You can visit the homepage to watch the high-definition version🥲.

Tianxing Chen avatar
Tianxing Chen@MarioChan2002·Jul 9

Fairness is our first principle. RoboDojo is operated by AI MMLab Club, a non-profit organization, together with academic partners worldwide. The benchmark, reproduced code, and checkpoints are open-sourced, with no commercial involvement. We welcome academic teams to join us.

Tianxing Chen avatar
Tianxing Chen@MarioChan2002·Jul 9

Together with RoboDojo, we introduce XPolicyLab: a unified framework for embodied model development, deployment, and evaluation. Using XPolicyLab, we reproduced 30+ models and built a comprehensive leaderboard for the community. XPolicyLab code: github.com/XPolicyLab/XPolic…

Tianxing Chen avatar
Tianxing Chen@MarioChan2002·Jul 9

RoboDojo includes 42 simulation tasks across 5 evaluation dimensions and 18 real-world tasks. We fully open-source the benchmark code and support global remote real-world evaluation, so researchers can evaluate models under a unified protocol.