Chenming Zhu

@chenming_eric

Ph.D. Candidate @ HKU-MMLab | 3D Vision & Language for Robotics 🤖

Hong Kong

zcmax.github.io

Replies Reposts Media

Chenming Zhu
@chenming_eric
Dec 18, 2025
MMSI-Video-Bench is the most challenging video spatial intelligence benchmark, which can also be used to test the world model, such as Veo3. 😊👏 Also, we introduce our online video spatial intelligence benchmark in streaming settings for #NeurIPS2025: OST-Bench 🚀
Runsen Xu
@runsen_xu
Dec 16, 2025
1/3 Introducing MMSI-Video-Bench, the most challenging Spatial Intelligence benchmark. Even the strongest MLLM, Gemini 3 Pro, scores only 38%, revealing a ~60% human–AI gap. 🤯 🌐 Project Page: rbler1234.github.io/MMSI-VIdeo-Ben… 💻 GitHub: github.com/InternRobotics… 🤗 Hugging Face:
00:00
1
Chenming Zhu
@chenming_eric
Nov 28, 2025
The most elegant framework for achieving spatial reconstruction and reasoning 👆 The story and analysis is pretty worth reading
Wenbo Hu
@gordonhu608
Nov 28, 2025
🚀 Introducing G^2VLM: Geometry Grounded Vision Language Model with Unified 3D Reconstruction and Spatial Reasoning G^2VLM can natively predicts 3D attributes (depth, camera pose, pointmaps) and uses them for spatial understanding via interleaved reasoning. 🔧 End-to-End
00:00
Chenming Zhu
@chenming_eric
Aug 5, 2025
This is crazy! I’m curious about its generalization capabilities, maybe a new approach for generating the embodied agent training data
Google DeepMind
@GoogleDeepMind
Aug 5, 2025
Replying to @GoogleDeepMind
🔘 Accelerating agent research To explore the potential for agent training, we placed our SIMA agent in a Genie 3 world with a goal. The agent acts, and Genie 3 simulates a response in the world without knowing the objective. This is key for building more capable embodied
00:00
00:00
00:00
Chenming Zhu
@chenming_eric
Nov 19, 2024
@cyodyssey 退圈快2年，感谢Siyuan的饭钱打赏@ABCDELabs
Chenming Zhu
@chenming_eric
Jan 6, 2024
#CPAL2024🇭🇰