Min-Hung (Steve) Chen (@CMHungSteven) / X

Min-Hung (Steve) Chen

1,034 posts

Min-Hung (Steve) Chen

@CMHungSteven

Staff Research Scientist, NVR TW @NVIDIAAI @NVIDIA (Project Lead: DoRA, EoRA, 4D-RGPT, SpatialClaw) | Ph.D. @GeorgiaTech | github.com/cmhungsteve

Taipei City, Taiwan

minhungchen.netlify.app

Joined July 2011

Pinned
Min-Hung (Steve) Chen
@CMHungSteven
May 11, 2022
(1/N) Are you looking for #Vision #Transformer papers in various areas? Check out this list of papers including a broad range of different tasks! github.com/cmhungsteve/Aw… Feel free to share with others😀 @Montreal_AI @machinelearnflx @hardmaru @ak92501 @arankomatsuzaki @omarsar0
GitHub - cmhungsteve/Awesome-Transformer-Attention: An ultimately comprehensive paper list of...
From github.com
Min-Hung (Steve) Chen
@CMHungSteven
Jul 6
2 denoising steps can capture better physics than the full 50-step video. 🤯 That's the insight behind PhaseLock: it locks that early motion prior back into the final high-fidelity output via Latent Δ Guidance — training-free, +6.2 pts physical consistency, ~zero extra compute
00:00
Woojung Han
@woojung0305
Jul 6
ᴅᴏ ᴠɪᴅᴇᴏ ᴅɪꜰꜰᴜꜱɪᴏɴ ᴍᴏᴅᴇʟꜱ ʀᴇᴀʟʟʏ ɴᴏᴛ ᴋɴᴏᴡ ᴘʜʏꜱɪᴄꜱ, ᴏʀ ᴅᴏ ᴛʜᴇʏ ꜰᴏʀɢᴇᴛ ɪᴛ ᴅᴜʀɪɴɢ ɢᴇɴᴇʀᴀᴛɪᴏɴ? 🧐🎬 Want to know more? Please drop by my #icml poster presentation at tomorrow's morning session (Hall A #905!)
Min-Hung (Steve) Chen
@CMHungSteven
Jun 23
Author here — huge thanks to @NVIDIAAI for sharing SpatialClaw 🙏 What excites me most: code becomes the agent’s workspace 🐍 SpatialClaw lets a VLM compose perception tools, inspect intermediate evidence, and revise before answering. #NVIDIAAI #SpatialAI
00:00
NVIDIA AI
@NVIDIAAI
Jun 16
Code is the right action interface for spatial reasoning agents. New from NVIDIA Research: SpatialClaw, a training-free agent that uses code as its action interface for complex visual tasks. Instead of calling a fixed set of pre-defined tools, the agent writes Python inside a
Min-Hung (Steve) Chen
@CMHungSteven
Jun 3
T4V @CVPR is starting!!
Min-Hung (Steve) Chen
@CMHungSteven
May 25
What comes after today’s visual backbones? At T4V @CVPR 2026, we’re bringing the community together for a focused half-day workshop on Transformers for Vision and Multimodal AI — covering image, video, 3D, MLLMs, efficient attention, SSMs/Mamba, and the next generation of visual
Min-Hung (Steve) Chen
@CMHungSteven
Jun 3
T4V @CVPR starts today🚀🚀 1:45–5:40pm, Room 607 Join talks/discussion on Transformers for image, video, 3D, MLLMs, efficient attention & SSMs/Mamba
Min-Hung (Steve) Chen
@CMHungSteven
May 25
What comes after today’s visual backbones? At T4V @CVPR 2026, we’re bringing the community together for a focused half-day workshop on Transformers for Vision and Multimodal AI — covering image, video, 3D, MLLMs, efficient attention, SSMs/Mamba, and the next generation of visual