正在加载视频...

视频加载失败

Introducing the USC Physical Superintelligence (PSI) Lab ( We are rebranding to better reflect our current focus. From here on out, we are tackling one thing: solving robotics and physical intelligence with every model, every bug, and every line of code. And yes, we are hiring at all levels,...

25,776 次观看 • 10 个月前 •via X (Twitter)

18 条评论

Yue Wang 的头像
Yue Wang10 个月前

About one year ago, we completely switched the gear and started the robotic journey. The road was bumpy as we didn’t have much robotic experience. Fortunately, over the year, we have produced some works in this domain and accumulated momentum as listed following. 2/9

Yue Wang 的头像
Yue Wang10 个月前

“RAM: Retrieval-Based Affordance Transfer for Generalizable Zero-Shot Robotic Manipulation” aims to solve zero-shot generalization for manipulation with an affordance-based retrieval model. 3/9

Yue Wang 的头像
Yue Wang10 个月前

“RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning” brings a unified simulation platform for scalable training and evaluation. 4/9

Yue Wang 的头像
Yue Wang10 个月前

“Learning from Massive Human Videos for Universal Humanoid Pose Control” tackles universal humanoid motion generation from human videos. 5/9

Yue Wang 的头像
Yue Wang10 个月前

“Robot Learning from Any Images” converts any in-the-wild images into robotic demonstration, followed by its video generation model version “Robot Learning from a Physical World Model”. and 6/9

Yue Wang 的头像
Yue Wang10 个月前

“AnchorDream: Repurposing Video Diffusion for Embodiment-Aware Robot Data Synthesis” (Junjie’s summer internship project at TRI) learns a video foundation model as a generative data engine. 7/9

Yue Wang 的头像
Yue Wang10 个月前

“ResMimic: From General Motion Tracking to Humanoid Whole-Body Loco-Manipulation via Residual Learning” (Siheng’s internship project at Amazon) enables whole-body manipulation with a two-stage residual framework. 8/9

Yue Wang 的头像
Yue Wang10 个月前

Of course, all of these couldn’t be possible without our great students from the lab and generous supports from TRI, NSF, Bosch, Nvidia, Qualcomm, Amazon, Google, Dolby, Capital One. 9/9

Sabeer Saeed 的头像
Sabeer Saeed10 个月前

Super Outstanding!

Zhijian Liu 的头像
Zhijian Liu10 个月前

Congrats, Yue! Super exciting!

Yue Wang 的头像
Yue Wang10 个月前

Thank you Zhijian!

Michael Elabd 的头像
Michael Elabd10 个月前

Congrats! Excited to see the new research that comes out of PSI

Yue Wang 的头像
Yue Wang10 个月前

Thank you Michael!

Zikang Jiang 的头像
Zikang Jiang10 个月前

congrats!

Eddy Xu 的头像
Eddy Xu10 个月前

lfg!

Jie Wang 的头像
Jie Wang10 个月前

really cool! Excited to see future releases from PSI

Himanshu Kumar 的头像
Himanshu Kumar10 个月前

Yue, this is quite a bold ambition, but I hope the PSI Lab can pull it off, kya? You might find this helpful:

Sam 的头像
Sam10 个月前

do you guys hire remote or help with visa?

相关视频