Yinpei Dai

@YinpeiD

University of Michigan, CSE PhD @UMichCSE @michigan_AI, research in emobodied AI and robotics.

Ann Arbor, MI

Joined March 2021

Replies Reposts Media

Pinned
Yinpei Dai
@YinpeiD
Mar 9
Robot memory methods are growing fast, but systematic evaluation is largely lacking. 📉 Introducing RoboMME: a new benchmark for memory-augmented robotic manipulation! 🤖🧠 Featuring 16 tasks across temporal, spatial, object, and procedural memory 🔗 robomme.github.io
00:00
7
Yinpei Dai
@YinpeiD
Jul 23
Great conversation with @micoolcho, @chris_j_paxton, and @liu_yuejiang ! Robot memory requires strong multimodal understanding to support reactive action prediction. Excited to see how people build on RoboMME and push this area forward!
RoboPapers
@RoboPapers
Jul 23
Memory is one of the most important problems in robotics. Long horizon memory is key for a variety of robot manipulation problems. However, there exist no good benchmarks for understanding progress in how well generalist robot policies can understand language. @YinpeiD and
00:00
1
Yinpei Dai
@YinpeiD
Jul 20
Big thanks to @micoolcho and @chris_j_paxton for hosting such a fun conversation! Memory helps robots tackle long-horizon tasks, adapt, and improve—and RoboMME shows how close we are. Can’t wait for the full episode!
RoboPapers
@RoboPapers
Jul 20
Full episode dropping soon! Geeking out with @YinpeiD @liu_yuejiang on RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies robomme.github.io Co-hosted by @micoolcho @chris_j_paxton
00:00
1
Yinpei Dai
@YinpeiD
Jun 2
Yuejiang is an amazing researcher and a great person to work with. If you’re interested in embodied AI and looking for a PhD position, LEMA Lab is absolutely worth considering!
Yuejiang Liu
@liu_yuejiang
Jun 2
Excited to share that I’ll join @NUSComputing as an Assistant Professor in 2027 🏛️ I’ll build LEMA Lab: lema-nus.github.io, study the principles of embodied intelligence, & empower every lab member to thrive 📢 Recruiting 3-6 PhD students in the next application cycles
1
Yinpei Dai
@YinpeiD
May 25
Excited to share that RoboMME has been selected for an oral presentation at ICML 2026! 🎉 Huge thanks to my collaborators and the reviewers for their valuable feedback and support!!
Yinpei Dai
@YinpeiD
Mar 9
Robot memory methods are growing fast, but systematic evaluation is largely lacking. 📉 Introducing RoboMME: a new benchmark for memory-augmented robotic manipulation! 🤖🧠 Featuring 16 tasks across temporal, spatial, object, and procedural memory 🔗 robomme.github.io
00:00
2