Hang Qiu
115
posts
@HangQiu
Irvine, CA
Joined March 2012
#CooperScene offers a deployable performance benchmark, pairing multi-vehicle sensor streams with real C-V2X network traces. It examines how cooperative scale stresses the network, and reveals open research gaps in both lightweight representation learning and efficient networking
🎉#ECCV2026 Excited to share our accepted paper: "CooperScene: Multi-Modal Cooperative Autonomy Benchmark with C-V2X Communication Characterization" Huge thanks to my advisor Prof.
@HangQiu, co-first-author Ruoshen Mo and all collaborators. 🔗 Website: cisl.ucr.edu/CooperScene/
00:00
Spatial representations enabled cooperative perception. Is cooperative autonomy feasible by exchanging texts? 🚘💬💬🚖 Check out our paper #TalkingVehicles
[1/n] Can LLM Agents learn to communicate and coordinate in natural language in driving scenarios 🚗 through self-play interactions? Our recent research shows the potential for such learning; interestingly, there is evidence that they develop concise protocols for collaboration!
00:00
Watching KTLA about launching in LA. Proud moment 😎
Angelenos and Austinites: we’ve got news for you. Beginning tomorrow, we’ll start inviting members of our waitlist to ride with us permanently in LA. And in Austin, we’ll continue fully autonomous testing before welcoming public riders later this year. Read more:
00:00
lol we are using some 8020s on our labs car too
Replying to @Waymo
Jan 20, 2009: the first Google owned Prius gets a Velodyne LiDAR on top using 80/20 metal from the Street View workshop on Charleston Rd
