Super excited to introduce 🐹 HAMSTER—our VLA model accepted to #ICLR2025! 🚀. We train a VLM to predict high-level trajectories, executed by a 3D policy like RVT-2. This decomposition enables better generalization and efficient use of off-domain cheap data. Check out.

4. Flowing from Words to Pixels. An insight that seems so simple in hindsight. For conditional generation, instead of starting from noise, why not flow directly from source to the target distri.? I'll be watching closely if this becomes the norm. Great Work by @Qihao Liu et al.

3. Prompting Depth Anything for 4K Metric Depth . It’s a very practical way to get dense and accurate metric depth. It upgrades a monocular depth models for metric accuracy by using data from metric sensors, getting the best of both worlds. Great work by Haotong Lin et al.

2. Reconstructing Animals and the Wild. This work generates complete scenes from natural images, trained with just synthetic Infinigen data. While working on Infinigen, I never thought it could be used so creatively. Fantastic work by Peter @Michael_J_Black @silvia_zuffi

That’s a wrap for #CVPR2025! Here's a 🧵 of some really cool works 👇. 1. Let Humanoids Hike!. Great work @ky_lin0305 and Stella Xu. They drove home the point that we can't treat locomotion and navigation as separate. The ultimate test: Can your robot complete a hike on its own?

Headed to Nashville for #CVPR2025! . Excited to present our robot learning work, "3D-MVP," and connect with fellow researchers. I'll be sharing my thoughts on the most exciting papers. Should be a fun week of learning and meeting friends, old and new! #Computervision #robotics

RT @Ishika_S_: VLAs have the potential to generalize over scenes and tasks, but require a ton of data to learn robust policies. We introdu….

Reasoning has taken off in LLMs – how does it translate to Robotics? . At RSS 2025, we're organizing the "R³: Reasoning for Robust Robot Manipulation" workshop. Please consider submitting your work and join us. We'd love to hear you!. Deadline: May 25.

RT @NVIDIARobotics: 🎊 That's a wrap on #ICLR2025. Shout out to all the amazing research in #robotics, machine vision, and more. Missed it….

3D-MVP has been accepted to #CVPR2025! . Our research demonstrates that 3D pretraining significantly outperforms other pretraining approaches in robotics. By integrating this pretraining into the Robotic View Transformer (RVT), we enhance downstream policy performance.

RT @abhishekunique7: Over the last few months, we’ve been thinking about how to learn from “off-domain” data - data from non-robot sources….

A few weeks have passed since DeepSeek became an overnight sensation, and I’ve had time to reflect on its impact. DeepSeek has undeniably opened up new possibilities for LLM research—both for researchers and smaller companies. — It’s not easy, but it demonstrates that a small,.

Applications are open for Robotics Research Scientists at NVIDIA. We are based in Seattle and the Bay Area and are led by Prof. Dieter Fox. A team-wide committee selects candidates. Please consider applying! . Priority topics include dexterous, bi-manual, and mobile robot.

Application link in case the one in the post doesn’t work:

Applications are open for PhD Research Intern in Robotics NVIDIA for Summer 2025. We are based in Seattle + Bay Area, led by Prof. Dieter Fox. Candidates are selected by a team-wide committee. Please consider applying! . Application: More info about.

A very well deserved recognition to the pioneers of neural networks! The Nobel prize is timely and marks the acknowledgment of the transformative power of AI.

Thanks for the invitation! I am looking forward to presenting @MontrealRobots.

Applications are now open for NVIDIA Graduate Fellowship. - For PhD students who have completed at least one year.- Deadline Sep 13.- Requires a research proposal and 2 LoRs.- Includes a summer internship.