Kashyap Chitta
@kashyap7x.bsky.social
kashyap7x.github.io
Postdoc at NVIDIA. Previously at the University of Tübingen and CMU. Robot Learning, Autonomous Driving.
Postdoc at NVIDIA. Previously at the University of Tübingen and CMU. Robot Learning, Autonomous Driving.
Reposted by Kashyap Chitta
Interested in ✨world models✨? I just open-sourced an implementation of the Dreamer 4 world model. It's in PyTorch and comes with a pretrained model + a neat little web interface that lets you interact with any of 30 DMControl tasks that I trained it on!
Link: github.com/nicklashanse...
Link: github.com/nicklashanse...
January 15, 2026 at 6:20 PM
Interested in ✨world models✨? I just open-sourced an implementation of the Dreamer 4 world model. It's in PyTorch and comes with a pretrained model + a neat little web interface that lets you interact with any of 30 DMControl tasks that I trained it on!
Link: github.com/nicklashanse...
Link: github.com/nicklashanse...
Reposted by Kashyap Chitta
cseweb.ucsd.edu/~tzli/novelt...
I gave an internal talk at UCSD last year regarding "novelty" in computer science research. In it I "debunked" some of the myth people seem to have about what is good research in computer science these days. People seemed to like it, so I thought I should share.
I gave an internal talk at UCSD last year regarding "novelty" in computer science research. In it I "debunked" some of the myth people seem to have about what is good research in computer science these days. People seemed to like it, so I thought I should share.
cseweb.ucsd.edu
January 9, 2026 at 5:21 PM
cseweb.ucsd.edu/~tzli/novelt...
I gave an internal talk at UCSD last year regarding "novelty" in computer science research. In it I "debunked" some of the myth people seem to have about what is good research in computer science these days. People seemed to like it, so I thought I should share.
I gave an internal talk at UCSD last year regarding "novelty" in computer science research. In it I "debunked" some of the myth people seem to have about what is good research in computer science these days. People seemed to like it, so I thought I should share.
Simple and efficient transformer based end-to-end driving. If we could give out another innovation award for NAVSIM, it would go to this work!
valeoai.github.io/driving-on-r...
valeoai.github.io/driving-on-r...
Driving on Registers
Driving on Registers: Simple and efficient transformer based end-to-end driving
valeoai.github.io
January 9, 2026 at 11:47 AM
Simple and efficient transformer based end-to-end driving. If we could give out another innovation award for NAVSIM, it would go to this work!
valeoai.github.io/driving-on-r...
valeoai.github.io/driving-on-r...
Reposted by Kashyap Chitta
While we also hype the moderation features, I'm really excited about the paper discovery tools @mariaa.bsky.social and I are starting to build. Open social means we can bootstrap onto existing discussions that are happening
January 6, 2026 at 12:58 AM
While we also hype the moderation features, I'm really excited about the paper discovery tools @mariaa.bsky.social and I are starting to build. Open social means we can bootstrap onto existing discussions that are happening
Reposted by Kashyap Chitta
Wrapping up 2025 with a review of some recent work, led by several amazing students and collaborators: Yixuan Pan, Ruoyi Qiao, @jiazhiyang.bsky.social, Shuhan Tan, Brayden Zhang, Shihao Li, @longpollehn.bsky.social, Peter Karkus, and @maxigl.bsky.social!
kashyap7x.substack.com/p/2025-resea...
kashyap7x.substack.com/p/2025-resea...
2025 Research Wrap-Up
Seven research contributions from this fall spanning heterogeneous datasets, latent reasoning, constrained trajectory diffusion, and robust driving policies
kashyap7x.substack.com
December 31, 2025 at 9:24 PM
Wrapping up 2025 with a review of some recent work, led by several amazing students and collaborators: Yixuan Pan, Ruoyi Qiao, @jiazhiyang.bsky.social, Shuhan Tan, Brayden Zhang, Shihao Li, @longpollehn.bsky.social, Peter Karkus, and @maxigl.bsky.social!
kashyap7x.substack.com/p/2025-resea...
kashyap7x.substack.com/p/2025-resea...
Wrapping up 2025 with a review of some recent work, led by several amazing students and collaborators: Yixuan Pan, Ruoyi Qiao, @jiazhiyang.bsky.social, Shuhan Tan, Brayden Zhang, Shihao Li, @longpollehn.bsky.social, Peter Karkus, and @maxigl.bsky.social!
kashyap7x.substack.com/p/2025-resea...
kashyap7x.substack.com/p/2025-resea...
2025 Research Wrap-Up
Seven research contributions from this fall spanning heterogeneous datasets, latent reasoning, constrained trajectory diffusion, and robust driving policies
kashyap7x.substack.com
December 31, 2025 at 9:24 PM
Wrapping up 2025 with a review of some recent work, led by several amazing students and collaborators: Yixuan Pan, Ruoyi Qiao, @jiazhiyang.bsky.social, Shuhan Tan, Brayden Zhang, Shihao Li, @longpollehn.bsky.social, Peter Karkus, and @maxigl.bsky.social!
kashyap7x.substack.com/p/2025-resea...
kashyap7x.substack.com/p/2025-resea...
Reposted by Kashyap Chitta
Our new E2E driving method, TransFuser v6, is out on ArXiv.
It outperforms all other methods on CARLA by a wide margin, 95 DS on Bench2Drive!
We show that minimizing the asymmetry between data annotator and policy is key for strong IL results.
Code, models, and paper:
ln2697.github.io/lead/
It outperforms all other methods on CARLA by a wide margin, 95 DS on Bench2Drive!
We show that minimizing the asymmetry between data annotator and policy is key for strong IL results.
Code, models, and paper:
ln2697.github.io/lead/
December 27, 2025 at 1:42 AM
Our new E2E driving method, TransFuser v6, is out on ArXiv.
It outperforms all other methods on CARLA by a wide margin, 95 DS on Bench2Drive!
We show that minimizing the asymmetry between data annotator and policy is key for strong IL results.
Code, models, and paper:
ln2697.github.io/lead/
It outperforms all other methods on CARLA by a wide margin, 95 DS on Bench2Drive!
We show that minimizing the asymmetry between data annotator and policy is key for strong IL results.
Code, models, and paper:
ln2697.github.io/lead/
Reposted by Kashyap Chitta
🧥 Live-stream robotic teamwork that folds clothes. 6 clothes in 3 minutes straight.
χ₀ = 20hrs data + 8 A100s + 3 key insights:
- Mode Consistency: align your distributions
- Model Arithmetic: merge, don't retrain
- Stage Advantage: pivot wisely
🔗 mmlab.hk/research/kai0 checkout 3mins demo
χ₀ = 20hrs data + 8 A100s + 3 key insights:
- Mode Consistency: align your distributions
- Model Arithmetic: merge, don't retrain
- Stage Advantage: pivot wisely
🔗 mmlab.hk/research/kai0 checkout 3mins demo
December 24, 2025 at 9:25 AM
🧥 Live-stream robotic teamwork that folds clothes. 6 clothes in 3 minutes straight.
χ₀ = 20hrs data + 8 A100s + 3 key insights:
- Mode Consistency: align your distributions
- Model Arithmetic: merge, don't retrain
- Stage Advantage: pivot wisely
🔗 mmlab.hk/research/kai0 checkout 3mins demo
χ₀ = 20hrs data + 8 A100s + 3 key insights:
- Mode Consistency: align your distributions
- Model Arithmetic: merge, don't retrain
- Stage Advantage: pivot wisely
🔗 mmlab.hk/research/kai0 checkout 3mins demo
Reposted by Kashyap Chitta
What's left to do in self-driving given Waymo is taking off? An argument that it's still a great research problem:
open.substack.com/pub/emergere...
open.substack.com/pub/emergere...
Why study self-driving?
Isn't it "solved"?
open.substack.com
December 21, 2025 at 4:21 AM
What's left to do in self-driving given Waymo is taking off? An argument that it's still a great research problem:
open.substack.com/pub/emergere...
open.substack.com/pub/emergere...
Reposted by Kashyap Chitta
More details at tuebingen.ai/news/ai-powe...
AI-powered assistants for scientific discovery
Andreas Geiger receives ERC Consolidator Grant
tuebingen.ai
December 11, 2025 at 3:56 PM
More details at tuebingen.ai/news/ai-powe...
Reposted by Kashyap Chitta
Reposted by Kashyap Chitta
Speaking of RL, Nvidia also just published a survey on the importance of closed-loop training (RL, etc.) in E2E driving.
research.nvidia.com/publication/...
research.nvidia.com/publication/...
Beyond Behavior Cloning in Autonomous Driving: a Survey of Closed-Loop Training Techniques | Research
Behavior cloning, the dominant approach for training autonomous vehicle (AV) policies, suffers from a fundamental gap: policies trained open-loop on temporally independent samples must operate in clos...
research.nvidia.com
December 9, 2025 at 7:29 PM
Speaking of RL, Nvidia also just published a survey on the importance of closed-loop training (RL, etc.) in E2E driving.
research.nvidia.com/publication/...
research.nvidia.com/publication/...
Reposted by Kashyap Chitta
Attending #Neurips2025? Get your personalized Scholar Inbox conference program now to easily navigate the poster sessions and find what you are looking for:
www.scholar-inbox.com/conference/n...
www.scholar-inbox.com/conference/n...
December 2, 2025 at 6:37 AM
Attending #Neurips2025? Get your personalized Scholar Inbox conference program now to easily navigate the poster sessions and find what you are looking for:
www.scholar-inbox.com/conference/n...
www.scholar-inbox.com/conference/n...
Reposted by Kashyap Chitta
I'll be at #NeurIPS2025 in San Diego from Thu to Sat, and I am looking for PostDocs in Embodied AI, particularly in world modeling and simulator learning. Please reach out if you are interested.
December 1, 2025 at 5:17 PM
I'll be at #NeurIPS2025 in San Diego from Thu to Sat, and I am looking for PostDocs in Embodied AI, particularly in world modeling and simulator learning. Please reach out if you are interested.
Reposted by Kashyap Chitta
Wondering how DeepSeek v3.2 rivals SOTA models (e.g., GPT5/Gemini 3 pro) while being ~30x cheaper? 🤔
Let's learn how the base model works!
We'll focus on attention, the need for KV caching, and key ideas for improving attention (MQA/GQA/MLA/DSA).
youtu.be/Y-o545eYjXM
Let's learn how the base model works!
We'll focus on attention, the need for KV caching, and key ideas for improving attention (MQA/GQA/MLA/DSA).
youtu.be/Y-o545eYjXM
December 1, 2025 at 6:23 PM
Wondering how DeepSeek v3.2 rivals SOTA models (e.g., GPT5/Gemini 3 pro) while being ~30x cheaper? 🤔
Let's learn how the base model works!
We'll focus on attention, the need for KV caching, and key ideas for improving attention (MQA/GQA/MLA/DSA).
youtu.be/Y-o545eYjXM
Let's learn how the base model works!
We'll focus on attention, the need for KV caching, and key ideas for improving attention (MQA/GQA/MLA/DSA).
youtu.be/Y-o545eYjXM
Reposted by Kashyap Chitta
🚀 Introducing TMLR Beyond PDF!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
November 25, 2025 at 4:12 PM
🚀 Introducing TMLR Beyond PDF!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
Reposted by Kashyap Chitta
TMLR (@tmlrorg.bsky.social) is now proud to support interactive HTML-based submissions, going "Beyond PDF" -- check it out!
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.
🚀 Introducing TMLR Beyond PDF!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
November 25, 2025 at 4:14 PM
TMLR (@tmlrorg.bsky.social) is now proud to support interactive HTML-based submissions, going "Beyond PDF" -- check it out!
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.
Reposted by Kashyap Chitta
Excellent speaker lineup for the @naverlabseurope.bsky.social AI for Robotics Workshop.
For those at home, the event is live-streamed on the landing page: europe.naverlabs.com/updates/ai4r...
For those at home, the event is live-streamed on the landing page: europe.naverlabs.com/updates/ai4r...
November 20, 2025 at 9:05 AM
Excellent speaker lineup for the @naverlabseurope.bsky.social AI for Robotics Workshop.
For those at home, the event is live-streamed on the landing page: europe.naverlabs.com/updates/ai4r...
For those at home, the event is live-streamed on the landing page: europe.naverlabs.com/updates/ai4r...
Reposted by Kashyap Chitta
We’re live! 🚀 Streaming: tinyurl.com/bdtk2nzs
The International Workshop on AI4Robotics by @naverlabseurope
2dys of Spatial AI, SLAM, robot learning, HRI, autonomy
This AM CET: @martinhumenberger.bsky.social @marcpollefeys.bsky.social Andrea Vedaldi Cordelia Schmid & @andrewdavidson.bsky.social ⬇️
The International Workshop on AI4Robotics by @naverlabseurope
2dys of Spatial AI, SLAM, robot learning, HRI, autonomy
This AM CET: @martinhumenberger.bsky.social @marcpollefeys.bsky.social Andrea Vedaldi Cordelia Schmid & @andrewdavidson.bsky.social ⬇️
November 20, 2025 at 8:40 AM
We’re live! 🚀 Streaming: tinyurl.com/bdtk2nzs
The International Workshop on AI4Robotics by @naverlabseurope
2dys of Spatial AI, SLAM, robot learning, HRI, autonomy
This AM CET: @martinhumenberger.bsky.social @marcpollefeys.bsky.social Andrea Vedaldi Cordelia Schmid & @andrewdavidson.bsky.social ⬇️
The International Workshop on AI4Robotics by @naverlabseurope
2dys of Spatial AI, SLAM, robot learning, HRI, autonomy
This AM CET: @martinhumenberger.bsky.social @marcpollefeys.bsky.social Andrea Vedaldi Cordelia Schmid & @andrewdavidson.bsky.social ⬇️
Reposted by Kashyap Chitta
A fascinating and historic panel discussion with six of the recipients of the 2025 Queen Elizabeth Prize for Engineering, honoring the critical interplay between Algorithms, Data, and Compute that gave rise to today’s remarkable advances in AI and Machine Learning
The Minds of Modern AI: Jensen Huang, Geoffrey Hinton, Yann LeCun & the AI Vision of the Future
YouTube video by FT Live
youtu.be
November 9, 2025 at 9:14 AM
A fascinating and historic panel discussion with six of the recipients of the 2025 Queen Elizabeth Prize for Engineering, honoring the critical interplay between Algorithms, Data, and Compute that gave rise to today’s remarkable advances in AI and Machine Learning
Reposted by Kashyap Chitta
Launching the Physical AI AV Dataset! 🚀
huggingface.co/datasets/nvi...
One of the largest, most diverse & commercially usable open-source datasets for AVs.
- 1727 hours of driving data
- Camera, LiDAR, & radar
- 25 countries, 2500+ cities
This is just the beginning, more features to come!
huggingface.co/datasets/nvi...
One of the largest, most diverse & commercially usable open-source datasets for AVs.
- 1727 hours of driving data
- Camera, LiDAR, & radar
- 25 countries, 2500+ cities
This is just the beginning, more features to come!
October 28, 2025 at 5:59 PM
Launching the Physical AI AV Dataset! 🚀
huggingface.co/datasets/nvi...
One of the largest, most diverse & commercially usable open-source datasets for AVs.
- 1727 hours of driving data
- Camera, LiDAR, & radar
- 25 countries, 2500+ cities
This is just the beginning, more features to come!
huggingface.co/datasets/nvi...
One of the largest, most diverse & commercially usable open-source datasets for AVs.
- 1727 hours of driving data
- Camera, LiDAR, & radar
- 25 countries, 2500+ cities
This is just the beginning, more features to come!
Reposted by Kashyap Chitta
X-mas came earlier this year!
Nvidia has just released the huge Physical AI AV Dataset
- 1727 hrs of driving data: 310K clips of 20s
- sensor rig: 7 cameras, lidar, radar
- 25 countries, 2.5K cities from US + Europe
Kudos to @kashyap7x.bsky.social et al.!
huggingface.co/datasets/nvi...
Nvidia has just released the huge Physical AI AV Dataset
- 1727 hrs of driving data: 310K clips of 20s
- sensor rig: 7 cameras, lidar, radar
- 25 countries, 2.5K cities from US + Europe
Kudos to @kashyap7x.bsky.social et al.!
huggingface.co/datasets/nvi...
October 28, 2025 at 9:59 PM
X-mas came earlier this year!
Nvidia has just released the huge Physical AI AV Dataset
- 1727 hrs of driving data: 310K clips of 20s
- sensor rig: 7 cameras, lidar, radar
- 25 countries, 2.5K cities from US + Europe
Kudos to @kashyap7x.bsky.social et al.!
huggingface.co/datasets/nvi...
Nvidia has just released the huge Physical AI AV Dataset
- 1727 hrs of driving data: 310K clips of 20s
- sensor rig: 7 cameras, lidar, radar
- 25 countries, 2.5K cities from US + Europe
Kudos to @kashyap7x.bsky.social et al.!
huggingface.co/datasets/nvi...
Reposted by Kashyap Chitta
Hundreds of hours of European driving data from NVIDIA! 1700 hours total
Big day for autonomous driving research.
Nvidia just dropped 1700 hours of public driving data on HuggingFace from over 2500 cities:
huggingface.co/datasets/nvi...
Nvidia just dropped 1700 hours of public driving data on HuggingFace from over 2500 cities:
huggingface.co/datasets/nvi...
huggingface.co
October 28, 2025 at 8:20 PM
Hundreds of hours of European driving data from NVIDIA! 1700 hours total
Reposted by Kashyap Chitta
Big day for autonomous driving research.
Nvidia just dropped 1700 hours of public driving data on HuggingFace from over 2500 cities:
huggingface.co/datasets/nvi...
Nvidia just dropped 1700 hours of public driving data on HuggingFace from over 2500 cities:
huggingface.co/datasets/nvi...
huggingface.co
October 28, 2025 at 6:03 PM
Big day for autonomous driving research.
Nvidia just dropped 1700 hours of public driving data on HuggingFace from over 2500 cities:
huggingface.co/datasets/nvi...
Nvidia just dropped 1700 hours of public driving data on HuggingFace from over 2500 cities:
huggingface.co/datasets/nvi...