Research Intern @NVIDIA | PhD Student @UofT @VectorInst
Toronto, Canada
Joined September 2011
- Tweets80
- Following407
- Followers397
- Likes1.1K
Sven Elflein retweeted
Excited to share our new Cosmos3-Nano–based ArtiFixer is now out 🚀 Almost matching the quality of the Wan 2.1 14B based one with just 8B parameters in the generation tower. And most importantly 3–4× faster inference while using about half the GPU memory! research.nvidia.com/labs/sil…
Sven Elflein retweeted
Check out Cosmos Simulation: Interactive World Simulation with Cosmos for Physical AI:
🌐Website: research.nvidia.com/labs/cos…
📜Report: research.nvidia.com/labs/cos…
🤖 Models: huggingface.co/collections/n…
Sven Elflein retweeted
We are hiring interns for next summer! If you are interested, reach out and apply here:
nvidia.wd5.myworkdayjobs.com…
Come build the future of simulation for AV and robotics at @nvidia! We are looking for interns to work on generative AI, world models, reconstruction, policy learning, VLMs, geometry and scene understanding. Apply here!
nvidia.wd5.myworkdayjobs.com…
Sven Elflein retweeted
At @VeedaAI we are growing our team across all fields and locations (Zurich, Bay Area, Toronto)! Come join us and help build the future of Physical AI! My DMs are open!
At @VeedaAI , we’re building for a future where robots learn by interacting with the world, at scale -- in simulated reality.
Getting there means bringing together the brightest minds to tackle some of the hardest problems in Physical AI.
Come build what’s next with us: jobs.ashbyhq.com/veeda-ai
Sven Elflein retweeted
Come build the future of simulation for AV and robotics at @nvidia! We are looking for interns to work on generative AI, world models, reconstruction, policy learning, VLMs, geometry and scene understanding. Apply here!
nvidia.wd5.myworkdayjobs.com…
Sven Elflein retweeted
We’re hiring PhD interns for NVIDIA Research's Learning and Perception Research (LPR) team!
We are passionate about embodied AI research that pushes boundaries while making real-world impact. We are particularly interested in agentic reasoning, multimodal and continual learning, AI scientists, and beyond.
To be considered, you must be a currently enrolled PhD student with a promising research track record (publications and/or core contributor to open-source projects). Please apply directly at the link below!
Listing: nvidia.wd5.myworkdayjobs.com…
Team homepage: research.nvidia.com/labs/lpr…
Sven Elflein retweeted
We'll present Déjà View as Oral at #NeurIPS 🎉
TL;DR: we beat VGGT-Ω on average with a 10x smaller, looped architecture.
Check it out 👇
Sven Elflein retweeted
This reminded me of NVIDIA’s Deja Vu.
Deja Vu repeats a single ~10M-parameter Transformer block K times and can approach VGGT-omega 1B performance.
For 3D reconstruction, this idea feels especially natural. The model may not need huge parameter count to store knowledge. A lot of the work is repeatedly updating correspondence, geometry, and consistency from the same observations.
Maybe scaling reconstruction/reasoning is less about parameter count, and more about recurrent compute depth.
OpenAI’s Astra AI uses a new reasoning approach called “recurrent depth.” Though it can help model costs and performance, researchers are concerned bc it obscures a model’s thinking process, making it more difficult to monitor.
w/ @amir @rocketalignment
theinformation.com/articles/…
Sven Elflein retweeted
New NVIDIA PhD research internships for 2027 🔍
Generative AI, LLMs, vision, graphics/simulation, robotics, autonomous vehicles.
I’m interested in AI systems that automate and improve research.
Apply: nvidia.wd5.myworkdayjobs.com…
Sven Elflein retweeted
CFP: Continual World Models Workshop @ NeurIPS 2026
🗓 Submission deadline: Aug 29, 2026
📍 Sydney, Australia
🌐 continual-world-models-works…
World models are increasingly capable of representing complex environments, yet most are still trained largely as static models. A central challenge is building world models that continually learn and adapt to a changing world.
At CWM @ NeurIPS 2026, we bring together researchers across world models, continual learning, robotics, multimodal learning, memory, and agents.
We are also recruiting reviewers! If you are interested in becoming a reviewer, please submit your information through this form:
docs.google.com/forms/d/e/1F…
We hope to see your work and have you join us at the workshop!
#NeurIPS2026 #WorldModels #ContinualLearning #EmbodiedAI #Robotics
Sven Elflein retweeted
Update: FlowBender code is public 🚀
✅ Depth → RGB
✅ Edge map → RGB
✅ Super-resolution
✅ JPEG restoration
✅ 3D texturing
Try it: github.com/DanielGilo/FlowBe… Project page: flow-bender.github.io/
Sven Elflein retweeted
Want to stay rooted in reality 🌍, embrace the power of AI 🤖, and get fully interactive with cutting-edge research and 3D Gaussian Splat captures? 📸✨ Join us at SIGGRAPH: kaolin.readthedocs.io/en/lat…
Sven Elflein retweeted
Excited to share our #ICCP2026 (PAMI Track!) work on Raw image denoising. We tackle two important problems:
- Low-light denoisers remove noise but produce inaccurate colors.
- The ground truth of SIDD, a standard benchmark in denoising, contains an unintended color shift.
🧵
Sven Elflein retweeted
🚀 Pixel Diffusion Decoder (PiD) v1.5 is out
🎨 No color-shifting problem, better 4K visual quality
📦 Undistilled checkpoint and full training code
🧩 Support FLUX, FLUX2, Qwen-Image, Z-Image, ...
Feel free to use it, reproduce it, and build on top of it
💻 Code:
github.com/nv-tlabs/PiD
🔗 Demo & comparison:
research.nvidia.com/labs/sil…
Sven Elflein retweeted
Excited to share several recent works at #ICML2026! 🇰🇷 Happy to meet & chat about world models, video generation & data centric multimodal ML! DMs are open
Motion Attribution for Video Generation will be presented as an oral + poster:
- Tue Jul 7, 10:30–10:45 AM KST📍Auditorium
- Tue Jul 7, 2:00–3:45 PM KST📍Hall A #2807
research.nvidia.com/labs/sil…
Also presenting two workshop works:
Addressable Memory for Video World Models at the From Frames to Stories (F2S) workshop
- Fri Jul 10, 14:10–14:25 KST📍Room 402
research.nvidia.com/labs/sil…
Variance Reduction for Expectations with Diffusion Teachers at the SPIGM workshop, led by amazing co-author Jesse Bettencourt @jessebett
- Fri Jul 10, 11:40 AM–1:30 PM KST📍Hall D1
research.nvidia.com/labs/sil…
research.nvidia.com/labs/sil…
New #NVIDIA Paper
We introduce Motive, a motion-centric, gradient-based data attribution method that traces which training videos help or hurt video generation.
By isolating temporal dynamics from static appearance, Motive identifies which training videos shape motion in video generation.
🔗 research.nvidia.com/labs/sil…
1/10
If you are interested in TTT (and its limitations) at all, you should talk to Junchen!
HALL A #2910
Thursday 10:30 AM – 12:15 PM
I’ll be at #ICML2026 next week presenting this work.
icml.cc/virtual/2026/poster/…
See you in Seoul! 🇰🇷
Sven Elflein retweeted
🎉 Excited to introduce TRON, a relighting framework for 3D captures.
💡TRON pairs a neural renderer with 3D Gaussian reconstructions, achieving realistic quality, with 3D, material, & lighting control at interactive frame rates.
arxiv.org/abs/2606.11314
research.nvidia.com/labs/sil…
Sven Elflein retweeted
Conditional diffusion/flow models often produce outputs inconsistent with the very signal conditioning them. The error is easily measurable, yet models are never trained to act on it.
In FlowBender (now on arXiv), we train the model to correct its own errors. 🧵
We are presenting today in the afternoon poster session (15:30-17:30) at Poster No. 28!
#CVPR2026
Sven Elflein retweeted
The ArtiFixer code and model weights are now released! Links to both on our project page: research.nvidia.com/labs/sil…
Per-scene methods like 3DGS look great on captured views but collapse off-trajectory. We repair those artifacts and beat prior SOTA by 1–3 dB PSNR. 🧵(1/6)