
Look forward to continuing to work with @tedlasai during his postdoc — great things coming soon! tedlasai.github.io/
Good to time to announce that I started my postdoc with the wonderful @kyroskutulakos and @DaveLindell!!

Assistant Professor @UofTCompSci Faculty Affiliate @VectorInst

Look forward to continuing to work with @tedlasai during his postdoc — great things coming soon! tedlasai.github.io/
Good to time to announce that I started my postdoc with the wonderful @kyroskutulakos and @DaveLindell!!


New Gaussian splatting representation based on polyline primitives! The primitives are optimized directly from images and can form long, connected chains. Perfect for reconstructing strand structures (spaghetti, hair, etc.) and it matches 3DGS performance on general scenes.
📢📢📢 Introducing Spaghetti Splatting 3DGS is a convolution of a Dirac with a Gaussian Kernel. We provide a strict generalization of 3DGS, extending its Dirac mass to lines and curvilinear structures project page: spaghetti-splatting.github.i… arXiv: arxiv.org/abs/2611.12716 🔊 on!

Pretty cool that this works with only 15 min of data! You can take a robot policy that doesn't do mobility and turn it into one that can move around (mobility + manipulation) with a very limited robot data.
We taught a robot how to move around in only 15 minutes. That’s all the mobility data our Direct Video-Action-powered robot needed to accomplish a task that requires mobility and manipulation.

📢🚀Happy to share our recent work. SpecKCast: Spectral Kalman Forecasting for Diffusion Acceleration 🌐Webpage: rajabi2001.github.io/speckca… 📄Paper: arxiv.org/abs/2610.26837 A huge thanks to @karaimer, @DaveLindell, @Babak_Taati, and Michael S. Brown. @Samsung_RA @VectorInst @UofTCompSci

It's amazing how much you can scale up the output resolution and speed of image generation models without any re-training. The trick here is to leverage the more local behavior of attention in later diffusion steps to accelerate attention within diffusion transformers.
📢🚀 Excited to share AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers Training-free, up to 6× the native resolution, up to 3.2× faster. 📄Paper: arxiv.org/abs/2610.17227 🌐Webpage: rajabi2001.github.io/area 🤗HF: huggingface.co/papers/2610.1…

Robots making rapid progress in "learning to learn"
What can a robot do without vision? We gave our robot a sense of touch and showed it can recognize and retrieve objects with its cameras blindfolded. Introducing Rhoda’s new tactile sensor—it feels pressure, texture, vibration, slip, and temperature, just like your fingertips.

(1/n) Watch our robot search a box and find textured balls, entirely without vision! Over the past few months, it’s been incredibly fun to work on bringing touch into our Direct Video-Action (DVA) model. Here’s how our video model is “seeing” through touch 🧵
What can a robot do without vision? We gave our robot a sense of touch and showed it can recognize and retrieve objects with its cameras blindfolded. Introducing Rhoda’s new tactile sensor—it feels pressure, texture, vibration, slip, and temperature, just like your fingertips.

6/ Also check out Roland Johansson's work: without tactile feedback humans have a lot of trouble with even simple tasks like picking up a match. It really seems like robots ought to have a sense of touch! youtube.com/watch?v=0LfJ3M3K…

5/ By combining the tactile sensor with our Direct Video-Action model, we created a robot that searches for, identifies, and grasps objects (even without vision). Full writeup: rhoda.ai/research/giving-rob…

4/ Humans use vibrations to help reveal surface texture and whether an object is slipping. Our sensor captures vibrations at around 500 Hz. Interestingly, by listening to the tactile signals as audio you can clearly tell the difference between different surface textures.

3/ The sensor detects light contact, measures normal and shear forces, and tracks rapid changes while handling substantial loads. That wide dynamic range lets it register both a feather’s touch and forces exceeding 40 N.

2/ Existing tactile sensors come with difficult tradeoffs: sensitivity, dynamic range, shear sensing, response speed, size, reliability, and cost. We built our own sensor to bring these capabilities together in a compact, low-cost package.

We gave our robots a sense of touch! Excited to finally share this work with @AndrewEJXie, @leooeld, and the @rhoda_ai team! Some technical details on the tactile sensor and robot policy: 🧵
What can a robot do without vision? We gave our robot a sense of touch and showed it can recognize and retrieve objects with its cameras blindfolded. Introducing Rhoda’s new tactile sensor—it feels pressure, texture, vibration, slip, and temperature, just like your fingertips.

There is a lot to appreciate in these demos—recovery from errors, the length and complexity of the demonstrations, and targeting actual, real-world tasks.
Continuing our RhodaAtWork series with another look at Direct Video-Action (DVA) in action. This time, a single model handles multiple tasks across two home appliances: refrigerator EPS foam installation and washing machine drain hose routing.

Great to see more robot demonstrations from @RhodaAI — recovery behavior, challenging customer use cases, and reliable autonomous operation over extended periods of time.
More industries. More real customer work. Introducing RhodaAtWork, a series on the customer workflows our robots are learning to handle across manufacturing and logistics. First up: unpacking, scanning and stacking component reels for electronics manufacturing.