"Science needs time to think. Science needs time to read, and time to fail. Science does not always know what it might be at right now." - quoted by https://t.co/RTRx6yHZCH
#ECCV2026
Thrilled to share that Tchaikovsky Reimagined, my first project as an individual artist and a leading experimental theater project (2026), has been accepted into the ECCV Art Gallery.
Many thanks to the ECCV Art Gallery Committee for this wonderful opportunity!!! 💗
Huge thanks to our amazing authors who made this work possible! Throughout one year, our survey “Simulating the Real World: A Unified Survey of Multimodal Generative Models” has been accepted by IEEE TPAMI 2026!🎉
Arxiv: https://t.co/mLaad3xIwx
Repo: https://t.co/S6u6n02OoR
I decided to write down some of my thoughts in blog posts. In this post: https://t.co/EjCdoPI14u, I share my thoughts on world model, purely in the context of using simulation data to train a dynamic model of an articulated rigid body system, e.g., a humanoid robot.
Excited to share our new work on perceptual ambiguity! 👀💗
We study how humans and machine learning models perceive ambiguous images by generating a controllable continuum, revealing systematic differences in how ambiguity is resolved.
📄 arXiv: https://t.co/CHlThUBbCu
🚨 We're very happy to introduce TRIBE v2: a foundation model of the brain's responses to sight, sound & language.
📄 Paper: https://t.co/uHwgOvTrRD
▶️ Demo: https://t.co/9ZX6XcOXSM
💻 Code: https://t.co/PCc2yKyh1D
🤗 Model: https://t.co/GiTKzsHUhY
1/N
🎨🤖Given only a static image of an oil painting by an expert artist, can a robot infer the corresponding control actions, such as trajectory, orientation, and applied force, to accurately reproduce the painting?
🖌️Introducing IMPASTO: a robotic oil-painting system that integrates learned pixel dynamics with model-based planning to reproduce paintings from target images.
This project is led by @yingke_wang18 (https://t.co/zbNCQSwp8k) – a roboticist and an artist.
Project website: https://t.co/d8CAD83hS3
Input > Process > Output.
That’s the story we’ve been told about the hidden lives of cells.
As biologists, we’ve grown familiar with measuring “output” - the final few snapshots of what we know is a more nuanced story. From there, we fill in the blanks and imagine everything that might have happened during the “process” phase.
What if we could record a molecular slideshow of the inner workings of cells and play it back to reveal the story as it actually unfolded?
Exciting new @Nature study shows that’s what GEMINI does.
GEMINI is a protein assembly inside the cell that grows predictably over time, capturing cellular events as fluorescent patterns that are encoded like tree rings with a traceable chronology.
Just as we can recount a history of a region’s climate from the rings of trees, GEMINI allows us to record molecular stories within cells.
Full read here: https://t.co/XReJPnAM0F
Our survey v3 is LIVE on arXiv! 🚀
A unified framework connecting 2D, video, 3D, and 4D generation for real-world simulation. Now with a glossary to make key concepts easier to follow. ✨
GitHub: https://t.co/KT0DZkn0gO
arXiv: https://t.co/Zvx3V5OWUq
#GenerativeAI#survey
1/ Today with my colleagues @PolymathicAI, I'm excited to release our latest project, Walrus, a cross-domain foundation model for physical dynamics, into the world.
https://t.co/ihv1MZGQM3
Paper: https://t.co/d6ah9LO4ud
Git: https://t.co/s3p8qGhZQR
HF: https://t.co/RufaBD9eJk