Lecture slides for my "Introduction to #ComputerVision" and "#DeepLearning in Computer Vision" courses. 🆕 Gaussian Splatting 🆕 Flow Matching 🆕 Sparse Autoencoders The included videos do not contain voiceovers yet, planned for a future revision.

Be sure to check out the picturesque Starfield Library! Also located within the Coex complex.

Some call this AI slop, I call this hilarious 🤣

Being at #ICML2026, new to the community, and trying to meet people. Have fun @HThasarathan and make sure to network! Harry will be previewing our new work on cross-model circuit discovery at the Mech Interp Workshop.

Submitting to #NeurIPS2026 and then getting the reviews 😬

On my to ICML 🇰🇷. Will present our Linearizer paper on Tuesday, July 7 at 10:30 AM poster session. Come and say hi! I’m happy to meet and chat about research, so please feel free to reach out and we can have a coffee together.

At @icmlconf in Seoul all week. Happy to chat about neural geometry, Vision Interp, or joining @GoodfireAI, come say hi (DMs are open) !🤠

🚀 **What if 3D generation did not need a complicated cascade of models?** Most recent 3D asset generation pipelines still follow a multi-stage recipe: 🖼️ generate multi-view images → 🧱 reconstruct geometry → 🔁 convert representations → 🛠️ refine artifacts… and hope that errors do not accumulate along the way. In our new paper, **PixGS: Pixel-Space Diffusion for Direct 3D Gaussian Splat Generation**, accepted at **ECCV 2026** 🎉, we ask a simpler question: > Can we generate 3D Gaussian Splats directly — in a single stage — while preserving fine-grained geometry and appearance? Our answer is **PixGS** ✨, a pixel-space diffusion framework that directly denoises 3D Gaussian attributes instead of relying on lossy latent compression. This allows the model to regularize the generated asset at the **splat level** 🎯, improving both geometry and appearance while avoiding many decoding artifacts seen in latent/cascade pipelines. The key idea is to bring the strength of large 2D generative priors into 3D generation 🌍, but without forcing the 3D representation through a compressed latent bottleneck. PixGS uses richer supervision signals — including 🔹 RGB 🔹 depth 🔹 surface normals 🔹 high-frequency structural cues to better preserve object details and view consistency. The result: high-quality text/image-conditioned 3DGS generation with strong performance and fast inference — around **1 second on a single A100 GPU** ⚡. I’m especially excited about this direction because 3D Gaussian Splatting has already transformed how we render and represent scenes. PixGS suggests that the next bottleneck may not be rendering anymore, but how cleanly, directly, and efficiently we can **generate** 3D assets in the first place. 📄 Paper: arxiv.org/abs/2607.01803 Huge thanks to my co-author Cao Duy and everyone who supported this work! 🙌 #ECCV2026 #ComputerVision #GenerativeAI #3DGeneration #GaussianSplatting #DiffusionModels #AIResearch #Graphics #Robotics #QualcommAIResearch

Nostalgic sounds from the 80s: all airs could go on to this list

Unitree G1 tap dancing at concerts in China

Keeping “8min presentation” in your Claude slop slides is crazy work #acl2026

We'll be in Seoul for ICML presenting "Interpreting Physics in Video World Models" at the main conference and mech interp workshop this week! 🌟 Main Conference - Thursday, July 9, 10:30-12:15 pm KST (Hall A 3406) 🌟 Mech Interp Workshop - Friday, July 10, 2-3 pm KST (Board 120) Stop by if you'd like to discuss interpretability for world models, physical reasoning, and scientific simulation!

Absolutely insane. 65 submissions from one author. #acl2026

This almost 100 year old masterpiece is set in 2026

Silicon Valley pundits love to preach the 80-hour work week; or what is it in China now, the 9-9-6 schedule. They tell you that building the future of AI, robotics, or a venture-backed startup requires constant, grinding hustle. They’re wrong. For a weekend read, I'm staring at "Rest: Why you get more done when you work less" by @AlexSoojun8374 Alex Soojung-Kim Pang. I had bought it last year and it finally rose to the top of the queue. Yet, it seems to be giving words to practices I've long espoused. I'm a "work hard everyday" kind of person, but I'm also a rest, create, expand type of person. One thing really from the beginning of the book jumped out at me. In the foreword, @ariannahuff Arianna Huffington points to a concept that completely refutes the tech-bro narrative: "Layer Work and Rest." The premise is simple but profound: supercreative people alternate periods of intensive work with periods of deliberate rest. She writes, and I'm paraphrasing hre: When you intentionally step away from a grueling technical problem, you aren't quitting. You are actually encouraging your subconscious mind into taking over. By giving yourself time and space immediately after focused effort, you allow your brain to discover the elegant solutions that eluded your conscious grasp. And I think that is actually layered into different temporal hierarchies: a given day, a given week, a given season. In STEM, we often treat humans like silicon processors that just need to be overclocked. But building responsible, impactful technology requires depth and clarity, not just raw compute. Deliberate rest isn't laziness. It’s the ultimate cheat code for bringing real value.