Viewing profile — tzumaoli

Trivia: Claude Shannon proposed the idea of predicting the next token (letter) using statistics/probabilities in the training data corpus in 1950: "Prediction and Entropy of Printe…

This really resonates. I am in the CS academia and I feel like I am losing "friends" who used to share the same values and social identity with me at an unprecedented speed. I wond…

It reminds me of this video: https://www.youtube.com/watch?v=FIFty-O4rOE TLDR: amazing things can happen, and people are actually nice in this kind of environment!

It could in theory. The model generates a depth image per frame, so each pixel becomes a small 3D point. It also assumes that the 3D scene is static. From this, you can then simply…

The same Cornell group has done more research in this area. Here is one for capturing the microgeometry of metals https://www.cs.cornell.edu/Projects/metalappearance/ Here's anothe…

It's interesting to see the trend of the attitude towards GenAI in Hacker News through out the years. This is totally vibe based and I don't have numbers to back it up, but back in…

When Fortran came out, I don't think a lot of people yelled at the assembly programmers and told them "learn Fortran or be replaced".

also "They will get better in no time"

Discussed in another thread https://news.ycombinator.com/item?id=41234415 As someone who has worked on diffusion model, it's a clear reject and not a very interesting architecture.…

I am a graphics researcher who has been following your blog posts on your GPU vector graphics engine. I've been benefitted by your posts and I think some of the vector graphics ren…

Check out our recent work on improving image vectorization using gradient descent: https://people.csail.mit.edu/tzumao/diffvg/

Also a self-plug: our differentiable renderer "redner" also has a bunch of Jupyter notebook tutorials: https://github.com/BachiLi/redner