evals evals evals

San Francisco, CA
Joined October 2018
exyzoteric retweeted
I feel there’s a lot of different sorts of daily motivations for me - is it delightful, is it fast, fix the fire, learn something new, cut the slop, get an efficiency win But I think when I look back and think about what I’ve learned, done, felt, that matters so much more
2
2
93
7,111
just in awe at how well-constructed the world is as an env. simulation or not.
1
28
1/ We've released Infinigen 2.0! Currently in preview. It creates indoor 3D scene files in 1min CPU time, and includes new and better materials --- all still fully procedural. Our new 2.0 design is highly efficient and allows easy control and recombination via Python APIs.
13
43
4
358
32,918
just got a cappuccino and saw them use instant coffee with a frother. paid 4.5 for it.
1
1
32
don’t ask me why i didn’t complain because i don’t want to tell you that i am quite enjoying the taste
26
exyzoteric retweeted
new post on harness engineering for AI self-improvement: lilianweng.github.io/posts/2… It is hard to forecast how much the future of RSI will rely on harnesses. Likely harness engineering will evolve in the direction of self-improvement and enable auto-research, and, in turn, smarter models keeps harnesses simple. Even when many harness improvement get eventually internalized into core model, the need to specify goals and context will not disappear.
137
795
135
5,468
916,410
follow enough advice and you arrive somewhere you never wanted.
19
roofline plot on Apple M2 for different tensor operations. (built with Fable 5)
72
exyzoteric retweeted
We've had a busy August!
August at Google DeepMind was like 🧞‍♂️ 🖼️ 🍌 🚀 🔍 🤏🏻 - Nano Banana (Gemini 2.5 Flash Image) - Gemini Embedding - Veo 3 Fast - Genie 3 - Imagen 4 Fast - Gemma 3 270M - Perch 2 - Kaggle Game Arena - Gemini API Url Context - AI Studio Builder (UI Rework, Prompt Suggestions, GitHub integration …) - AI studio UI (Model Picker, Chat, Scrolling…) and much more!
31
35
1
787
80,944
This is a super exciting opportunity! Having worked with Utkarsh, I can confirm he’s an amazing mentor and you absolutely do not want to miss out on collaborating with him 🚀
📢Excited to share that I’ve joined @MBZUAI as an Assistant Professor of Computer Vision this fall! If you’re interested in CV4Science: building the next generation of foundation models & discovery tools for science, consider applying to MBZUAI. I’ll be recruiting PhD students!
1
2
277
📢Excited to share that I’ve joined @MBZUAI as an Assistant Professor of Computer Vision this fall! If you’re interested in CV4Science: building the next generation of foundation models & discovery tools for science, consider applying to MBZUAI. I’ll be recruiting PhD students!
16
14
5
169
15,878
Have you wondered "How Video Meetings Change Your Expression?" We will be presenting our work at #ECCV2024 @eccvconf tomorrow Oct 1 @ 10:30AM (poster #195). Come say hi! facet.cs.columbia.edu Huge thanks to my amazing collaborators @utkarshmall13 @PurvaTendulkar @cvondrick
1
5
911
Excited to share our new paper on large-angle monocular dynamic novel view synthesis! Given a single RGB video, we propose a method that can imagine what that scene would look like from any other viewpoint. Website: gcd.cs.columbia.edu/ Paper: arxiv.org/abs/2405.14868 🧵(1/5)
5
30
4
133
33,599
🛰️📷 Need a Vision-Language Model (VLM) for satellite images but cannot collect billions of text-satellite pairs for training? Fret not! We introduce GRAFT, a VLM that can do better than CLIP on satellite imagery but is trained without textual annotation ✏️! 🧵 (1/7)
2
32
3
116
18,773