Chaitanya K. Joshi
@chaitjo
Followers
9K
Following
19K
Media
248
Statuses
3K
PhD student @Cambridge_CL. Research Intern @AIatMeta and @PrescientDesign. Interested in Deep Learning for biomolecule design.
Cambridge, UK
Joined July 2010
Introducing All-atom Diffusion Transformers . — towards Foundation Models for generative chemistry, from my internship with the FAIR Chemistry team @OpenCatalyst @AIatMeta . There are a couple ML ideas which I think are new and exciting in here 👇
4
105
527
RT @teortaxesTex: a good argument for why Transformers or their functional equivalents are likely sufficient as a substrate to learn whatev….
0
2
0
RT @MirunaCretu2: Check out our latest work on unconditional molecule generation! Some cool 👀 insights below! 🧵.
0
1
0
Congratulations to Carlos for a 🔥paper! . For SOTA molecular generation with 3D Diffusion Transformers - they show you don’t even need latent diffusion. Tabasco is perhaps as simple and elegant as it gets!.
🌶️spicy paper alert🌶️. We present TABASCO, a new model for small molecule generation that achieves state-of-the-art PoseBusters validity while also being ~10x faster. This is all achieved despite.✖️no equivariance.✖️no self-conditioning.✖️no bond modelling. what we found🧵(1/n)
0
1
20
RT @msikic: Our RiNALMo RNA language model has been published in Nat Comm Great work by @RJPenic and @TinVlasic,….
0
14
0
RT @ZhongingAlong: CryoDRGN-AI ❄️🐉🤖 is now published in @naturemethods!!! So excited to see this out and a huge congrats to @axlevy0 and te….
0
31
0
RT @srikosuri: It’s kind of interesting, but in some ways AI is making CS much more like biology. We can observe and tweak what it does, bu….
0
53
0
This!!!.
This is absolutely amazing! The most interesting part of this breakthrough for me is how many people were saying “we don’t have enough antibody data” - well, the @chaidiscovery team have shown that we might have enough to solve the problem.
0
0
4
RT @ArdaGoreci: This is absolutely amazing! The most interesting part of this breakthrough for me is how many people were saying “we don’t….
0
1
0
RT @zackulissi: uma-s-1.1 is now live in the UMA educational demo! The demo featured in two tutorials recently - o….
0
7
0
Actually thought about this (having done such a stint in a wet lab now). I think the SF type will still be super duper optimistic and excited!. But it will contextualize and physically embed all the AI work in something more meaningful.
We desperately need to take all the people in SF who talk about accelerating biology but have only ever done math or CS and have them do a six month rotation in a wet lab.
1
0
11
RT @BhaktaVee: @chaitjo my take is that transformers exemplify the bitter lesson of architectures that leverage hardware-driven scaling oft….
0
1
0
RT @thegautamkamath: ICML's election for their board of directors has begun. I've thrown my hat in the ring. Please consider voting for Gau….
0
22
0
RT @itsbautistam: @chaitjo Nice! My realization back then was that in transformers it's much easier to encode the geometric structure of th….
0
1
0