Rishabh

@rishabh16_

AI for structural biology @stanford @rdaslab • mech interp for BioML • prev @yale @genentech @cambridge_cl 🧬🛠

🇺🇸🇸🇬

rish-16.github.io

Replies Reposts Media Articles

Pinned
Rishabh
@rishabh16_
Mar 20, 2022
🚨 My Graph Deep Learning extended tutorial has a new home!!! It's finally up on my personal website: rish-16.github.io/posts/gnn-math/ Do RT, share, and tag your GDL buddies if you find it useful 🙌🏻🙌🏻
Rishabh
@rishabh16_
Dec 10, 2021
🚨NEW WEEKEND READING🚨 I'm publishing "Graph Neural Networks for Novice Math Fanatics" – a primer on the math behind GNNs using colourful drawings & diagrams rish16.notion.site/Graph-Neural-N… (I hope this becomes the definitive guide to GDL for those hoping to enter the field!!!)
10
Rishabh
@rishabh16_
3h
Rumor has it Anthropic is finishing up GTA6. Hope someone else doesn't scoop them on it ...
Rishabh
@rishabh16_
12h
Kinda related but this reminds me of @jaschasd ‘s ultra finegrained hparam tuning work: arxiv.org/abs/2402.06184
William Gilpin
@wgilpin0
14h
Our group discovered that reasoning models produce fractals when asked to solve hard problems. We can use nonlinear dynamics to probe the thinking processes of recurrent depth models on Sudoku, mathematics, and even ARC-AGI (1/N) arxiv.org/abs/2609.04963
00:00
Rishabh
@rishabh16_
Sep 7
This was awesome to read :D If anyone has more play recs, please send!!
2
Rishabh
@rishabh16_
Sep 4
Can attest to how versatile it is -- RoPE really is awesome! It was really fun seeing how powerful it can be when modeling 3D geometric objects (like molecules) in our Platonic Transformers architecture :D arxiv.org/abs/2510.03511
eigenron
@eigenron
Sep 4
RoPE is the most mathematically beautiful part of modern models. fight me.