Om Prakash retweeted
This note from Bao Chau Ngo (Fields Medal 2010, Clay Research Award) on Alpoge's work for Riemann is incredible. It's rare to see a mathematician of his caliber with such an open mind about AI.
Come on, @AP He deserves better than being called an "Anthropic worker." He is Dr. Levent Alpöge, a Princeton-trained mathematician who proved the Riemann Hypothesis, one of the most famous problems in mathematics. His story deserves proper recognition. @leanprover
7
399
Om Prakash retweeted
1/5 EBMs are usually taught as one model family among many. In our new tutorial with @du_yilun, we use them as a single lens on machine learning.
"An Energy-Based Perspective on Machine Learning" arxiv.org/abs/2611.14383
Om Prakash retweeted
The Lean FRO and Mathlib Initiative have released a joint statement on the Lean verification of the Riemann Hypothesis.
On November 20, Levent Alpöge released a proof of the Riemann Hypothesis as a Lean library of 17 million lines of code. Over the past day, the Lean and Mathlib communities worked together to verify it. The statement certifies the proof as correct and describes how it meets all necessary verification criteria.
Congratulations to Levent on this achievement, and thank you to everyone in the community who contributed to the verification.
Full statement: leanprover-community.github.…
Om Prakash retweeted
We don't yet have full confirmation from the paranoid-mode comparator test, but we are close to seeing a formal artifact generated in Lean that provides a proof of the famous Riemann Hypothesis.
This is such a historic moment. The code was generated by @__alpoge__ using Claude and possibly other software.
There is still some uncertainty, but we may be on the verge of a historic announcement. For the first time in history, we may have such a magnificent proof that we DON'T fully understand yet, but that is probably correct.
Lean experts are currently investigating the validity of the claimed formalised proof of the Riemann Hypothesis - follow along at
leanprover.zulipchat.com/#na…
Om Prakash retweeted
Today at Starkly Speaking: @AdrDobbel, @rneschneuing and @igashov will present 'A Unified 3D Generative Model for Synthesizable Structure-Based Drug Design' biorxiv.org/content/10.64898…
On Zoom 1pm ET / 7pm CET: hannes-stark.com/starkly-spe…
Om Prakash retweeted
To me, the promise of unified simulation is that we can spend less effort rebuilding tools and more effort exploring physical systems.
WorldParticle gives us a common architecture across six kinds of physics. Extending it to more systems, and learning across them with shared weights, is the direction I'm excited to keep pushing.
Huge thanks to my co-first authors Caoliwen Wang and Siyuan Chen, co-corresponding author Peter Yichen Chen @peterchencyc, Wojciech Matusik @wojmatusik, and all our collaborators. See you at SIGGRAPH Asia 2026! (7/7)
Paper: arxiv.org/abs/2605.15305
Project page: worldparticle.github.io/Worl…
#AI4Science #PhysicalAI #PhysicsSimulation #SIGGRAPH #SIGGRAPHAsia2026 #MITCSAIL
How did the 3D genome architecture evolve? 🧬 Our new review @AnnualReviews by @zocean636 & Evan Trop in the group explores conserved genome folding principles, lineage-specific architectures, and how comparative genomics and #AI may connect sequence evolution to genome folding and function. annualreviews.org/content/jo…
Network science has inspired the design of materials that spread out their deformation when a force is applied, have tunable stiffness and can control vibrational waves
go.nature.com/4iMUiyT
Om Prakash retweeted
Beautiful paper! A uniquely elegant example of how statistical physics can help us make sense of complex ecosystems. From fairy circles to ecological tipping points, seemingly disparate phenomena emerge from simple universal rules @sonia_kefi pnas.org/doi/abs/10.1073/pna…
Om Prakash retweeted
Strong potential gradients can localize individual particles, but interactions can still generate collective motion. An experiment with ultracold atoms in a tilted lattice now shows this collective motion by tracking mobile bound states of fractons.
nature.com/articles/s41567-0…
Om Prakash retweeted
Our latest work in collaboration with @barabasi is out in @Nature today.
We explore the benefits of length scale heterogeneity in metamaterials, blending mechanics with network science concepts!
📄Article: nature.com/articles/s41586-0…
🗞️MIT Mews: news.mit.edu/2026/engineers-…
What if we designed materials the way nature builds its networks?
Our @Nature paper introduces Scale-Rich Materials with remarkable properties: tunable stiffness, impact absorption and wave control.
PDF at barabasi.com/science/publica…
#NetworkScience #NetSci #NetworkMaterials
Om Prakash retweeted
In Nov 2016, my team and I published a paper on Neuro-Symbolic Program Synthesis. This was a time before transformers and LLMs, and the programs we generated with neural networks were tiny macros used to perform simple data edits (think autocomplete) in sheets.
Fast forward 10 years: we used LLMs to generate over 6 million lines of Lean mathematical proofs to formally verify key security properties of Signal's cryptographic protocol, all the way down to the machine-code level.
These advances have been transformational. The next 10 years are going to be important and I hope we all appreciate the responsibility that comes with leveraging these capabilities for the benefit of humanity.
Read more:
Nov 2016: arxiv.org/abs/1611.01855
Nov 2026: deepmind.google/documents/36…
Om Prakash retweeted
We are excited to share our first nanopore sequencing study, now published in @NatureComms!
We developed a deep learning-based method for accurate tRNA classification, absolute quantification, and single-molecule detection of RNA modifications.
nature.com/articles/s41467-0…
Om Prakash retweeted
Mathematician Terence Tao introduces readers to six core ideas that have shaped mathematics. Welcoming, wise, and “full of piercing insight” (@benorlin), SIX MATH ESSENTIALS is available today, wherever books are sold: quantabooks.org/terence-taos…
Om Prakash retweeted
Introducing e-means, an algorithm beating k-means in 10x fewer epochs.
With order of magnitude wall-clock speedups, e-means is released in FAISS and needs no hyperparameter tuning beyond k. It is robust to various scales and datasets. Still skeptical ? Just try e-means [1/10]🧵
Om Prakash retweeted
Venki Ramakrishnan (Molecular biologist, Nobel Prize): "The experimental part [of drug discovery] is always going to be the bottleneck."
Recent headlines about using AI to speed up drug discovery made me curious about what that looks like and where it is today. So I flew to the UK to interview Venki Ramakrishnan who is on the scientific advisory board of Demis Hassabis' Isomorphic Labs.
He has described himself as a "failed physicist" starting originally with a physics PhD who changed fields because he felt his work was incremental. After pivoting into molecular biology without a degree he discovered the field of work that eventually led to his Nobel prize.
So aside from drug discovery, I also asked him about how AI is impacting the field of biology, how biology compares to physics, and the politics behind the Nobel prize.
Chapters:
00:00 Intro
00:43 How drugs are designed
04:54 Why drug testing remains difficult
07:02 How AI will impact the field of biology
10:35 Switching from physics to biology
15:00 How to choose research problems that matter
23:19 Did lacking a credential hurt his career
26:50 Physics vs biology
33:21 Why physicists succeed in other fields
35:10 The politics of the Nobel Prize
38:37 Top scientific book recommendations
40:05 Advice for young people
41:43 Outro
Where to watch:
• YouTube - youtube.com/watch?v=-1q0zA21…
• Spotify - open.spotify.com/episode/2M0…
• Apple Podcasts - podcasts.apple.com/us/podcas…
• Transcript - developing.dev/p/e074-venki-…
Thank you to the sponsor of this episode for supporting my work:
• WorkOS: makes your app Enterprise Ready with easy to use APIs to add SSO, SCIM, RBAC, and more in just a few lines of code, check them out at workos.com/
Om Prakash retweeted
Excited to share our new paper @NaturePhysics! 🎉
We show how motor-driven flows remodel DNA into a viscoelastic network that feeds back to transform chaotic motion into synchronized, millimeter-scale oscillations!
Read more here
doi.org/10.1038/s41567-026-0…
Om Prakash retweeted
I built Recall for the Meta Global AI Developer Hackathon: a voice-first memory of your home, built first for blind and low-vision people.
Walk through a room once, then ask "where's my watch?" and be walked there by voice.
A blindfolded run, one take: youtu.be/vRAR3FHNSks





























