If you will be in Portland on Friday, November 20th, 2026, I'm giving a talk at 11:30am at PSU, in Fariborz Maseeh Hall, room 462 about a Lean faithfulness review tool for mathematicians.
Free Pizza 🍕 and Coffee ☕ too!
I'm going to be in Seattle tonight and for the next few days, anyone interested in math, formal methods and AI, DM me
2
5
398
AI is rapidly advancing and mogging humans at advanced math.
Rocketships are returning from orbit and landing in one piece.
As amazing as this technology is, we should avoid turning it into an idol. It should serve us, while we serve God. Reject transhumanism.
Today, the promises of transhumanism and posthumanism are multiplying. These are ideological currents according to which, sooner or later, technology will eliminate all disease and pain, as well as aging and even death itself. Precisely because scientific progress is a supreme achievement of human intelligence, it must not be confused with a privilege reserved for millionaires seeking immortality, while entire peoples suffer from poverty and oppression. In the face of this mirage, which promotes the illusion of human beings perfected by cybernetics and biotechnology, Christian anthropology reminds us that a limitation is not a manufacturing defect, but rather the very condition of human grandeur. #ApostolicJourney #Peru #MagnificaHumanitas vatican.va/content/leo-xiv/e…
1
127
This was a theme in @mathematics_inc's recent post math.inc/position
You need to be working with the abstractions and engaging more with mathematical theories
Proving binaries one at a time won't scale, even with AI writing the proofs.
Prove the language, compiler and runtime once. Then what each program proves at source level carries to the binary.
That is where I expect verification investment to go next.
1
14
952
Tobi Lehman retweeted
Our shift-symmetry paper is out on arXiv 🥳
How many patterns on a wrap-around grid look exactly the same after a shift? We give the exact count - in every dimension, for any grid size and number of colors.
With Prof. John Caughman and @tlehman #Combinatorics #CellularAutomata #FormalVerification #Lean4
Tobi Lehman retweeted
Replying to @arpit_bhayani
Exactly the gap - Lean proves your spec, not that it's the spec you meant. Navier-Stokes-gate, anyone? 🙃
For our new paper we stated every result in Lean first, then a co-author hand-audited that the 27 main results say what the paper says 👇
nitter.cf/0xbnd/status/212125476…
Our shift-symmetry paper is out on arXiv 🥳
How many patterns on a wrap-around grid look exactly the same after a shift? We give the exact count - in every dimension, for any grid size and number of colors.
With Prof. John Caughman and @tlehman #Combinatorics #CellularAutomata #FormalVerification #Lean4
Tobi Lehman retweeted
Paper 👉: arxiv.org/abs/2611.15751
Repo with scripts, Lean proofs and the audit viewer: github.com/peterbanda/d-dim-…
Story behind it: medium.com/@0xbnd/when-does-…
This paper with the brilliant @0xbnd and Dr. Caughman was a big effort, I'm happy to share the preprint now that it's live: arxiv.org/abs/2611.15751
It uses group theory to analyze how the size of a set of undesirable symmetric configurations scales with n and d (dimension)
Tobi Lehman retweeted
Replying to @tlehman
two algebraically equivalent algorithms can have very different memory costs, so representation still matters in production
Learning Abstract Algebra makes you realize there are only like a handful of algorithms.
If you remove different representations, you find out that many algorithms are either the same, or almost the same.
Shortest path is semiring based computation.
More here: sciencedirect.com/science/ar…
I realized this when I was writing my book on OR algorighms. There are hundreds of algos I went through. Sometimes, you could find 40-50 algorithms, that are different only in one parameter, or one function in the function composition.
---
I am 100% sure, there are abstract algorithms in space search + reduction, that could simplify what already exist and provide dozens of real world useful algos for thousands of various usecases.
It goes deeper, check out Physics, Topology, Logic and Computation: A Rosetta Stone arxiv.org/abs/0903.0340
We are reaching levels of "why not?" previously thought impossible!
I think I took my theorem proving classes too seriously. Thanks to @145k4 I was able to run (Lean)DOOM, tick by tick. We exit Hangar without firing a shot. A fireball is still in flight.
We proved complete runs of Hangar and Nuclear Plant in our version of (Lean)DOOM.
Each run is completely verified in Lean and the runs of the players can be certified too. So can you prove to me that you can finish each level? The game is fully playable and we can build live the proofs from the game too. So much fun!
7
298























