Lean Verified Transformers
Lean Verified Transformers is Lean Verified Tokenizers ( ). It is ranked #776 on the AI Radar, in AI research, first seen 51 days ago and shared in 4 posts (65.3k views).
What Lean Verified Transformers says about itself
Rational vectors, sequence models, and verified parallelism in Lean.
Lean Verified Transformers — Lean Verified Transformers Skip to content Lean Verified Transformers Lean Verified Transformers This post explores writing formally verified ML code in Lean. Since the cost of proofs is declining rapidly and the amount of code generated is skyrocketing, the value of verified code seems likely to climb. While understanding proofs remains challenging, collaborating with AI to get proofs of easy-to-understand properties seems like a natural middle ground. The goal of this post is to verify foundational properties of Transformers. These are critical properties that are used for parallelization and optimization,…
What people said about Lean Verified Transformers on X
Deep Learning Speedrun - (https://srush.github.io/prereqs-of-deep-learning/) What is the minimal amount of Math you need to train an MNIST model?
— @srush_nlp, 8 days ago · 489 likes · see the post
Lean Verified Tokenizers (https://srush.github.io/verified-tokenizer/) Training-inference mismatch kills RL runs, but yet we build messages in Jinja? Prototypes a Lean-verified Rust agent pipeline.
— @srush_nlp, 9 days ago · 390 likes · see the post
Love seeing ideas from Miles' Token-In-Token-Out work inspire new research! Instead of just testing that training sees the exact tokens the model generated, Sasha takes it a step further, formally verifying key properties of a Rust tokenizer in Lean. That's the kind of rigor agentic RL needs. Fascinating work,…
— @radixark, 9 days ago · 31 likes · see the post
Alternatives to Lean Verified Transformers
- academy.dair.ai — Chat with Paper
- alphaXiv — Now on @askalphaxiv
- flashprove — Further progress in provable inference will likely be an important part of improving trust in AI deployment. To that…
- moe-blogpost — A hands-on guide to training Mixture-of-Experts models
- liberai.com — Technical blog
- Science's Last Exam — Can AI do research, end to end? Science's Last Exam defines and measures whether an AI agent can carry a research goal…
Lean Verified Transformers in numbers
- Rank on the AI Radar: #776 of 10402
- Shared in 4 posts by 2 accounts: @srush_nlp, @radixark
- 65.3k views on those posts
- First seen 51 days ago, last shared 8 days ago
- Pricing seen by AGK Intelligence: open source
- Market: AI research
FAQ
What is Lean Verified Transformers?
Lean Verified Tokenizers ( ) It was first shared on X 51 days ago and is ranked #776 on the AI Radar.
Is Lean Verified Transformers free?
It is open source.
Who shared Lean Verified Transformers?
2 accounts on X, including @srush_nlp, @radixark, in 4 posts totalling 65.3k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.6k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 15:10 UTC. Full method.