Formal Mind. (@FormalMindAI) header
Formal Mind. (@FormalMindAI) profile picture

Verifying Ethereum smart contracts in Lean.

USA
Joined April 2025
Photo by Formal Mind. (@FormalMindAI)Photo by Formal Mind. (@FormalMindAI)
Formal Mind. retweeted
Exponential Time Hypothesis (ETH), a 27-year-old conjecture in complexity theory, just fell. On Oct 16, OpenAI and Anthropic models independently solved 3-SAT with the same 2^(n/√log n) bound. Not P = NP, but faster than anyone thought possible. We live in extraordinary times.
90
79
23
968
58,568
Formal Mind. retweeted
Ryan Williams (@rrwilliams) is a professor at MIT and the winner of the Gödel Prize in theoretical computer science. I interviewed him all about his work starting by asking him a popular Leetcode question (3 SUM). In this episode: • Solving Leetcode faster than popular "optimal" solutions • SAT problems and solvers • Hot takes on famous open questions • How to pick good research direction Where to watch: • YouTube - youtu.be/AaK1SL2i_4Y • Spotify - open.spotify.com/episode/0JH… • Apple Podcasts - podcasts.apple.com/us/podcas… • Transcript - developing.dev/p/mit-complex… 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/ Chapters: 00:00 - Intro 00:41 - Asking him a popular Leetcode question 03:54 - Doing better than the popular optimal solution 08:26 - Fine grained complexity 17:00 - A severe strengthening of P vs NP 24:38 - SAT problems and solvers 34:51 - Hot takes on famous open questions 46:57 - Simulating space with time 01:01:02 - Why he solves hard problems 01:02:35 - How to pick good research direction 01:07:14 - Technical book recommendations 01:08:31 - Advice for his younger self 01:11:56 - Outro
6
60
5
684
151,051
An internal model at OpenAI has produced a claimed refutation of ETH! Russell Impagliazzo, Mohan Paturi, and I have been asked to help organize an open and independent examination of this claim. 1/4
303
770
544
7,398
3,771,178
Formal Mind. retweeted
Software Verification in Lean (SVIL) comes to Stanford University on Saturday, November 7. Co-organized by Beneficial AI Foundation and Lean FRO, it is a day of public lectures on advancing Lean’s software verification ecosystem toward production use, with a focus on end-to-end verification of security-critical systems. Talks will cover Rust verification with the Aeneas framework, low-level software verification, and formally verified OS kernels. Speakers include Max Tegmark, Shaowei Lin, Leonardo de Moura, Jonathan Protzenko, Andres Erbsen, and Mike Dodds. In-person space is limited. Non-local and waitlisted registrants will receive a livestream link, and recordings will be available afterward. Program and registration: beneficial-ai-foundation.git…
4
60
2
364
33,527
Formal Mind. retweeted
I trained models across MacBooks using Apple's AirDrop protocol. grove is a distributed training library for Apple Silicon. Devices discover each other over AWDL, a direct radio link. If there's a shared WiFi network it upgrades to that for speed, otherwise everything goes over the direct link. No router, no cloud, no setup. grove start <script> -n 4 grove join
156
252
86
3,266
669,869
Cray Distinguished Colloquium at UMN, next Monday. An AI converted zlib to Lean and proved it correct. 10 AI agents built a verified DSL in a weekend. Three IMO teams, no competing platform. The slides are written in Verso: checked by Lean. leodemoura.github.io/static/…
8
38
2
130
11,080
Formal Mind. retweeted
Geoffrey Hinton says people still think LLMs are different from us, but they’re not LLMs don’t just generate words; they reflect how we process meaning: "they’re our best model of how language works" old linguistic theories failed to explain language; neural nets using feature vectors finally do
107
242
53
1,480
193,411
🚨 NotebookLM Dethroned?! 🚨 Meet vui: The new open-source dialogue generation model. 💪100M Params, 40k hours audio! 🎙️ Multi-speaker audio 😂 Non-speech sounds (like [laughs]!) 📜 MIT License Is this the future of audio? Watch the demo and decide for yourself!
9
36
1
201
16,406
Formal Mind. retweeted
omg.. this cant be real China’s 4DV AI just dropped 4D Gaussian Splatting, you can turn 2D video into 4D with sound.. imagine.. we will be able to change camera angle, zoom in/out while watching movies 5 examples:
704
3,681
862
35,008
3,617,436
Formal Mind. retweeted
LLMs: 💬 Take text input 📝 Provide text output AI Agents: 🧠 Use LLMs for reasoning and planning 🦾 Act in their environment 🔄 Receive and use feedback 📚 Access memory or knowledge base 🛠️ Have tools for execution 🎲 LLM can be multi-modal, generating text, images, videos etc
1
1
1
101
Formal Mind. retweeted
I'm writing my notes on Lecture 5 from the LLM Mooc by UC Berkeley. Here's my favorite quote from Charles Sutton. @RandomlyWalking "Your career is long, technologies change. Think across disciplines."
1
1
59
Formal Mind. retweeted
how i set up 12+ claude 4 code agents to work in parallel mostly autonomously for several hours without conflicts or slop code
79
164
16
2,036
317,705
Formal Mind. retweeted
Over the weekend, I built an AI agent that can review my code, suggest changes, and create new repos. 👀 I'm thinking about how to extend this functionality.
1
56
Formal Mind. retweeted
Meet my GitHub Formal Mind agents, designed to collaborate with you. - Install the GitHub App - Create a repository - Push your code - Open pull requests - Use "@ agent verify" 🕵️‍♂️ See the agent review and verify your code! 💬
1
1
82
Formal Mind. retweeted
Patrick's insights are spot on: For AI and crypto to achieve mass adoption, we must establish clear frameworks for agent liability. Ensuring that autonomous agents are accountable is essential. It's imperative that we maintain control over AI, rather than being controlled by it.
1
2
1
4
710
Formal Mind. retweeted
Wondering what happened with the Cetus hack? Join @MitchellAmador and Magnus partner @dedaub Monday for a line by line analysis with @neville_grech and how you can build safely in the most hostile environments. 🗓️ Date: June 2 ⏰ Time: 6pm London teapaw.org/i/spaces/1RDGlzdjzlMxL

5
6
5
27
11,080
Prompt Theory (Made with Veo 3) What if AI-generated characters refused to believe they were AI-generated?
971
3,765
2,312
23,967
6,043,899
Formal Mind. retweeted
Oh, we just learned that at LlamaCon, Meta announced llama-prompt-ops, a Python package that "transforms prompts that work well with other LLMs into prompts optimized for Llama models". They demonstrated neat gains across tasks. Built in DSPy and via our MIPROv2 optimizer.
HUGE performance gains with the new llama-prompt-ops tool to optimize prompts from other models for llama this is crazy
5
48
3
234
48,449
If you’re picking your AI models based on public generalist leaderboards, you’re doing it wrong. In my opinion, evaluation & model picking is at least 30% of the work of any great AI builder and it’s a mix of public generalist and specialized leaderboards, social signals (likes on HF, community tabs,…) and private evaluation on your data. And the results depend a lot on your use-case/domain/language/tasks/constraints so don’t be a social media sheep!
20
46
11
320
45,245
Formal Mind. retweeted
🚀 ZeroGPU v2 update We just switch to Nvidia H200 last week It means that @huggingface Spaces are now equipped with: - 🧠 70GB vram - ⚡ 2.5x more flops 🔓This will hopefully unlock unseen use cases 💰 It also makes Pro plan a seriously cheap CUDA compute option
14
43
7
197
81,497