Mathlib
PulseAugur coverage of Mathlib — every cluster mentioning Mathlib across labs, papers, and developer communities, ranked by signal.
- used by Lean 4 Programming Language 90%
- authored by Andrew Wiles 90%
- developed Gotit.pub 90%
- used by Lean Finder V2 90%
- developed by MathlibQR 90%
- affiliated with Lean 4 Programming Language 70%
- instance of Lean 4 Programming Language 70%
- developed by Lean 4 Programming Language 70%
- used by Gotit.pub 70%
- used by MathlibQR 70%
- developed by Lean Finder V2 70%
- used by CatalyzeX Code Finder for Papers 70%
12 day(s) with sentiment data
-
1,000 Anthropic Claude Agents Prove Fermat's Last Theorem Autonomously
Anthropic's Claude Agents, an early version of Claude Fable 5.2, successfully proved Fermat's Last Theorem by developing their own organizational structure. A team of 1,000 agents completed the complex mathematical proo…
-
Axiom formalizes 7,500-page Stacks Project in Lean, creating largest math library
Axiom has released a formalization of the entire Stacks Project, a foundational reference in algebraic geometry, into the Lean proof assistant. This extensive library, comprising 16 million lines of code, represents the…
-
New research quantifies imbalance in Latin rectangles
This paper explores the concept of balance in Latin rectangles, viewing them as points in a grid and measuring their imbalance using the L1 metric. Researchers developed a simulated annealing approach that improves upon…
-
New DAPS system enhances LLM theorem proving by leveraging dependency signals
Researchers have developed DAPS, a new dependency-aware premise selection system designed to improve Large Language Model (LLM) theorem proving in the Lean 4 programming language. Unlike previous methods that rely on se…
-
Vampire-Hammer tool integrates automated theorem prover with Lean
Researchers have developed Vampire-Hammer, a new tool designed to assist the Lean interactive theorem prover. This tool leverages the Vampire automated theorem prover to solve mathematical goals and reconstruct proofs. …
-
Vampire-Hammer tool enhances Lean theorem prover with automated proof capabilities
A new tool called Vampire-Hammer has been developed to assist the Lean interactive theorem prover. This tool integrates with the Lean-auto framework to send goals and premises to the Vampire automated theorem prover. Va…
-
New tool translates model counting certificates into Lean theorems
Researchers have developed a tool called count-catcher that translates certificates from the exact model counter d4 into Lean theorems. These theorems assert that a given CNF formula, defined within Lean, has a specific…
-
AI tool KAPPA automates proof-checking in statistics papers
Researchers have developed KAPPA, an open-source software system designed to automatically check proofs in statistical theory papers. KAPPA utilizes language model agents, specifically those powered by Claude Code, to t…
-
Lean Finder V2 enhances semantic search for Mathlib
Lean Finder V2 is a new semantic search tool designed to improve query disambiguation for the Mathlib mathematical library. It features an "Arena mode" where users can compare Lean Finder side-by-side with other models,…
-
New method aligns informal math statements with formal libraries
Researchers have developed a method to align informal mathematical statements with their formal counterparts in libraries like Lean's mathlib. This work constructs a dataset linking informal statements from the Stacks P…
-
Lean Finder V2 enhances Mathlib semantic search with query disambiguation
Researchers have developed Lean Finder V2, an advanced semantic search framework designed for Mathlib, a large mathematical library. This new system employs a combination of model-level adversarial training and input-le…
-
PaperFormalizer system aids authors in auto-formalizing research papers with Lean
Researchers have developed PaperFormalizer, a system that uses the Lean proof assistant to help authors auto-formalize research papers. This agentic pipeline takes LaTeX source files and converts them into a dependency …
-
Lean Finder V2 advances Mathlib search; Anthropic SDK updates released
Researchers have developed Lean Finder V2, an advanced semantic search framework for Mathlib that significantly improves query disambiguation through adversarial training and query rewriting. This new system achieves an…
-
LeanGraphSearch enhances formal theorem proving with graph-based lemma retrieval
Researchers have developed LeanGraphSearch, a novel graph-based retrieval method designed to enhance formal theorem proving in the Lean 4 programming language. This system constructs a dependency graph of mathematical d…
-
New dataset formalizes theorems from Annals of Mathematics for AI benchmarking
Researchers have introduced AnnalsChallenge, a new dataset comprising 50 formalized theorem statements from papers published in the Annals of Mathematics since 2020. These statements were meticulously hand-written in Le…
-
New Lean 4 theorem prover NanoProof released with open data and tooling · 2 sources tracked
Researchers have introduced NanoProof, an open-source automated theorem prover for Lean 4 that achieves 50.8% pass@16 on the MiniF2F-Test benchmark. This system is notable for releasing its training data, extraction too…
-
AI-assisted proof for optimal packing of 11 squares verified
An AI-assisted proof for the optimal packing of 11 squares has been successfully verified using the Lean theorem prover. The proof, formalized in Lean 4.34.1 and Mathlib revision d13f23b723b8a846827a245b89c10fc7d3f11612…
-
Scott Armstrong streamlines formalization of complex math proofs with Lean/Mathlib
Scott Armstrong has demonstrated that formalizing complex mathematical proofs, including those related to PDEs and probability theory, can be achieved within 24-48 hours using the Lean theorem prover and its associated …
-
Anthropic's Fermat Proof Analysis: 13M Lines Due to Step Count, Not Verbosity
An analysis of Anthropic's formalized proof of Fermat's Last Theorem reveals that the reported 13 million lines of Lean code are accurate, but the length is due to an exceptionally high number of generated steps rather …
-
New StochBench benchmark tests LLMs on stochastic processes in Lean · 2 sources tracked
Researchers have introduced StochBench, a new benchmark designed to evaluate large language models on stochastic processes in the Lean 4 programming language. This benchmark features 450 graduate-level problems, address…