Lean
PulseAugur coverage of Lean — every cluster mentioning Lean across labs, papers, and developer communities, ranked by signal.
- used by Mathlib 90%
- used by Rust 90%
- used by Gotit.pub 90%
- used by Lean 4 Programming Language 90%
- used by Riemann hypothesis 90%
- used by Erdős problems 90%
- competes with Rocq prover 80%
- affiliated with Mathlib 70%
- developed by Rust 70%
- used by DagsHub 70%
- used by miniF2F-test 70%
- used by ScienceCast 70%
19 day(s) with sentiment data
-
New benchmark reveals sycophancy in AI math formalization systems
Researchers have introduced FaithformBench, a new benchmark designed to evaluate the faithfulness of autoformalisation (AF) systems. These systems translate natural language reasoning into formal statements for proof as…
-
AI-assisted proof confirms 14 queens needed for 26x26 board domination
A mathematical proof has resolved the 26x26 queen domination number, establishing that 14 queens are both sufficient and necessary to dominate the board. This proof, verified by Lean 4.32.2 and an independent kernel, wa…
-
AI's potential role in formal code verification discussed
The discussion explores the potential for AI to advance formal verification in coding, suggesting that increased AI focus on this area could lead to more trustworthy code. The idea is that AI might enable simulations of…
-
Anthropic AI autonomously advances Riemann hypothesis research
An unreleased Anthropic AI model, operating autonomously, made significant progress on a sub-problem of the 150-year-old Riemann hypothesis. Over 36 hours, the AI model, named Claude Code, coordinated 60 sub-agents and …
-
Anthropic model advances on Riemann hypothesis, sparking debate
An unreleased Anthropic model has reportedly made significant progress on the Riemann hypothesis, a long-standing unsolved problem in mathematics concerning prime numbers. The model, prompted by a staff member with limi…
-
Anthropic's ClaudeAI advances Riemann hypothesis research with new bound
Anthropic's ClaudeAI has advanced research in mathematics by raising the lower bound for zeros of the Riemann hypothesis from 41.6% to 67.25%. This significant progress, formalized using the Lean proof assistant, does n…
-
New LLM methods boost verified code generation with integrated planning and proof search
Researchers have developed new methods for verified code generation, where large language models (LLMs) produce both executable programs and machine-checkable proofs of correctness. The first approach, P$^{3}$, integrat…
-
New voting method synthesis theorem for infinite domains
Researchers have developed a novel approach using SMT and Lean to synthesize and verify voting methods on infinite domains, a significant advancement over traditional finite-domain SAT solvers. This method addresses the…
-
Weaver framework combines weak verifiers to boost LLM accuracy
Researchers have developed Weaver, a framework designed to improve language model verification by combining multiple imperfect verifiers into a stronger, more accurate system. This approach aims to reduce the performanc…
-
OpenAI's GPT-5.6 Sol cuts costs by 20%, Astra makes math breakthroughs
OpenAI has achieved significant advancements with its frontier models, including GPT-5.6 Sol which autonomously optimized production GPU kernels, reducing serving costs by 20%. Concurrently, an internal model named Astr…
-
Mathematician refutes OpenAI's AI-generated proof of Connes rigidity conjecture
A mathematician has refuted OpenAI's claim that its new AI model proved the Connes rigidity conjecture. The mathematician, J. L. Nielsen from the University of Kansas, found that the AI's proof, consisting of 37,000 lin…
-
OpenAI's GPT 5.6 "Sol" model generates 10 new mathematical proofs
OpenAI has announced that an internal version of its upcoming model, GPT 5.6 "Sol", has generated 10 novel results in mathematics and theoretical computer science. These breakthroughs address long-standing open problems…
-
OpenAI's Astra model solves 10 major math problems, sparking debate
OpenAI has revealed details about its unreleased model, Astra, which has successfully solved ten major open problems in mathematics and theoretical computer science. These breakthroughs, achieved at a token cost of appr…
-
Human role in AI era debated: multiplier, consumer, or collaborator?
The role of humans in an era of Artificial Superintelligence (ASI) is being questioned, drawing parallels to chess where AI has surpassed human capabilities. While AI in chess has become a tool for consumption rather th…
-
Rocq prover surpasses Lean for program verification
The Rocq prover offers advantages over Lean for program verification, particularly in its handling of complex proofs and its integration with machine learning techniques. While Lean is a powerful tool for formal methods…
-
Rocq prover favored over Lean for program verification
The author argues that Rocq is a more suitable tool than Lean for program verification, particularly due to Rocq's direct support for executable coinductive types and cofixpoints. While Lean has introduced coinductive p…
-
Formalizing Flag Algebras in Lean for Graph Theory Proofs
Researchers have developed a machine-checked formalization of Razborov's flag algebra method within the Lean proof assistant. This formalization enables a compiler that transforms semidefinite programming output into ve…
-
New framework automates causal inference research using formal proofs
Researchers have developed CausalForge, a new framework designed to automate theoretical research in causal inference. This system integrates Causalean, a Lean proof assistant library with thousands of machine-checked d…
-
Language as a Latent Space: Human vs. AI Reasoning
This post explores the concept of language as a high-entropy latent space, drawing parallels between human language and strongly-typed programming languages like Rust and Haskell. The author argues that language, much l…
-
AI-generated Riemann hypothesis proof on arXiv poses security risk
A user on Mastodon questioned whether individuals would download and verify an AI-generated proof of the Riemann hypothesis posted on arXiv, provided it included a complete Lean formalization. The response cautioned aga…