kitft / funsearchLinks
Genetic programming using LLMs
☆48Updated 7 months ago
Alternatives and similar repositories for funsearch
Users that are interested in funsearch are comparing it to the libraries listed below
Sorting:
- ☆193Updated 2 years ago
- An environment for learning formal mathematical reasoning from scratch☆71Updated last year
- Retrieval-Augmented Theorem Provers for Lean☆298Updated 8 months ago
- An updated version of miniF2F with lots of fixes and informal statements / solutions.☆92Updated 9 months ago
- Benchmark for undergraduate-level formal mathematics☆109Updated last year
- Learning Formal Mathematics from Intrinsic Motivation☆36Updated 3 months ago
- Tutorial on neural theorem proving☆177Updated last year
- Harmonic Datasets☆51Updated last year
- ☆209Updated 6 months ago
- llmstep: [L]LM proofstep suggestions in Lean 4.☆145Updated last year
- LLMs + Lean, on your laptop or in the cloud☆189Updated last week
- fast + parallel AlphaZero in JAX☆103Updated 9 months ago
- Latent Program Network (from the "Searching Latent Program Spaces" paper)☆98Updated 2 weeks ago
- The official repository for the paper Multilingual Mathematical Autoformalization☆37Updated last year
- A long-horizon, sparse-reward math environment for reinforcement learning. Official code repo for "What makes Math problems hard for rein…☆30Updated 2 months ago
- Official code for paper: INT: An Inequality Benchmark for Evaluating Generalization in Theorem Proving☆39Updated 2 years ago
- LeanUniverse: A Library for Consistent and Scalable Lean4 Dataset Management☆71Updated 9 months ago
- Scaling scaling laws with board games.☆53Updated 2 years ago
- An evaluation benchmark for undergraduate competition math in Lean4, Isabelle, Coq, and natural language.☆166Updated last week
- Neural theorem proving tutorial, version II☆39Updated last year
- ☆66Updated last month
- Automatic solver for plane geometry problems.☆68Updated last month
- A Gymnasium-based Environment of the Abstraction and Reasoning Corpus (ARC)☆68Updated last year
- LeanEuclid is a benchmark for autoformalization in the domain of Euclidean geometry, targeting the proof assistant Lean.☆110Updated 5 months ago
- A "code intepreter" for Lean☆70Updated last month
- ☆73Updated 3 years ago
- ProofNet dataset ported into Lean 4☆25Updated 4 months ago
- ☆114Updated last month
- Transformers for Mathematics Tutorial | Simons/SLMath Workshop on AI for Mathematics 2025☆40Updated 6 months ago
- Proof artifact co-training for Lean☆45Updated 2 years ago