{"success":true,"data":{"author":"Google DeepMind and formal-conjectures contributors; extraction by phanerozoic","bucketAccessType":"public","bucketName":"phanerozoic/Lean4-FormalConjectures","bucketPrefix":"","category":"Mathematics","cloudProvider":"huggingface","createdAt":"2026-11-21T22:16:38.791Z","description":"Declarations from Google DeepMind's formal-conjectures repository, a collection of Lean 4 formalizations of open mathematical conjectures and problems, such as Erdős problems, together with solved variants and supporting lemmas. There are 4,557 entries from 896 files in 88 sub-libraries, extracted at upstream commit b9b8aa0f. 3,759 entries carry docstrings with the informal problem statement and source. Each row gives the formal statement, the proof body (often sorry for statements that are still open), the declaration kind, the name, the library and file, the imports, and the identifiers it uses. The repository serves as a benchmark target for automated theorem provers and for checking autoformalization against human-written formal statements of research-level problems. It is one Parquet file; load it with pandas or Hugging Face datasets. The upstream formal-conjectures repository is released by Google DeepMind under the Apache-2.0 license.","displayTitle":"Formal Conjectures: 4,557 Lean 4 Statements of Open Problems and Conjectures (Google DeepMind formal-conjectures, phanerozoic extraction)","huggingFaceUrl":"https://huggingface.co/datasets/phanerozoic/Lean4-FormalConjectures","id":"2a13146c-ebb4-5e29-b15e-0b8ebfd1878d","license":"Apache-2.0","licenseEvidenceUrl":"https://github.com/google-deepmind/formal-conjectures/blob/main/LICENSE","objectCountEstimate":3,"probeStatus":"ok","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"phanerozoic/Lean4-FormalConjectures\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"phanerozoic/Lean4-FormalConjectures\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=phanerozoic, license:apache-2.0, size_categories:1K<n<10K, format:parquet, downloads=73"],"format":"mixed","generatedAt":"2026-11-21T22:31:57.788Z","layout":{"example":"data/train-00000-of-00001.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"data/train-00000-of-00001.parquet","truncatedAt":512}],"sampleObjects":["data/train-00000-of-00001.parquet"],"sidecars":[{"bytes":"---\r\nlicense: apache-2.0\r\ntask_categories:\r\n  - text-generation\r\n  - feature-extraction\r\nlanguage:\r\n  - en\r\ntags:\r\n  - theorem-proving\r\n  - formal-methods\r\n  - lean4\r\nsize_categories:\r\n  - 1K<n<10K\r\ndataset_info:\r\n  features:\r\n  - {name: statement, dtype: string}\r\n  - {name: proof, dtype: string}\r\n  - {name: type, dtype: string}\r\n  - {name: symbolic_name, dtype: string}\r\n  - {name: library, dtype: string}\r\n  - {name: filename, dtype: string}\r\n  - {name: imports, list: string}\r\n  - {name: deps, list: string}\r\n  - {name: docstring, dtype: string}\r\n  - {name: source_url, dtype: string}\r\n  - {name: commit, dtype: string}\r\n  splits:\r\n  - {name: train, num_examples: 4557}\r\n  config_name: default\r\nconfigs:\r\n- config_name: default\r\n  data_files:\r\n  - split: train\r\n    path: data/train-*\r\n---\r\n\r\n# Lean4-FormalConjectures\r\n\r\nStructured dataset from **formal-conjectures** — Google DeepMind formalized conjectures.\r\n\r\n## Source\r\n\r\n- Repository: https://github.com/google-deepmind/formal-conjectures\r\n- Commit: `b9b8aa0fd6170e482798f7c78c163acacd40e083`\r\n- Files: 896\r\n- License: apache-2.0\r\n\r\n## Schema\r\n\r\n| Column | Type | Description |\r\n|--------|------|-------------|\r\n| statement | string | Declaration signature/claim with the leading keyword removed (verbatim slice); the full declaration minus its proof |\r\n| proof | string | Verbatim proof/body, empty if the declaration has none |\r\n| type | string | Declaration keyword |\r\n| symbolic_name | string | Declaration identifier |\r\n| library | string | Sub-library |\r\n| filename | string | Repository-relative source path |\r\n| imports | list[string] | File-level `Require`/`Import` modules |\r\n| deps | list[string] | Intra-corpus identifiers referenced |\r\n| docstring | string | Preceding documentation comment, empty if absent |\r\n| source_url | string | Upstream repository |\r\n| commit | string | Upstream commit extracted |\r\n\r\n## Statistics\r\n\r\n- Entries: 4,557\r\n- With proof: 4,472 (98.1%)\r\n- With docstring: 3,759 (82.5%)\r\n- Libraries: 88\r\n\r\n### By type\r\n\r\n| Type | Count |\r\n|---|---|\r\n| theorem | 2,859 |\r\n| def | 1,143 |\r\n| lemma | 364 |\r\n| abbrev | 81 |\r\n| structure | 53 |\r\n| instance | 24 |\r\n| inductive | 15 |\r\n| class | 13 |\r\n| elab | 3 |\r\n| macro | 2 |\r\n\r\n## Example\r\n\r\n```\r\nS (S₀ : List ℤ) (n : ℕ) : List ℤ\r\nmatch n with\r\n  | 0 => S₀\r\n  | n + 1 => (S S₀ n) ++ [Int.ofNat (k (S S₀ n))]\r\n```\r\n\r\n- type: def | symbolic_name: `Arxiv..S` | FormalConjectures/Arxiv/0912.2382/CurlingNumberConjecture.lean\r\n\r\n## Use\r\n\r\nEach declaration is split into a `statement` (signature/claim) and a `proof` (body) that are disjoint\r\nand together form the complete declaration, for proof modeling, autoformalization, retrieval, and\r\ndependency analysis via `deps`.\r\n\r\n## Citation\r\n\r\n```bibtex\r\n@misc{lean4_formalconjectures_dataset,\r\n  title  = {Lean4-FormalConjectures},\r\n  author = {Norton, Charles},\r\n  year   = {2026},\r\n  note   = {Extracted from https://github.com/google-deepmind/formal-conjectures, commit b9b8aa0fd617},\r\n  url    = {https://huggingface.co/datasets/phanerozoic/Lean4-FormalConjectures}\r\n}\r\n```\r\n","encoding":"utf8","key":"README.md","name":"README.md","size":3081,"truncated":false}],"sourceObjectKey":"data/train-00000-of-00001.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"phanerozoic/Lean4-FormalConjectures\")\nprint(ds)","starterSamplePath":"data/train-00000-of-00001.parquet","totalObjects":3},"summary":"Declarations from Google DeepMind's formal-conjectures repository, a collection of Lean 4 formalizations of open mathematical conjectures and problems, such as Erdős problems,…","title":"Formal Conjectures: 4,557 Lean 4 Statements of Open Problems and Conjectures (Google DeepMind formal-conjectures, phanerozoic extraction)","updatedAt":"2026-11-22T01:08:47.419Z","visibility":"public","websiteUrl":"https://github.com/google-deepmind/formal-conjectures"}}