{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:44TYLDSL4T3TSIEBATL5Q6RUTX","short_pith_number":"pith:44TYLDSL","schema_version":"1.0","canonical_sha256":"e727858e4be4f739208104d7d87a349dc87bd37999fc5090224a67782bb173c1","source":{"kind":"arxiv","id":"2410.10878","version":2},"attestation_state":"computed","paper":{"title":"Herald: A Natural Language Annotated Lean 4 Dataset","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.LO"],"primary_cat":"cs.CL","authors_text":"Bin Dong, Guoxiong Gao, Jiedong Jiang, Qi Gao, Tianyi Xu, Yutong Wang, Zihan Qin","submitted_at":"2024-10-09T10:11:24Z","abstract_excerpt":"Verifiable formal languages like Lean have profoundly impacted mathematical reasoning, particularly through the use of large language models (LLMs) for automated reasoning. A significant challenge in training LLMs for these formal languages is the lack of parallel datasets that align natural language with formal language proofs. To address this challenge, this paper introduces a novel framework for translating the Mathlib4 corpus (a unified library of mathematics in formal language Lean 4) into natural language. Building upon this, we employ a dual augmentation strategy that combines tactic-ba"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.10878","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-09T10:11:24Z","cross_cats_sorted":["cs.AI","cs.LG","cs.LO"],"title_canon_sha256":"fb2d380c6215367dff0a66eb8911cc3a3f7588fc2ce294804337179e2961bc52","abstract_canon_sha256":"f2054f8ccdca06462f87e487cbf24b7a11fa5ca7c114ff1d5ddbb71932440806"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:20:40.627367Z","signature_b64":"aIx/J/qnwnFdSB8UAy8hITVmGqU/Gq2RntHWuukH04d79zP/k9mlbIKu3ucjwpZL7QfqgphgmktnNdSw6ufpCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e727858e4be4f739208104d7d87a349dc87bd37999fc5090224a67782bb173c1","last_reissued_at":"2026-07-05T10:20:40.626743Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:20:40.626743Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Herald: A Natural Language Annotated Lean 4 Dataset","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.LO"],"primary_cat":"cs.CL","authors_text":"Bin Dong, Guoxiong Gao, Jiedong Jiang, Qi Gao, Tianyi Xu, Yutong Wang, Zihan Qin","submitted_at":"2024-10-09T10:11:24Z","abstract_excerpt":"Verifiable formal languages like Lean have profoundly impacted mathematical reasoning, particularly through the use of large language models (LLMs) for automated reasoning. A significant challenge in training LLMs for these formal languages is the lack of parallel datasets that align natural language with formal language proofs. To address this challenge, this paper introduces a novel framework for translating the Mathlib4 corpus (a unified library of mathematics in formal language Lean 4) into natural language. Building upon this, we employ a dual augmentation strategy that combines tactic-ba"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.10878","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2410.10878/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.10878","created_at":"2026-07-05T10:20:40.626823+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.10878v2","created_at":"2026-07-05T10:20:40.626823+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.10878","created_at":"2026-07-05T10:20:40.626823+00:00"},{"alias_kind":"pith_short_12","alias_value":"44TYLDSL4T3T","created_at":"2026-07-05T10:20:40.626823+00:00"},{"alias_kind":"pith_short_16","alias_value":"44TYLDSL4T3TSIEB","created_at":"2026-07-05T10:20:40.626823+00:00"},{"alias_kind":"pith_short_8","alias_value":"44TYLDSL","created_at":"2026-07-05T10:20:40.626823+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05400","citing_title":"LeanMarathon: Toward Reliable AI Co-Mathematicians through Long-Horizon Lean Autoformalization","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14061","citing_title":"MathAtlas: A Benchmark for Autoformalization in the Wild","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13137","citing_title":"LeanSearch v2: Global Premise Retrieval for Lean 4 Theorem Proving","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13137","citing_title":"LeanSearch v2: Global Premise Retrieval for Lean 4 Theorem Proving","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03789","citing_title":"Automated Conjecture Resolution with Formal Verification","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13515","citing_title":"SFT-GRPO Data Overlap as a Post-Training Hyperparameter for Autoformalization","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX","json":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX.json","graph_json":"https://pith.science/api/pith-number/44TYLDSL4T3TSIEBATL5Q6RUTX/graph.json","events_json":"https://pith.science/api/pith-number/44TYLDSL4T3TSIEBATL5Q6RUTX/events.json","paper":"https://pith.science/paper/44TYLDSL"},"agent_actions":{"view_html":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX","download_json":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX.json","view_paper":"https://pith.science/paper/44TYLDSL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.10878&json=true","fetch_graph":"https://pith.science/api/pith-number/44TYLDSL4T3TSIEBATL5Q6RUTX/graph.json","fetch_events":"https://pith.science/api/pith-number/44TYLDSL4T3TSIEBATL5Q6RUTX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX/action/storage_attestation","attest_author":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX/action/author_attestation","sign_citation":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX/action/citation_signature","submit_replication":"https://pith.science/pith/44TYLDSL4T3TSIEBATL5Q6RUTX/action/replication_record"}},"created_at":"2026-07-05T10:20:40.626823+00:00","updated_at":"2026-07-05T10:20:40.626823+00:00"}