{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ET75EIP3GNUMCAOH3UHD6TYXVD","short_pith_number":"pith:ET75EIP3","schema_version":"1.0","canonical_sha256":"24ffd221fb3368c101c7dd0e3f4f17a8c9fa884a22378344e73018acd646609b","source":{"kind":"arxiv","id":"2606.02991","version":1},"attestation_state":"computed","paper":{"title":"Pretraining Language Models on Historical Text","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Freda Shi, Junchi Yu, Niclas Griesshaber, Philip Torr, Xiaoxi Luo, Yao Lu, Yixuan Wang, Zachary Shinnick","submitted_at":"2026-06-02T00:59:06Z","abstract_excerpt":"We introduce TypewriterLM, a 7.24B History language model (LM) trained exclusively on English text predating 1913. Developing History LMs requires addressing challenges in data quality and availability, preventing temporal leakage, designing temporally consistent post-training pipelines, and constructing reliable evaluations. To address these issues, we construct TypewriterCorpus, a 54B-token historical corpus collected from diverse archival and linguistically annotated sources with extensive data cleaning and leakage mitigation procedures. Furthermore, we introduce lexically grounded instruct"},"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":"2606.02991","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-02T00:59:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"abb6d070159cf9695a6468e7abb84cb68b0b330d07224963fcfae08a6980fb2b","abstract_canon_sha256":"6e70dbb297999322c837c45399be4e67b79a617c4b426e6910e558a65b34be95"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:28.694991Z","signature_b64":"gUk1KLRdTtJbkQDMT3ofH4cOWzBm5QVcnwj1KDSJrtxgRfCemkHZUNffxNl5UzKfQc5cCLp2d6SDOAhRuk0DDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24ffd221fb3368c101c7dd0e3f4f17a8c9fa884a22378344e73018acd646609b","last_reissued_at":"2026-06-03T01:05:28.694576Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:28.694576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pretraining Language Models on Historical Text","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Freda Shi, Junchi Yu, Niclas Griesshaber, Philip Torr, Xiaoxi Luo, Yao Lu, Yixuan Wang, Zachary Shinnick","submitted_at":"2026-06-02T00:59:06Z","abstract_excerpt":"We introduce TypewriterLM, a 7.24B History language model (LM) trained exclusively on English text predating 1913. Developing History LMs requires addressing challenges in data quality and availability, preventing temporal leakage, designing temporally consistent post-training pipelines, and constructing reliable evaluations. To address these issues, we construct TypewriterCorpus, a 54B-token historical corpus collected from diverse archival and linguistically annotated sources with extensive data cleaning and leakage mitigation procedures. Furthermore, we introduce lexically grounded instruct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02991","kind":"arxiv","version":1},"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/2606.02991/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":"2606.02991","created_at":"2026-06-03T01:05:28.694628+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.02991v1","created_at":"2026-06-03T01:05:28.694628+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02991","created_at":"2026-06-03T01:05:28.694628+00:00"},{"alias_kind":"pith_short_12","alias_value":"ET75EIP3GNUM","created_at":"2026-06-03T01:05:28.694628+00:00"},{"alias_kind":"pith_short_16","alias_value":"ET75EIP3GNUMCAOH","created_at":"2026-06-03T01:05:28.694628+00:00"},{"alias_kind":"pith_short_8","alias_value":"ET75EIP3","created_at":"2026-06-03T01:05:28.694628+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD","json":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD.json","graph_json":"https://pith.science/api/pith-number/ET75EIP3GNUMCAOH3UHD6TYXVD/graph.json","events_json":"https://pith.science/api/pith-number/ET75EIP3GNUMCAOH3UHD6TYXVD/events.json","paper":"https://pith.science/paper/ET75EIP3"},"agent_actions":{"view_html":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD","download_json":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD.json","view_paper":"https://pith.science/paper/ET75EIP3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.02991&json=true","fetch_graph":"https://pith.science/api/pith-number/ET75EIP3GNUMCAOH3UHD6TYXVD/graph.json","fetch_events":"https://pith.science/api/pith-number/ET75EIP3GNUMCAOH3UHD6TYXVD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD/action/storage_attestation","attest_author":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD/action/author_attestation","sign_citation":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD/action/citation_signature","submit_replication":"https://pith.science/pith/ET75EIP3GNUMCAOH3UHD6TYXVD/action/replication_record"}},"created_at":"2026-06-03T01:05:28.694628+00:00","updated_at":"2026-06-03T01:05:28.694628+00:00"}