{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:G2NFR3QGCKDYGQ72MTR73M3MCM","short_pith_number":"pith:G2NFR3QG","canonical_record":{"source":{"id":"2412.15285","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-18T18:41:18Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b321559c408e2019c6282fafa7e98f10d555b26e29e38cee8d204b7907809a22","abstract_canon_sha256":"94ca532d542ecf4c40eb09d2c0dcabb99bef9e40479653e52448d2965008f655"},"schema_version":"1.0"},"canonical_sha256":"369a58ee0612878343fa64e3fdb36c1331bc0a9f99fbf2d7a98f4c75d4d98fd6","source":{"kind":"arxiv","id":"2412.15285","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15285","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15285v1","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15285","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"pith_short_12","alias_value":"G2NFR3QGCKDY","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"pith_short_16","alias_value":"G2NFR3QGCKDYGQ72","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"pith_short_8","alias_value":"G2NFR3QG","created_at":"2026-07-05T09:52:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:G2NFR3QGCKDYGQ72MTR73M3MCM","target":"record","payload":{"canonical_record":{"source":{"id":"2412.15285","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-18T18:41:18Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b321559c408e2019c6282fafa7e98f10d555b26e29e38cee8d204b7907809a22","abstract_canon_sha256":"94ca532d542ecf4c40eb09d2c0dcabb99bef9e40479653e52448d2965008f655"},"schema_version":"1.0"},"canonical_sha256":"369a58ee0612878343fa64e3fdb36c1331bc0a9f99fbf2d7a98f4c75d4d98fd6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:20.765225Z","signature_b64":"343YZineBPAYJihPwEb/cfZvnUMo/oz/rIkRwBLOvTJM6y36DMBHzN6GcW90ZZ7/niW3yA5CwUc5/6/Q5978BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"369a58ee0612878343fa64e3fdb36c1331bc0a9f99fbf2d7a98f4c75d4d98fd6","last_reissued_at":"2026-07-05T09:52:20.764786Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:20.764786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.15285","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:52:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nQEuLtnB3yTYO0SbpzQHjPfOqaW2stTrvyySBicqEeuYeBusHZRQJ0J5BpNpggV+/3avFPbTR1OeVZ33L5lJCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:32:34.561911Z"},"content_sha256":"d5576f80c53403c10a4ea45f1470e002a239d3da63c39db8ae7191fcd915ca4d","schema_version":"1.0","event_id":"sha256:d5576f80c53403c10a4ea45f1470e002a239d3da63c39db8ae7191fcd915ca4d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:G2NFR3QGCKDYGQ72MTR73M3MCM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Maximize Your Data's Potential: Enhancing LLM Accuracy with Two-Phase Pretraining","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bryan Catanzaro, Dan Su, Kezhi Kong, Mohammad Shoeybi, Mostofa Patwary, Shrimai Prabhumoye, Steven Feng","submitted_at":"2024-12-18T18:41:18Z","abstract_excerpt":"Pretraining large language models effectively requires strategic data selection, blending and ordering. However, key details about data mixtures especially their scalability to longer token horizons and larger model sizes remain underexplored due to limited disclosure by model developers. To address this, we formalize the concept of two-phase pretraining and conduct an extensive systematic study on how to select and mix data to maximize model accuracies for the two phases. Our findings illustrate that a two-phase approach for pretraining outperforms random data ordering and natural distributio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15285","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/2412.15285/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:52:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OdmRPMVHnQeQKwVhGq4R2brDPIWGuAOx9/gljzrWMeB8N+fOQp0je3CsxCwnTRkJ5PS5qrfogqeXtUD8WrerCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:32:34.562422Z"},"content_sha256":"fa6ac92d0c2fc4312842b91f9f098df84482cb66f0ac063146f1fde9a0cba31e","schema_version":"1.0","event_id":"sha256:fa6ac92d0c2fc4312842b91f9f098df84482cb66f0ac063146f1fde9a0cba31e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/G2NFR3QGCKDYGQ72MTR73M3MCM/bundle.json","state_url":"https://pith.science/pith/G2NFR3QGCKDYGQ72MTR73M3MCM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/G2NFR3QGCKDYGQ72MTR73M3MCM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T21:32:34Z","links":{"resolver":"https://pith.science/pith/G2NFR3QGCKDYGQ72MTR73M3MCM","bundle":"https://pith.science/pith/G2NFR3QGCKDYGQ72MTR73M3MCM/bundle.json","state":"https://pith.science/pith/G2NFR3QGCKDYGQ72MTR73M3MCM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/G2NFR3QGCKDYGQ72MTR73M3MCM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:G2NFR3QGCKDYGQ72MTR73M3MCM","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"94ca532d542ecf4c40eb09d2c0dcabb99bef9e40479653e52448d2965008f655","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-18T18:41:18Z","title_canon_sha256":"b321559c408e2019c6282fafa7e98f10d555b26e29e38cee8d204b7907809a22"},"schema_version":"1.0","source":{"id":"2412.15285","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15285","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15285v1","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15285","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"pith_short_12","alias_value":"G2NFR3QGCKDY","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"pith_short_16","alias_value":"G2NFR3QGCKDYGQ72","created_at":"2026-07-05T09:52:20Z"},{"alias_kind":"pith_short_8","alias_value":"G2NFR3QG","created_at":"2026-07-05T09:52:20Z"}],"graph_snapshots":[{"event_id":"sha256:fa6ac92d0c2fc4312842b91f9f098df84482cb66f0ac063146f1fde9a0cba31e","target":"graph","created_at":"2026-07-05T09:52:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2412.15285/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pretraining large language models effectively requires strategic data selection, blending and ordering. However, key details about data mixtures especially their scalability to longer token horizons and larger model sizes remain underexplored due to limited disclosure by model developers. To address this, we formalize the concept of two-phase pretraining and conduct an extensive systematic study on how to select and mix data to maximize model accuracies for the two phases. Our findings illustrate that a two-phase approach for pretraining outperforms random data ordering and natural distributio","authors_text":"Bryan Catanzaro, Dan Su, Kezhi Kong, Mohammad Shoeybi, Mostofa Patwary, Shrimai Prabhumoye, Steven Feng","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-18T18:41:18Z","title":"Maximize Your Data's Potential: Enhancing LLM Accuracy with Two-Phase Pretraining"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15285","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d5576f80c53403c10a4ea45f1470e002a239d3da63c39db8ae7191fcd915ca4d","target":"record","created_at":"2026-07-05T09:52:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"94ca532d542ecf4c40eb09d2c0dcabb99bef9e40479653e52448d2965008f655","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-18T18:41:18Z","title_canon_sha256":"b321559c408e2019c6282fafa7e98f10d555b26e29e38cee8d204b7907809a22"},"schema_version":"1.0","source":{"id":"2412.15285","kind":"arxiv","version":1}},"canonical_sha256":"369a58ee0612878343fa64e3fdb36c1331bc0a9f99fbf2d7a98f4c75d4d98fd6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"369a58ee0612878343fa64e3fdb36c1331bc0a9f99fbf2d7a98f4c75d4d98fd6","first_computed_at":"2026-07-05T09:52:20.764786Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:52:20.764786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"343YZineBPAYJihPwEb/cfZvnUMo/oz/rIkRwBLOvTJM6y36DMBHzN6GcW90ZZ7/niW3yA5CwUc5/6/Q5978BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:52:20.765225Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.15285","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d5576f80c53403c10a4ea45f1470e002a239d3da63c39db8ae7191fcd915ca4d","sha256:fa6ac92d0c2fc4312842b91f9f098df84482cb66f0ac063146f1fde9a0cba31e"],"state_sha256":"6a812748263ae54e4d4be39f83b8ce7f27e81048998b5a1c9de7c43f7da848c2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sLO6PSqsUJSZCM58bJIDYqciM/uzS8pZQV3Y4jxho/frn1BpEN+Sx+HTXHv4cMZz2Emhs9o4ga5p+40NT8vqCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T21:32:34.566043Z","bundle_sha256":"e94262b112635fd15a4c10d68bdb10cd549b71014c109a79e49ee723df382949"}}