{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PQPV64P75PVTI7QNQX5RZLLB5B","short_pith_number":"pith:PQPV64P7","schema_version":"1.0","canonical_sha256":"7c1f5f71ffebeb347e0d85fb1cad61e85cd361eb754a6acedcd53634c5e8f4ed","source":{"kind":"arxiv","id":"2606.03938","version":1},"attestation_state":"computed","paper":{"title":"q0: Primitives for Hyper-Epoch Pretraining","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Akshay Vegesna, Bishwas Mandal, Samip Dahal, Shmuel Berman","submitted_at":"2026-06-02T17:27:48Z","abstract_excerpt":"Multi-epoch training is becoming the standard now that compute is growing faster than the supply of high-quality text. But pretraining a single model saturates within a few passes, long before the compute budget is exhausted. We argue this calls for a conceptual shift from training a single model toward exploring a population of models and aggregating their predictions. We introduce hyper-epoch pretraining (q0), which turns a multi-epoch budget into a population of diverse models whose combined predictions reach a lower validation loss than a single refined model. q0 reduces to three core prim"},"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":"2606.03938","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T17:27:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2c7bb240633faa9ed5320f3330bdf330544e8d1b5098e7556f45e2f19fe17797","abstract_canon_sha256":"e521f01f9a76a8aa87193161a238d59a8f1eca1abe878a2ad6a65ae938ebc584"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T02:06:07.449388Z","signature_b64":"vUBw7cpt1ZaNDQBadjChQa0JENuBlcXW9/lSqnWYAEpRwG/gfZ1Z0U2IvWEx7zZfLwPq16p6JBu8hUMzrPoFBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7c1f5f71ffebeb347e0d85fb1cad61e85cd361eb754a6acedcd53634c5e8f4ed","last_reissued_at":"2026-06-03T02:06:07.448915Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T02:06:07.448915Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"q0: Primitives for Hyper-Epoch Pretraining","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Akshay Vegesna, Bishwas Mandal, Samip Dahal, Shmuel Berman","submitted_at":"2026-06-02T17:27:48Z","abstract_excerpt":"Multi-epoch training is becoming the standard now that compute is growing faster than the supply of high-quality text. But pretraining a single model saturates within a few passes, long before the compute budget is exhausted. We argue this calls for a conceptual shift from training a single model toward exploring a population of models and aggregating their predictions. We introduce hyper-epoch pretraining (q0), which turns a multi-epoch budget into a population of diverse models whose combined predictions reach a lower validation loss than a single refined model. q0 reduces to three core prim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03938","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/2606.03938/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":"2606.03938","created_at":"2026-06-03T02:06:07.448980+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03938v1","created_at":"2026-06-03T02:06:07.448980+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03938","created_at":"2026-06-03T02:06:07.448980+00:00"},{"alias_kind":"pith_short_12","alias_value":"PQPV64P75PVT","created_at":"2026-06-03T02:06:07.448980+00:00"},{"alias_kind":"pith_short_16","alias_value":"PQPV64P75PVTI7QN","created_at":"2026-06-03T02:06:07.448980+00:00"},{"alias_kind":"pith_short_8","alias_value":"PQPV64P7","created_at":"2026-06-03T02:06:07.448980+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/PQPV64P75PVTI7QNQX5RZLLB5B","json":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B.json","graph_json":"https://pith.science/api/pith-number/PQPV64P75PVTI7QNQX5RZLLB5B/graph.json","events_json":"https://pith.science/api/pith-number/PQPV64P75PVTI7QNQX5RZLLB5B/events.json","paper":"https://pith.science/paper/PQPV64P7"},"agent_actions":{"view_html":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B","download_json":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B.json","view_paper":"https://pith.science/paper/PQPV64P7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03938&json=true","fetch_graph":"https://pith.science/api/pith-number/PQPV64P75PVTI7QNQX5RZLLB5B/graph.json","fetch_events":"https://pith.science/api/pith-number/PQPV64P75PVTI7QNQX5RZLLB5B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B/action/storage_attestation","attest_author":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B/action/author_attestation","sign_citation":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B/action/citation_signature","submit_replication":"https://pith.science/pith/PQPV64P75PVTI7QNQX5RZLLB5B/action/replication_record"}},"created_at":"2026-06-03T02:06:07.448980+00:00","updated_at":"2026-06-03T02:06:07.448980+00:00"}