{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:7LY3HRXZKNLHQR4VALA5AFXK7P","short_pith_number":"pith:7LY3HRXZ","canonical_record":{"source":{"id":"2606.16246","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-15T05:48:57Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4ff4490126be0156d7c57edfbc0e4c948f418f131e317057ddbed034275c9c19","abstract_canon_sha256":"3ef618b32d03211fc4c417661e7e5ece3fc71c9ad537e461da74851beae6528c"},"schema_version":"1.0"},"canonical_sha256":"faf1b3c6f9535678479502c1d016eafbca43730b0d672fff26c0a2cd9515b645","source":{"kind":"arxiv","id":"2606.16246","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.16246","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.16246v2","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.16246","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"pith_short_12","alias_value":"7LY3HRXZKNLH","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"pith_short_16","alias_value":"7LY3HRXZKNLHQR4V","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"pith_short_8","alias_value":"7LY3HRXZ","created_at":"2026-06-23T01:13:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:7LY3HRXZKNLHQR4VALA5AFXK7P","target":"record","payload":{"canonical_record":{"source":{"id":"2606.16246","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-15T05:48:57Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4ff4490126be0156d7c57edfbc0e4c948f418f131e317057ddbed034275c9c19","abstract_canon_sha256":"3ef618b32d03211fc4c417661e7e5ece3fc71c9ad537e461da74851beae6528c"},"schema_version":"1.0"},"canonical_sha256":"faf1b3c6f9535678479502c1d016eafbca43730b0d672fff26c0a2cd9515b645","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:13:06.827542Z","signature_b64":"wUfSl0jZ1/sEjmydz/UOOWWcxnwMLJWvbJSzzXYGzUhZJY+4lSttqhet1iVpUVL49yJ/r0fTeZBD9v5PTtplDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"faf1b3c6f9535678479502c1d016eafbca43730b0d672fff26c0a2cd9515b645","last_reissued_at":"2026-06-23T01:13:06.826929Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:13:06.826929Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.16246","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-23T01:13:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DCM46AdYsu76077iu2OtaZgEBWNU0toWG+rE73E12rtlzAnDyLyfDc15ogWvV2ul1BKMU2Hn5eE+A4J0EyEOBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T03:04:21.345150Z"},"content_sha256":"efe381e122f0c51949e091efbadf23857d31d6ae3d1c59a96cece528bac27d03","schema_version":"1.0","event_id":"sha256:efe381e122f0c51949e091efbadf23857d31d6ae3d1c59a96cece528bac27d03"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:7LY3HRXZKNLHQR4VALA5AFXK7P","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Demystifying Training-Time Augmentation for Data-Constrained Language Model Pretraining","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Fan Bai, Michael K. Chen, Xikun Zhang, Zhengding Hu, Zhen Wang","submitted_at":"2026-06-15T05:48:57Z","abstract_excerpt":"As AI labs approach a data ceiling where compute capacity outpaces the rate of new high-quality text generation, language model pretraining is shifting toward a data-constrained, compute-abundant regime that demands productive multi-epoch training on fixed corpora. Standard autoregressive (AR) pretraining overfits severely in this setting, reaching its optimum early and then continuously deteriorating. We investigate training-time data augmentation as a regularizer to mitigate this overfitting and enable productive training for hundreds of epochs on the same data. We introduce three orthogonal"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.16246","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/2606.16246/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":"2026-06-23T01:13:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9D8W6xpPZ6t/QWqFMoaOk9iXtbZlx2byuOcOeUmaYwT1e0akCfewJL77ZDeJIAMGi5A7vNizBhXRSFYxZu4nAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T03:04:21.345830Z"},"content_sha256":"45ce52cbd3bbb4fda72da4f230b12aeec90e90c31409539f1978a81017a0ab4f","schema_version":"1.0","event_id":"sha256:45ce52cbd3bbb4fda72da4f230b12aeec90e90c31409539f1978a81017a0ab4f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7LY3HRXZKNLHQR4VALA5AFXK7P/bundle.json","state_url":"https://pith.science/pith/7LY3HRXZKNLHQR4VALA5AFXK7P/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7LY3HRXZKNLHQR4VALA5AFXK7P/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":"2026-06-30T03:04:21Z","links":{"resolver":"https://pith.science/pith/7LY3HRXZKNLHQR4VALA5AFXK7P","bundle":"https://pith.science/pith/7LY3HRXZKNLHQR4VALA5AFXK7P/bundle.json","state":"https://pith.science/pith/7LY3HRXZKNLHQR4VALA5AFXK7P/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7LY3HRXZKNLHQR4VALA5AFXK7P/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7LY3HRXZKNLHQR4VALA5AFXK7P","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":"3ef618b32d03211fc4c417661e7e5ece3fc71c9ad537e461da74851beae6528c","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-15T05:48:57Z","title_canon_sha256":"4ff4490126be0156d7c57edfbc0e4c948f418f131e317057ddbed034275c9c19"},"schema_version":"1.0","source":{"id":"2606.16246","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.16246","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.16246v2","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.16246","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"pith_short_12","alias_value":"7LY3HRXZKNLH","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"pith_short_16","alias_value":"7LY3HRXZKNLHQR4V","created_at":"2026-06-23T01:13:06Z"},{"alias_kind":"pith_short_8","alias_value":"7LY3HRXZ","created_at":"2026-06-23T01:13:06Z"}],"graph_snapshots":[{"event_id":"sha256:45ce52cbd3bbb4fda72da4f230b12aeec90e90c31409539f1978a81017a0ab4f","target":"graph","created_at":"2026-06-23T01:13:06Z","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/2606.16246/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As AI labs approach a data ceiling where compute capacity outpaces the rate of new high-quality text generation, language model pretraining is shifting toward a data-constrained, compute-abundant regime that demands productive multi-epoch training on fixed corpora. Standard autoregressive (AR) pretraining overfits severely in this setting, reaching its optimum early and then continuously deteriorating. We investigate training-time data augmentation as a regularizer to mitigate this overfitting and enable productive training for hundreds of epochs on the same data. We introduce three orthogonal","authors_text":"Fan Bai, Michael K. Chen, Xikun Zhang, Zhengding Hu, Zhen Wang","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-15T05:48:57Z","title":"Demystifying Training-Time Augmentation for Data-Constrained Language Model Pretraining"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.16246","kind":"arxiv","version":2},"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:efe381e122f0c51949e091efbadf23857d31d6ae3d1c59a96cece528bac27d03","target":"record","created_at":"2026-06-23T01:13:06Z","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":"3ef618b32d03211fc4c417661e7e5ece3fc71c9ad537e461da74851beae6528c","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-15T05:48:57Z","title_canon_sha256":"4ff4490126be0156d7c57edfbc0e4c948f418f131e317057ddbed034275c9c19"},"schema_version":"1.0","source":{"id":"2606.16246","kind":"arxiv","version":2}},"canonical_sha256":"faf1b3c6f9535678479502c1d016eafbca43730b0d672fff26c0a2cd9515b645","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"faf1b3c6f9535678479502c1d016eafbca43730b0d672fff26c0a2cd9515b645","first_computed_at":"2026-06-23T01:13:06.826929Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T01:13:06.826929Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wUfSl0jZ1/sEjmydz/UOOWWcxnwMLJWvbJSzzXYGzUhZJY+4lSttqhet1iVpUVL49yJ/r0fTeZBD9v5PTtplDA==","signature_status":"signed_v1","signed_at":"2026-06-23T01:13:06.827542Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.16246","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:efe381e122f0c51949e091efbadf23857d31d6ae3d1c59a96cece528bac27d03","sha256:45ce52cbd3bbb4fda72da4f230b12aeec90e90c31409539f1978a81017a0ab4f"],"state_sha256":"1e454d7d9ceb85a32290f289b1e6ad55fd385c2421c90801968d261ebd98cc92"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nj2dXwToFa9IMMsi6A/2+vsTmqCuMoVeB3kCG/giu25iJZpRqcVB2vir6NTBlCYGHpXqeyWBZKMnZAhvD/yOBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T03:04:21.349446Z","bundle_sha256":"4f79ac6357503747928fd7cd3d43615f929830fb4a88f30f185ce8aee6521d33"}}