{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SF5W7II5T2I3ZGCJLAYC4AITQU","short_pith_number":"pith:SF5W7II5","schema_version":"1.0","canonical_sha256":"917b6fa11d9e91bc984958302e01138538bf477536724567f9317c6e96f00855","source":{"kind":"arxiv","id":"2210.14891","version":17},"attestation_state":"computed","paper":{"title":"Broken Neural Scaling Laws","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"David Krueger, Ethan Caballero, Irina Rish, Kshitij Gupta","submitted_at":"2022-10-26T17:45:01Z","abstract_excerpt":"We present a smoothly broken power law functional form (that we refer to as a Broken Neural Scaling Law (BNSL)) that accurately models & extrapolates the scaling behaviors of deep neural networks (i.e. how the evaluation metric of interest varies as amount of compute used for training (or inference), number of model parameters, training dataset size, model input size, number of training steps, or upstream performance varies) for various architectures & for each of various tasks within a large & diverse set of upstream & downstream tasks, in zero-shot, prompted, & finetuned settings. This set i"},"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":"2210.14891","kind":"arxiv","version":17},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-10-26T17:45:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6402f203e6bee551f066e1fb6a903ad97b19c8838da4cfd1a1cb38b2bfe2e9b0","abstract_canon_sha256":"3f643982fc3eccae559e9ca46e91c5eb7f9d8f22e8de58a4d7b449ae16b23069"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:33:41.557271Z","signature_b64":"rBH4ivWOOfrULyFKa6hr8o15pv3ddPNELA/ynCMg5de+Nj2H/q4KEzkG0OvfQneQ4rhB4HgYqtdwKh0NM5GdDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"917b6fa11d9e91bc984958302e01138538bf477536724567f9317c6e96f00855","last_reissued_at":"2026-07-05T06:33:41.556804Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:33:41.556804Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Broken Neural Scaling Laws","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"David Krueger, Ethan Caballero, Irina Rish, Kshitij Gupta","submitted_at":"2022-10-26T17:45:01Z","abstract_excerpt":"We present a smoothly broken power law functional form (that we refer to as a Broken Neural Scaling Law (BNSL)) that accurately models & extrapolates the scaling behaviors of deep neural networks (i.e. how the evaluation metric of interest varies as amount of compute used for training (or inference), number of model parameters, training dataset size, model input size, number of training steps, or upstream performance varies) for various architectures & for each of various tasks within a large & diverse set of upstream & downstream tasks, in zero-shot, prompted, & finetuned settings. This set i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.14891","kind":"arxiv","version":17},"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/2210.14891/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":"2210.14891","created_at":"2026-07-05T06:33:41.556861+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.14891v17","created_at":"2026-07-05T06:33:41.556861+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.14891","created_at":"2026-07-05T06:33:41.556861+00:00"},{"alias_kind":"pith_short_12","alias_value":"SF5W7II5T2I3","created_at":"2026-07-05T06:33:41.556861+00:00"},{"alias_kind":"pith_short_16","alias_value":"SF5W7II5T2I3ZGCJ","created_at":"2026-07-05T06:33:41.556861+00:00"},{"alias_kind":"pith_short_8","alias_value":"SF5W7II5","created_at":"2026-07-05T06:33:41.556861+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.26248","citing_title":"Unified Neural Scaling Laws","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27914","citing_title":"Does Capability Transfer to Subjective Behavior -- and Would Our Instruments Tell Us? A Self-Evolving, Trust-by-Construction Evaluation Paradigm","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18838","citing_title":"Lying Is Just a Phase: The Hidden Alignment Transition in Language Model Scaling","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18840","citing_title":"The Growing Pains of Frontier Models: When Leaderboards Stop Separating and What to Measure Next","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2304.01373","citing_title":"Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling","ref_index":210,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05209","citing_title":"Are Flat Minima an Illusion?","ref_index":118,"is_internal_anchor":false},{"citing_arxiv_id":"2308.01825","citing_title":"Scaling Relationship on Learning Mathematical Reasoning with Large Language Models","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24294","citing_title":"AI-Native Autonomous Infrastructure (ANAI): A Formal Framework for the Next General-Purpose Technology","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU","json":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU.json","graph_json":"https://pith.science/api/pith-number/SF5W7II5T2I3ZGCJLAYC4AITQU/graph.json","events_json":"https://pith.science/api/pith-number/SF5W7II5T2I3ZGCJLAYC4AITQU/events.json","paper":"https://pith.science/paper/SF5W7II5"},"agent_actions":{"view_html":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU","download_json":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU.json","view_paper":"https://pith.science/paper/SF5W7II5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.14891&json=true","fetch_graph":"https://pith.science/api/pith-number/SF5W7II5T2I3ZGCJLAYC4AITQU/graph.json","fetch_events":"https://pith.science/api/pith-number/SF5W7II5T2I3ZGCJLAYC4AITQU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU/action/storage_attestation","attest_author":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU/action/author_attestation","sign_citation":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU/action/citation_signature","submit_replication":"https://pith.science/pith/SF5W7II5T2I3ZGCJLAYC4AITQU/action/replication_record"}},"created_at":"2026-07-05T06:33:41.556861+00:00","updated_at":"2026-07-05T06:33:41.556861+00:00"}