{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GLMMTYDPAIF3RG3JBKW2SZGAS4","short_pith_number":"pith:GLMMTYDP","schema_version":"1.0","canonical_sha256":"32d8c9e06f020bb89b690aada964c09702084aa88bf2692f3c6aa4d8318773fa","source":{"kind":"arxiv","id":"2410.03083","version":1},"attestation_state":"computed","paper":{"title":"Scaling Parameter-Constrained Language Models with Quality Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Changsheng Zhao, Ernie Chang, Matteo Paltenghi, Patrick Huber, Pin-Jie Lin, Rastislav Rabatin, Vikas Chandra, Yang Li, Yangyang Shi, Zechun Liu","submitted_at":"2024-10-04T02:07:17Z","abstract_excerpt":"Scaling laws in language modeling traditionally quantify training loss as a function of dataset size and model parameters, providing compute-optimal estimates but often neglecting the impact of data quality on model generalization. In this paper, we extend the conventional understanding of scaling law by offering a microscopic view of data quality within the original formulation -- effective training tokens -- which we posit to be a critical determinant of performance for parameter-constrained language models. Specifically, we formulate the proposed term of effective training tokens to be a co"},"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":"2410.03083","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T02:07:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"16aad6771e347c44b47333380d47c73ca00c3e46ec591788728052cc9307a9c4","abstract_canon_sha256":"005310cdf4999c73d66ed8bcf80359751805d2d87ca52cab12913aa19108abac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:43.989378Z","signature_b64":"5EcUylrjewZVTtN3O0U/OsYWg2lSFH2bc/gPR4YTMk6iqQ5SoeCuAeeK3m4RKTdk+2VwIAyB8hPAindZozlhDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32d8c9e06f020bb89b690aada964c09702084aa88bf2692f3c6aa4d8318773fa","last_reissued_at":"2026-07-05T09:15:43.988871Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:43.988871Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Parameter-Constrained Language Models with Quality Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Changsheng Zhao, Ernie Chang, Matteo Paltenghi, Patrick Huber, Pin-Jie Lin, Rastislav Rabatin, Vikas Chandra, Yang Li, Yangyang Shi, Zechun Liu","submitted_at":"2024-10-04T02:07:17Z","abstract_excerpt":"Scaling laws in language modeling traditionally quantify training loss as a function of dataset size and model parameters, providing compute-optimal estimates but often neglecting the impact of data quality on model generalization. In this paper, we extend the conventional understanding of scaling law by offering a microscopic view of data quality within the original formulation -- effective training tokens -- which we posit to be a critical determinant of performance for parameter-constrained language models. Specifically, we formulate the proposed term of effective training tokens to be a co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03083","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/2410.03083/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":"2410.03083","created_at":"2026-07-05T09:15:43.988926+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.03083v1","created_at":"2026-07-05T09:15:43.988926+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03083","created_at":"2026-07-05T09:15:43.988926+00:00"},{"alias_kind":"pith_short_12","alias_value":"GLMMTYDPAIF3","created_at":"2026-07-05T09:15:43.988926+00:00"},{"alias_kind":"pith_short_16","alias_value":"GLMMTYDPAIF3RG3J","created_at":"2026-07-05T09:15:43.988926+00:00"},{"alias_kind":"pith_short_8","alias_value":"GLMMTYDP","created_at":"2026-07-05T09:15:43.988926+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18380","citing_title":"QSTRBench: a New Benchmark to Evaluate the Ability of Language Models to Reason with Qualitative Spatial and Temporal Calculi","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07546","citing_title":"On the Invariance and Generality of Neural Scaling Laws","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4","json":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4.json","graph_json":"https://pith.science/api/pith-number/GLMMTYDPAIF3RG3JBKW2SZGAS4/graph.json","events_json":"https://pith.science/api/pith-number/GLMMTYDPAIF3RG3JBKW2SZGAS4/events.json","paper":"https://pith.science/paper/GLMMTYDP"},"agent_actions":{"view_html":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4","download_json":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4.json","view_paper":"https://pith.science/paper/GLMMTYDP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.03083&json=true","fetch_graph":"https://pith.science/api/pith-number/GLMMTYDPAIF3RG3JBKW2SZGAS4/graph.json","fetch_events":"https://pith.science/api/pith-number/GLMMTYDPAIF3RG3JBKW2SZGAS4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4/action/storage_attestation","attest_author":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4/action/author_attestation","sign_citation":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4/action/citation_signature","submit_replication":"https://pith.science/pith/GLMMTYDPAIF3RG3JBKW2SZGAS4/action/replication_record"}},"created_at":"2026-07-05T09:15:43.988926+00:00","updated_at":"2026-07-05T09:15:43.988926+00:00"}