{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TRTY7OSUZCY2KDVIDJUEHI4E5Q","short_pith_number":"pith:TRTY7OSU","schema_version":"1.0","canonical_sha256":"9c678fba54c8b1a50ea81a6843a384ec2e8052e4dd40f04f38a399fd7d457511","source":{"kind":"arxiv","id":"2306.13840","version":4},"attestation_state":"computed","paper":{"title":"Beyond Scale: The Diversity Coefficient as a Data Quality Metric for Variability in Natural Language Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.NE"],"primary_cat":"cs.CL","authors_text":"Allison Casasola, Alycia Lee, Brando Miranda, Elyas Obbad, Rylan Schaeffer, Sanmi Koyejo, Sudharsan Sundar","submitted_at":"2023-06-24T02:25:56Z","abstract_excerpt":"Current trends in pre-training Large Language Models (LLMs) primarily focus on the scaling of model and dataset size. While the quality of pre-training data is considered an important factor for training powerful LLMs, it remains a nebulous concept that has not been rigorously characterized. To this end, we propose a formalization of one key aspect of data quality -- measuring the variability of natural language data -- specifically via a measure we call the diversity coefficient. Our empirical analysis shows that the proposed diversity coefficient aligns with the intuitive properties of diver"},"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":"2306.13840","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-24T02:25:56Z","cross_cats_sorted":["cs.AI","cs.LG","cs.NE"],"title_canon_sha256":"bbc9e5c34fe77a1ab9d2ae736abf8077b891814a065368c4867a0654294560cd","abstract_canon_sha256":"51c052b3bd2c0404e1bb88623d5bebe071830e7b8d8feb65b6bdccf3eb41949c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:00.307000Z","signature_b64":"lWoSkfuiEeLPJTzGbplRqu7UfpzvV8AV5L0wlz5H+tK2DHCU79ZhGoq2xBt28w4SlY7AVE+6wizV6paJ/7mKCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c678fba54c8b1a50ea81a6843a384ec2e8052e4dd40f04f38a399fd7d457511","last_reissued_at":"2026-07-05T11:31:00.306437Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:00.306437Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Scale: The Diversity Coefficient as a Data Quality Metric for Variability in Natural Language Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.NE"],"primary_cat":"cs.CL","authors_text":"Allison Casasola, Alycia Lee, Brando Miranda, Elyas Obbad, Rylan Schaeffer, Sanmi Koyejo, Sudharsan Sundar","submitted_at":"2023-06-24T02:25:56Z","abstract_excerpt":"Current trends in pre-training Large Language Models (LLMs) primarily focus on the scaling of model and dataset size. While the quality of pre-training data is considered an important factor for training powerful LLMs, it remains a nebulous concept that has not been rigorously characterized. To this end, we propose a formalization of one key aspect of data quality -- measuring the variability of natural language data -- specifically via a measure we call the diversity coefficient. Our empirical analysis shows that the proposed diversity coefficient aligns with the intuitive properties of diver"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.13840","kind":"arxiv","version":4},"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/2306.13840/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":"2306.13840","created_at":"2026-07-05T11:31:00.306510+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.13840v4","created_at":"2026-07-05T11:31:00.306510+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.13840","created_at":"2026-07-05T11:31:00.306510+00:00"},{"alias_kind":"pith_short_12","alias_value":"TRTY7OSUZCY2","created_at":"2026-07-05T11:31:00.306510+00:00"},{"alias_kind":"pith_short_16","alias_value":"TRTY7OSUZCY2KDVI","created_at":"2026-07-05T11:31:00.306510+00:00"},{"alias_kind":"pith_short_8","alias_value":"TRTY7OSU","created_at":"2026-07-05T11:31:00.306510+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.18361","citing_title":"Synthetic Data Generation for Training Diversified Commonsense Reasoning Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02442","citing_title":"Measuring AI Reasoning: A Guide for Researchers","ref_index":145,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q","json":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q.json","graph_json":"https://pith.science/api/pith-number/TRTY7OSUZCY2KDVIDJUEHI4E5Q/graph.json","events_json":"https://pith.science/api/pith-number/TRTY7OSUZCY2KDVIDJUEHI4E5Q/events.json","paper":"https://pith.science/paper/TRTY7OSU"},"agent_actions":{"view_html":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q","download_json":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q.json","view_paper":"https://pith.science/paper/TRTY7OSU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.13840&json=true","fetch_graph":"https://pith.science/api/pith-number/TRTY7OSUZCY2KDVIDJUEHI4E5Q/graph.json","fetch_events":"https://pith.science/api/pith-number/TRTY7OSUZCY2KDVIDJUEHI4E5Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q/action/storage_attestation","attest_author":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q/action/author_attestation","sign_citation":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q/action/citation_signature","submit_replication":"https://pith.science/pith/TRTY7OSUZCY2KDVIDJUEHI4E5Q/action/replication_record"}},"created_at":"2026-07-05T11:31:00.306510+00:00","updated_at":"2026-07-05T11:31:00.306510+00:00"}