{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RBKZ3U6ZGYSQP7R5CMAQNRLJ4D","short_pith_number":"pith:RBKZ3U6Z","schema_version":"1.0","canonical_sha256":"88559dd3d9362507fe3d130106c569e0d0e5f6b1a3ed500bedab06bfdda54a8f","source":{"kind":"arxiv","id":"2504.01542","version":2},"attestation_state":"computed","paper":{"title":"Register Always Matters: Analysis of LLM Pretraining Data Through the Lens of Language Variation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amanda Myntti, Erik Henriksson, Sampo Pyysalo, Veronika Laippala","submitted_at":"2025-04-02T09:30:24Z","abstract_excerpt":"Pretraining data curation is a cornerstone in Large Language Model (LLM) development, leading to growing research on quality filtering of large web corpora. From statistical quality flags to LLM-based labelling systems, datasets are divided into categories, frequently reducing to a binary: those passing the filters are deemed as valuable examples, others are discarded as useless or detrimental. However, a more detailed understanding of the contribution of different kinds of texts to model performance is still largely lacking. In this article, we present the first study utilising registers or g"},"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":"2504.01542","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-02T09:30:24Z","cross_cats_sorted":[],"title_canon_sha256":"805bf0be4def17dbd56214d6d1f4280eb9b9333c1de4ca05c9f7d73c985e1213","abstract_canon_sha256":"19883d0ac1ffe5c0dabe7235bc68ad7f6c5841e1edbf0eb6cb07e66439135f8d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:03.230498Z","signature_b64":"XxDgw6353iSJYbLLV8a3KgebbujWO6d7qwy+pxjEtbvZz96Rd1ebwZziXRnFJzU9l8q/DvWh0CqdpGwSpk1oDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88559dd3d9362507fe3d130106c569e0d0e5f6b1a3ed500bedab06bfdda54a8f","last_reissued_at":"2026-05-25T02:01:03.229798Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:03.229798Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Register Always Matters: Analysis of LLM Pretraining Data Through the Lens of Language Variation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amanda Myntti, Erik Henriksson, Sampo Pyysalo, Veronika Laippala","submitted_at":"2025-04-02T09:30:24Z","abstract_excerpt":"Pretraining data curation is a cornerstone in Large Language Model (LLM) development, leading to growing research on quality filtering of large web corpora. From statistical quality flags to LLM-based labelling systems, datasets are divided into categories, frequently reducing to a binary: those passing the filters are deemed as valuable examples, others are discarded as useless or detrimental. However, a more detailed understanding of the contribution of different kinds of texts to model performance is still largely lacking. In this article, we present the first study utilising registers or g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.01542","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/2504.01542/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":"2504.01542","created_at":"2026-05-25T02:01:03.229919+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.01542v2","created_at":"2026-05-25T02:01:03.229919+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.01542","created_at":"2026-05-25T02:01:03.229919+00:00"},{"alias_kind":"pith_short_12","alias_value":"RBKZ3U6ZGYSQ","created_at":"2026-05-25T02:01:03.229919+00:00"},{"alias_kind":"pith_short_16","alias_value":"RBKZ3U6ZGYSQP7R5","created_at":"2026-05-25T02:01:03.229919+00:00"},{"alias_kind":"pith_short_8","alias_value":"RBKZ3U6Z","created_at":"2026-05-25T02:01:03.229919+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.23651","citing_title":"How Human-Like Are Large Language Models? A Register-Aware Linguistic Evaluation Framework","ref_index":143,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23651","citing_title":"How Human-Like Are Large Language Models? A Register-Aware Linguistic Evaluation Framework","ref_index":71,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17650","citing_title":"Measuring Distribution Shift in User Prompts and Its Effects on LLM Performance","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D","json":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D.json","graph_json":"https://pith.science/api/pith-number/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/graph.json","events_json":"https://pith.science/api/pith-number/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/events.json","paper":"https://pith.science/paper/RBKZ3U6Z"},"agent_actions":{"view_html":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D","download_json":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D.json","view_paper":"https://pith.science/paper/RBKZ3U6Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.01542&json=true","fetch_graph":"https://pith.science/api/pith-number/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/graph.json","fetch_events":"https://pith.science/api/pith-number/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/action/storage_attestation","attest_author":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/action/author_attestation","sign_citation":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/action/citation_signature","submit_replication":"https://pith.science/pith/RBKZ3U6ZGYSQP7R5CMAQNRLJ4D/action/replication_record"}},"created_at":"2026-05-25T02:01:03.229919+00:00","updated_at":"2026-05-25T02:01:03.229919+00:00"}