{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:YTMYTTIY6DF7I7GD5MG5KCPGUO","short_pith_number":"pith:YTMYTTIY","schema_version":"1.0","canonical_sha256":"c4d989cd18f0cbf47cc3eb0dd509e6a3954d162c417e9c3e53305d811a1099e8","source":{"kind":"arxiv","id":"2411.14343","version":1},"attestation_state":"computed","paper":{"title":"UnifiedCrawl: Aggregated Common Crawl for Affordable Adaptation of LLMs on Low-Resource Languages","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"(2) Independent Researcher), Akhil Kedia (2), Bethel Melesse Tessema (1), Tae-Sun Chung (1) ((1) Ajou University","submitted_at":"2024-11-21T17:41:08Z","abstract_excerpt":"Large language models (LLMs) under-perform on low-resource languages due to limited training data. We present a method to efficiently collect text data for low-resource languages from the entire Common Crawl corpus. Our approach, UnifiedCrawl, filters and extracts common crawl using minimal compute resources, yielding mono-lingual datasets much larger than previously available sources. We demonstrate that leveraging this data to fine-tuning multilingual LLMs via efficient adapter methods (QLoRA) significantly boosts performance on the low-resource language, while minimizing VRAM usage. Our exp"},"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":"2411.14343","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-21T17:41:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5bb845b1cd593b74f139864246d9deebdd60a475a17eb0ccfae940b1f718f57b","abstract_canon_sha256":"ec60cefba0f6de8c5b9cebeef954ecd48f46a7c30255dd05003ad1dcfb8487b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:45.492741Z","signature_b64":"MJyVjX7yxN3GoZJ/7XJPkRXY+PGjZ0sFnkjThj0a3jV1jCfkC7oXXIdiGzR4UBvUvxo6DIWp0KRcahtpnP4GBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c4d989cd18f0cbf47cc3eb0dd509e6a3954d162c417e9c3e53305d811a1099e8","last_reissued_at":"2026-07-05T09:38:45.492224Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:45.492224Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UnifiedCrawl: Aggregated Common Crawl for Affordable Adaptation of LLMs on Low-Resource Languages","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"(2) Independent Researcher), Akhil Kedia (2), Bethel Melesse Tessema (1), Tae-Sun Chung (1) ((1) Ajou University","submitted_at":"2024-11-21T17:41:08Z","abstract_excerpt":"Large language models (LLMs) under-perform on low-resource languages due to limited training data. We present a method to efficiently collect text data for low-resource languages from the entire Common Crawl corpus. Our approach, UnifiedCrawl, filters and extracts common crawl using minimal compute resources, yielding mono-lingual datasets much larger than previously available sources. We demonstrate that leveraging this data to fine-tuning multilingual LLMs via efficient adapter methods (QLoRA) significantly boosts performance on the low-resource language, while minimizing VRAM usage. Our exp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.14343","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/2411.14343/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":"2411.14343","created_at":"2026-07-05T09:38:45.492284+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.14343v1","created_at":"2026-07-05T09:38:45.492284+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.14343","created_at":"2026-07-05T09:38:45.492284+00:00"},{"alias_kind":"pith_short_12","alias_value":"YTMYTTIY6DF7","created_at":"2026-07-05T09:38:45.492284+00:00"},{"alias_kind":"pith_short_16","alias_value":"YTMYTTIY6DF7I7GD","created_at":"2026-07-05T09:38:45.492284+00:00"},{"alias_kind":"pith_short_8","alias_value":"YTMYTTIY","created_at":"2026-07-05T09:38:45.492284+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.16356","citing_title":"Evaluating Binary Decision Biases in Large Language Models: Implications for Fair Agent-Based Financial Simulations","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO","json":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO.json","graph_json":"https://pith.science/api/pith-number/YTMYTTIY6DF7I7GD5MG5KCPGUO/graph.json","events_json":"https://pith.science/api/pith-number/YTMYTTIY6DF7I7GD5MG5KCPGUO/events.json","paper":"https://pith.science/paper/YTMYTTIY"},"agent_actions":{"view_html":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO","download_json":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO.json","view_paper":"https://pith.science/paper/YTMYTTIY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.14343&json=true","fetch_graph":"https://pith.science/api/pith-number/YTMYTTIY6DF7I7GD5MG5KCPGUO/graph.json","fetch_events":"https://pith.science/api/pith-number/YTMYTTIY6DF7I7GD5MG5KCPGUO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO/action/storage_attestation","attest_author":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO/action/author_attestation","sign_citation":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO/action/citation_signature","submit_replication":"https://pith.science/pith/YTMYTTIY6DF7I7GD5MG5KCPGUO/action/replication_record"}},"created_at":"2026-07-05T09:38:45.492284+00:00","updated_at":"2026-07-05T09:38:45.492284+00:00"}