{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PXQP3MZ5QKL6DK5M3SDE2JHIUK","short_pith_number":"pith:PXQP3MZ5","schema_version":"1.0","canonical_sha256":"7de0fdb33d8297e1abacdc864d24e8a28f4892d7ade4e8b71fd8dfe56e4debcb","source":{"kind":"arxiv","id":"2407.12481","version":2},"attestation_state":"computed","paper":{"title":"Krutrim LLM: A Novel Tokenization Strategy for Multilingual Indic Languages with Petabyte-Scale Data Processing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chandra Khatri, Daud Ibrahim, Deepak Kumarr, Divyansh Rajput, Gautam Bhargava, Pidathala Sowjanya, Rahul Kumar, Rishabh Nahata, Shubham Kakde","submitted_at":"2024-07-17T11:06:27Z","abstract_excerpt":"We present a novel approach to data preparation for developing multilingual Indic large language model. Our meticulous data acquisition spans open-source and proprietary sources, including Common Crawl, Indic books, news articles, and Wikipedia, ensuring a diverse and rich linguistic representation. For each Indic language, we design a custom preprocessing pipeline to effectively eliminate redundant and low-quality text content. Additionally, we perform deduplication on Common Crawl data to address the redundancy present in 70% of the crawled web pages. This study focuses on developing high-qu"},"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":"2407.12481","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-17T11:06:27Z","cross_cats_sorted":[],"title_canon_sha256":"b66efe9dd54889d0043e391d56359527db74211f50016f9d3e6a68e93f7953ef","abstract_canon_sha256":"8cb0be24c7f945b2fc36498492c025943eb597f2459a446fbec5578ee6f420b3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:42:36.222543Z","signature_b64":"izdpLVsLqrfST6rA+R5battj5D95XEgMvrCxyVBHsdnFn/BO4KRbhzHBnIiTt1N3ebf3kDtnF/XtDi91hfK2Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7de0fdb33d8297e1abacdc864d24e8a28f4892d7ade4e8b71fd8dfe56e4debcb","last_reissued_at":"2026-07-05T10:42:36.222012Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:42:36.222012Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Krutrim LLM: A Novel Tokenization Strategy for Multilingual Indic Languages with Petabyte-Scale Data Processing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chandra Khatri, Daud Ibrahim, Deepak Kumarr, Divyansh Rajput, Gautam Bhargava, Pidathala Sowjanya, Rahul Kumar, Rishabh Nahata, Shubham Kakde","submitted_at":"2024-07-17T11:06:27Z","abstract_excerpt":"We present a novel approach to data preparation for developing multilingual Indic large language model. Our meticulous data acquisition spans open-source and proprietary sources, including Common Crawl, Indic books, news articles, and Wikipedia, ensuring a diverse and rich linguistic representation. For each Indic language, we design a custom preprocessing pipeline to effectively eliminate redundant and low-quality text content. Additionally, we perform deduplication on Common Crawl data to address the redundancy present in 70% of the crawled web pages. This study focuses on developing high-qu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.12481","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/2407.12481/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":"2407.12481","created_at":"2026-07-05T10:42:36.222073+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.12481v2","created_at":"2026-07-05T10:42:36.222073+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.12481","created_at":"2026-07-05T10:42:36.222073+00:00"},{"alias_kind":"pith_short_12","alias_value":"PXQP3MZ5QKL6","created_at":"2026-07-05T10:42:36.222073+00:00"},{"alias_kind":"pith_short_16","alias_value":"PXQP3MZ5QKL6DK5M","created_at":"2026-07-05T10:42:36.222073+00:00"},{"alias_kind":"pith_short_8","alias_value":"PXQP3MZ5","created_at":"2026-07-05T10:42:36.222073+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.29379","citing_title":"BrahmicTokenizer-131K: An Indic-Capable Drop-In Replacement for o200k_base","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11181","citing_title":"Code Mixologist : A Practitioner's Guide to Building Code-Mixed LLMs","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK","json":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK.json","graph_json":"https://pith.science/api/pith-number/PXQP3MZ5QKL6DK5M3SDE2JHIUK/graph.json","events_json":"https://pith.science/api/pith-number/PXQP3MZ5QKL6DK5M3SDE2JHIUK/events.json","paper":"https://pith.science/paper/PXQP3MZ5"},"agent_actions":{"view_html":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK","download_json":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK.json","view_paper":"https://pith.science/paper/PXQP3MZ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.12481&json=true","fetch_graph":"https://pith.science/api/pith-number/PXQP3MZ5QKL6DK5M3SDE2JHIUK/graph.json","fetch_events":"https://pith.science/api/pith-number/PXQP3MZ5QKL6DK5M3SDE2JHIUK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK/action/storage_attestation","attest_author":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK/action/author_attestation","sign_citation":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK/action/citation_signature","submit_replication":"https://pith.science/pith/PXQP3MZ5QKL6DK5M3SDE2JHIUK/action/replication_record"}},"created_at":"2026-07-05T10:42:36.222073+00:00","updated_at":"2026-07-05T10:42:36.222073+00:00"}