{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TYTJNVXUH4AG5N5XGXQBMQ7MS2","short_pith_number":"pith:TYTJNVXU","schema_version":"1.0","canonical_sha256":"9e2696d6f43f006eb7b735e01643ec969145eeeff1ef3b43781328c523cc8f38","source":{"kind":"arxiv","id":"2407.00416","version":2},"attestation_state":"computed","paper":{"title":"Too Late to Train, Too Early To Use? A Study on Necessity and Viability of Low-Resource Bengali LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hasnaen Adil, Haz Sameen Shahgir, Khondker Salman Sayeed, Ruwad Naswan, Satak Kumar Dey, Tamzeed Mahfuz","submitted_at":"2024-06-29T11:50:16Z","abstract_excerpt":"Each new generation of English-oriented Large Language Models (LLMs) exhibits enhanced cross-lingual transfer capabilities and significantly outperforms older LLMs on low-resource languages. This prompts the question: Is there a need for LLMs dedicated to a particular low-resource language? We aim to explore this question for Bengali, a low-to-moderate resource Indo-Aryan language native to the Bengal region of South Asia.\n  We compare the performance of open-weight and closed-source LLMs such as LLaMA-3 and GPT-4 against fine-tuned encoder-decoder models across a diverse set of Bengali downst"},"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":"2407.00416","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-29T11:50:16Z","cross_cats_sorted":[],"title_canon_sha256":"2ff511462df05bd93c77bef3d291504142f4b461a64fcf6fd30256c58784dfec","abstract_canon_sha256":"74f217269f00786bc75093298fc34e14a524e01ecb23b32f486cb8ab15be068d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:48:34.062008Z","signature_b64":"GGjKwzzoqyBkdZloqCSMM7nE8dRR66KccCN6KwxvUBFhHJPzI9/Ap23lc5Ijd1lkNx5BJq5RnumZQG11cYNLAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e2696d6f43f006eb7b735e01643ec969145eeeff1ef3b43781328c523cc8f38","last_reissued_at":"2026-07-05T09:48:34.061519Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:48:34.061519Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Too Late to Train, Too Early To Use? A Study on Necessity and Viability of Low-Resource Bengali LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hasnaen Adil, Haz Sameen Shahgir, Khondker Salman Sayeed, Ruwad Naswan, Satak Kumar Dey, Tamzeed Mahfuz","submitted_at":"2024-06-29T11:50:16Z","abstract_excerpt":"Each new generation of English-oriented Large Language Models (LLMs) exhibits enhanced cross-lingual transfer capabilities and significantly outperforms older LLMs on low-resource languages. This prompts the question: Is there a need for LLMs dedicated to a particular low-resource language? We aim to explore this question for Bengali, a low-to-moderate resource Indo-Aryan language native to the Bengal region of South Asia.\n  We compare the performance of open-weight and closed-source LLMs such as LLaMA-3 and GPT-4 against fine-tuned encoder-decoder models across a diverse set of Bengali downst"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00416","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/2407.00416/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":"2407.00416","created_at":"2026-07-05T09:48:34.061578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.00416v2","created_at":"2026-07-05T09:48:34.061578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00416","created_at":"2026-07-05T09:48:34.061578+00:00"},{"alias_kind":"pith_short_12","alias_value":"TYTJNVXUH4AG","created_at":"2026-07-05T09:48:34.061578+00:00"},{"alias_kind":"pith_short_16","alias_value":"TYTJNVXUH4AG5N5X","created_at":"2026-07-05T09:48:34.061578+00:00"},{"alias_kind":"pith_short_8","alias_value":"TYTJNVXU","created_at":"2026-07-05T09:48:34.061578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.04497","citing_title":"Opportunities and Challenges of Large Language Models for Low-Resource Languages in Humanities Research","ref_index":94,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2","json":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2.json","graph_json":"https://pith.science/api/pith-number/TYTJNVXUH4AG5N5XGXQBMQ7MS2/graph.json","events_json":"https://pith.science/api/pith-number/TYTJNVXUH4AG5N5XGXQBMQ7MS2/events.json","paper":"https://pith.science/paper/TYTJNVXU"},"agent_actions":{"view_html":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2","download_json":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2.json","view_paper":"https://pith.science/paper/TYTJNVXU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.00416&json=true","fetch_graph":"https://pith.science/api/pith-number/TYTJNVXUH4AG5N5XGXQBMQ7MS2/graph.json","fetch_events":"https://pith.science/api/pith-number/TYTJNVXUH4AG5N5XGXQBMQ7MS2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2/action/storage_attestation","attest_author":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2/action/author_attestation","sign_citation":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2/action/citation_signature","submit_replication":"https://pith.science/pith/TYTJNVXUH4AG5N5XGXQBMQ7MS2/action/replication_record"}},"created_at":"2026-07-05T09:48:34.061578+00:00","updated_at":"2026-07-05T09:48:34.061578+00:00"}