{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RZTIHC23UFC23VG2IXCNWFOBO2","short_pith_number":"pith:RZTIHC23","schema_version":"1.0","canonical_sha256":"8e66838b5ba145add4da45c4db15c176a93dadf0f6d79e869340977149e0dd92","source":{"kind":"arxiv","id":"2406.14882","version":1},"attestation_state":"computed","paper":{"title":"70B-parameter large language models in Japanese medical question-answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Issey Sukeda, Risa Kishikawa, Satoshi Kodera","submitted_at":"2024-06-21T06:04:10Z","abstract_excerpt":"Since the rise of large language models (LLMs), the domain adaptation has been one of the hot topics in various domains. Many medical LLMs trained with English medical dataset have made public recently. However, Japanese LLMs in medical domain still lack its research. Here we utilize multiple 70B-parameter LLMs for the first time and show that instruction tuning using Japanese medical question-answering dataset significantly improves the ability of Japanese LLMs to solve Japanese medical license exams, surpassing 50\\% in accuracy. In particular, the Japanese-centric models exhibit a more signi"},"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":"2406.14882","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-21T06:04:10Z","cross_cats_sorted":[],"title_canon_sha256":"51a1e728b9ea6c6fc9a65de2f60c29f7232dce4a8c417adca7a8f658510176a9","abstract_canon_sha256":"70d611cd98ec98b6b0c8bdeb783232ee99e31f757ec9469e972c83dada4e527b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:35:09.828809Z","signature_b64":"ZNTOz1h0DwK+CrzyecG+mDEcGOs1UW08GsvsYgECK4Odgqy3n8EVmS0eHi41CRR6stu0Lyhddhy/yjnIrTNGAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e66838b5ba145add4da45c4db15c176a93dadf0f6d79e869340977149e0dd92","last_reissued_at":"2026-07-05T08:35:09.828288Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:35:09.828288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"70B-parameter large language models in Japanese medical question-answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Issey Sukeda, Risa Kishikawa, Satoshi Kodera","submitted_at":"2024-06-21T06:04:10Z","abstract_excerpt":"Since the rise of large language models (LLMs), the domain adaptation has been one of the hot topics in various domains. Many medical LLMs trained with English medical dataset have made public recently. However, Japanese LLMs in medical domain still lack its research. Here we utilize multiple 70B-parameter LLMs for the first time and show that instruction tuning using Japanese medical question-answering dataset significantly improves the ability of Japanese LLMs to solve Japanese medical license exams, surpassing 50\\% in accuracy. In particular, the Japanese-centric models exhibit a more signi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.14882","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/2406.14882/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":"2406.14882","created_at":"2026-07-05T08:35:09.828350+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.14882v1","created_at":"2026-07-05T08:35:09.828350+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.14882","created_at":"2026-07-05T08:35:09.828350+00:00"},{"alias_kind":"pith_short_12","alias_value":"RZTIHC23UFC2","created_at":"2026-07-05T08:35:09.828350+00:00"},{"alias_kind":"pith_short_16","alias_value":"RZTIHC23UFC23VG2","created_at":"2026-07-05T08:35:09.828350+00:00"},{"alias_kind":"pith_short_8","alias_value":"RZTIHC23","created_at":"2026-07-05T08:35:09.828350+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.16661","citing_title":"A Japanese Language Model and Three New Evaluation Benchmarks for Pharmaceutical NLP","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2","json":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2.json","graph_json":"https://pith.science/api/pith-number/RZTIHC23UFC23VG2IXCNWFOBO2/graph.json","events_json":"https://pith.science/api/pith-number/RZTIHC23UFC23VG2IXCNWFOBO2/events.json","paper":"https://pith.science/paper/RZTIHC23"},"agent_actions":{"view_html":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2","download_json":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2.json","view_paper":"https://pith.science/paper/RZTIHC23","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.14882&json=true","fetch_graph":"https://pith.science/api/pith-number/RZTIHC23UFC23VG2IXCNWFOBO2/graph.json","fetch_events":"https://pith.science/api/pith-number/RZTIHC23UFC23VG2IXCNWFOBO2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2/action/storage_attestation","attest_author":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2/action/author_attestation","sign_citation":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2/action/citation_signature","submit_replication":"https://pith.science/pith/RZTIHC23UFC23VG2IXCNWFOBO2/action/replication_record"}},"created_at":"2026-07-05T08:35:09.828350+00:00","updated_at":"2026-07-05T08:35:09.828350+00:00"}