{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ACYLMDBJGR4BYMFKABYD5DC2RC","short_pith_number":"pith:ACYLMDBJ","schema_version":"1.0","canonical_sha256":"00b0b60c2934781c30aa00703e8c5a88aca8b59d5daa92380da9445b29f50414","source":{"kind":"arxiv","id":"2507.08924","version":2},"attestation_state":"computed","paper":{"title":"From KMMLU-Redux to KMMLU-Pro: A Professional Korean Benchmark Suite for LLM Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guijin Son, Jinsik Lee, Seokhee Hong, Soyeon Kim, Sunkyoung Kim, Yeonjung Hong","submitted_at":"2025-07-11T17:56:32Z","abstract_excerpt":"The development of Large Language Models (LLMs) requires robust benchmarks that encompass not only academic domains but also industrial fields to effectively evaluate their applicability in real-world scenarios. In this paper, we introduce two Korean expert-level benchmarks. KMMLU-Redux, reconstructed from the existing KMMLU, consists of questions from the Korean National Technical Qualification exams, with critical errors removed to enhance reliability. KMMLU-Pro is based on Korean National Professional Licensure exams to reflect professional knowledge in Korea. Our experiments demonstrate th"},"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":"2507.08924","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T17:56:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"920437f53e5eb30220824d396eced4d1af9592754a9b2eca38ba595bbfa97922","abstract_canon_sha256":"3dc5d511c88b041149b68e1636b77b9d795066adc642af0df8e24b7ad1d58a1f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:39:13.400552Z","signature_b64":"fsVozNAXQnDIlbj/aWf2UrCiHQEYGi1SysNMhcAUdjQ2dc+z2H/zNxF0fe8pdZttbmHpoExHewFIXN5/WCo7Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00b0b60c2934781c30aa00703e8c5a88aca8b59d5daa92380da9445b29f50414","last_reissued_at":"2026-07-05T11:39:13.400106Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:39:13.400106Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From KMMLU-Redux to KMMLU-Pro: A Professional Korean Benchmark Suite for LLM Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guijin Son, Jinsik Lee, Seokhee Hong, Soyeon Kim, Sunkyoung Kim, Yeonjung Hong","submitted_at":"2025-07-11T17:56:32Z","abstract_excerpt":"The development of Large Language Models (LLMs) requires robust benchmarks that encompass not only academic domains but also industrial fields to effectively evaluate their applicability in real-world scenarios. In this paper, we introduce two Korean expert-level benchmarks. KMMLU-Redux, reconstructed from the existing KMMLU, consists of questions from the Korean National Technical Qualification exams, with critical errors removed to enhance reliability. KMMLU-Pro is based on Korean National Professional Licensure exams to reflect professional knowledge in Korea. Our experiments demonstrate th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08924","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/2507.08924/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":"2507.08924","created_at":"2026-07-05T11:39:13.400161+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.08924v2","created_at":"2026-07-05T11:39:13.400161+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08924","created_at":"2026-07-05T11:39:13.400161+00:00"},{"alias_kind":"pith_short_12","alias_value":"ACYLMDBJGR4B","created_at":"2026-07-05T11:39:13.400161+00:00"},{"alias_kind":"pith_short_16","alias_value":"ACYLMDBJGR4BYMFK","created_at":"2026-07-05T11:39:13.400161+00:00"},{"alias_kind":"pith_short_8","alias_value":"ACYLMDBJ","created_at":"2026-07-05T11:39:13.400161+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.17448","citing_title":"Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17448","citing_title":"Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC","json":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC.json","graph_json":"https://pith.science/api/pith-number/ACYLMDBJGR4BYMFKABYD5DC2RC/graph.json","events_json":"https://pith.science/api/pith-number/ACYLMDBJGR4BYMFKABYD5DC2RC/events.json","paper":"https://pith.science/paper/ACYLMDBJ"},"agent_actions":{"view_html":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC","download_json":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC.json","view_paper":"https://pith.science/paper/ACYLMDBJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.08924&json=true","fetch_graph":"https://pith.science/api/pith-number/ACYLMDBJGR4BYMFKABYD5DC2RC/graph.json","fetch_events":"https://pith.science/api/pith-number/ACYLMDBJGR4BYMFKABYD5DC2RC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC/action/storage_attestation","attest_author":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC/action/author_attestation","sign_citation":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC/action/citation_signature","submit_replication":"https://pith.science/pith/ACYLMDBJGR4BYMFKABYD5DC2RC/action/replication_record"}},"created_at":"2026-07-05T11:39:13.400161+00:00","updated_at":"2026-07-05T11:39:13.400161+00:00"}