{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:66YZ2SCWBX5AWHJ5KA5K5DR3MV","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"fc47aa9c60069c86e280d9d9199db31c3b7095ff2a53df44e578b8a20ef656af","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-31T02:05:45Z","title_canon_sha256":"db4a33a8662b3e750208b5d19117b5a3cfe66f5a46ad8d707ac586f2299d4bdf"},"schema_version":"1.0","source":{"id":"2405.20574","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.20574","created_at":"2026-07-05T08:56:21Z"},{"alias_kind":"arxiv_version","alias_value":"2405.20574v2","created_at":"2026-07-05T08:56:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.20574","created_at":"2026-07-05T08:56:21Z"},{"alias_kind":"pith_short_12","alias_value":"66YZ2SCWBX5A","created_at":"2026-07-05T08:56:21Z"},{"alias_kind":"pith_short_16","alias_value":"66YZ2SCWBX5AWHJ5","created_at":"2026-07-05T08:56:21Z"},{"alias_kind":"pith_short_8","alias_value":"66YZ2SCW","created_at":"2026-07-05T08:56:21Z"}],"graph_snapshots":[{"event_id":"sha256:be560410ddcd506f3ea60f5c6c3d6d2cdae46a6ed6ca2f2e5e4c1650719031d2","target":"graph","created_at":"2026-07-05T08:56:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2405.20574/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces the Open Ko-LLM Leaderboard and the Ko-H5 Benchmark as vital tools for evaluating Large Language Models (LLMs) in Korean. Incorporating private test sets while mirroring the English Open LLM Leaderboard, we establish a robust evaluation framework that has been well integrated in the Korean LLM community. We perform data leakage analysis that shows the benefit of private test sets along with a correlation study within the Ko-H5 benchmark and temporal analyses of the Ko-H5 score. Moreover, we present empirical support for the need to expand beyond set benchmarks. We hope th","authors_text":"Chanjun Park, Dahyun Kim, Hwalsuk Lee, Hyeonwoo Kim, Sanghoon Kim, Seonghwan Cho, Sukyung Lee, Yungi Kim","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-31T02:05:45Z","title":"Open Ko-LLM Leaderboard: Evaluating Large Language Models in Korean with Ko-H5 Benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.20574","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7dbe37d3ce76921395017acee2edc113f4ad8cfa25dc52c861e1c7f27de65142","target":"record","created_at":"2026-07-05T08:56:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"fc47aa9c60069c86e280d9d9199db31c3b7095ff2a53df44e578b8a20ef656af","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-31T02:05:45Z","title_canon_sha256":"db4a33a8662b3e750208b5d19117b5a3cfe66f5a46ad8d707ac586f2299d4bdf"},"schema_version":"1.0","source":{"id":"2405.20574","kind":"arxiv","version":2}},"canonical_sha256":"f7b19d48560dfa0b1d3d503aae8e3b6556739dff35bded03910ddb59c80a45ab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f7b19d48560dfa0b1d3d503aae8e3b6556739dff35bded03910ddb59c80a45ab","first_computed_at":"2026-07-05T08:56:21.347533Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:56:21.347533Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JcvuPnls4abg89D6vHv6adEXoffEvU5pVG5e6BEklfvmjSyc+dRvpUcKeh7o+KrHewRTdC5v76CJUQmJL5vtBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:56:21.348038Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.20574","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7dbe37d3ce76921395017acee2edc113f4ad8cfa25dc52c861e1c7f27de65142","sha256:be560410ddcd506f3ea60f5c6c3d6d2cdae46a6ed6ca2f2e5e4c1650719031d2"],"state_sha256":"3b006271361b59b28ef02e7cec1b62809490fd37ff3d2fc0cb128cfec5f1f331"}