{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GOI3GGPNMPK37Y6HGWBV6TSZN6","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":"b0b98e68b2af26b5f0f42dc4c68ea29272f98f2380861e51e1dff5eedf8e8be5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-31T18:43:49Z","title_canon_sha256":"c0d2b3c3b2ff7ba7403267e5cfa68ed7fb729d2414feaddad59baa977cfaa5a8"},"schema_version":"1.0","source":{"id":"2501.00593","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.00593","created_at":"2026-07-05T09:56:58Z"},{"alias_kind":"arxiv_version","alias_value":"2501.00593v2","created_at":"2026-07-05T09:56:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.00593","created_at":"2026-07-05T09:56:58Z"},{"alias_kind":"pith_short_12","alias_value":"GOI3GGPNMPK3","created_at":"2026-07-05T09:56:58Z"},{"alias_kind":"pith_short_16","alias_value":"GOI3GGPNMPK37Y6H","created_at":"2026-07-05T09:56:58Z"},{"alias_kind":"pith_short_8","alias_value":"GOI3GGPN","created_at":"2026-07-05T09:56:58Z"}],"graph_snapshots":[{"event_id":"sha256:c9c0d400c6440b028a544c1f9363932fd224c5a9a88f8aada5c133749cac26eb","target":"graph","created_at":"2026-07-05T09:56:58Z","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/2501.00593/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models have made remarkable advancements in understanding and generating human language, achieving notable success across a wide array of applications. However, evaluating these models remains a significant challenge, particularly for resource-limited languages such as Turkish. To address this gap, we introduce the Turkish MMLU (TR-MMLU) benchmark, a comprehensive evaluation framework designed to assess the linguistic and conceptual capabilities of large language models (LLMs) in Turkish. TR-MMLU is constructed from a carefully curated dataset comprising 6200 multiple-choice questions","authors_text":"Ahmet Semih G\\\"um\\\"u\\c{s}, Ali Arda Fincan, Banu Diri, M. Ali Bayram, \\\"Oner Ayta\\c{s}, Sava\\c{s} Y{\\i}ld{\\i}r{\\i}m","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-31T18:43:49Z","title":"Setting Standards in Turkish NLP: TR-MMLU for Large Language Model Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.00593","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:e75fb44556a43fe14090fc919f97db15ed79e1c5b3166c044050e10b82fca9e7","target":"record","created_at":"2026-07-05T09:56:58Z","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":"b0b98e68b2af26b5f0f42dc4c68ea29272f98f2380861e51e1dff5eedf8e8be5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-31T18:43:49Z","title_canon_sha256":"c0d2b3c3b2ff7ba7403267e5cfa68ed7fb729d2414feaddad59baa977cfaa5a8"},"schema_version":"1.0","source":{"id":"2501.00593","kind":"arxiv","version":2}},"canonical_sha256":"3391b319ed63d5bfe3c735835f4e596f9a8fb1623c0d7780560a6275891f538f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3391b319ed63d5bfe3c735835f4e596f9a8fb1623c0d7780560a6275891f538f","first_computed_at":"2026-07-05T09:56:58.893968Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:56:58.893968Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MGXIoF/agUje0IRH1q6JxtqUhLQ7sujIqRn5veGapKtvL4Ly40AG3rJuxGUaTZnJ+0I4Db0aW+3C5Y88hJmkDA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:56:58.894475Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.00593","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e75fb44556a43fe14090fc919f97db15ed79e1c5b3166c044050e10b82fca9e7","sha256:c9c0d400c6440b028a544c1f9363932fd224c5a9a88f8aada5c133749cac26eb"],"state_sha256":"a6f7382c35be9c793362e59ef3481c39e3dee1ab744460e5ccd6ea1bbbed0e2d"}