{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:J6KJHU3E4OLGHC3GGMROENRJGQ","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":"b60a29fc92e9b0a3b50fb012ee7f3be3e262b418cbf65744b9123f7244aedec9","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T14:49:01Z","title_canon_sha256":"b6f62a8d6ba67e7ff319aa10a189a8db55bf7c3e4fb5c73bf1af416d5b762b85"},"schema_version":"1.0","source":{"id":"2506.18710","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.18710","created_at":"2026-07-05T11:30:02Z"},{"alias_kind":"arxiv_version","alias_value":"2506.18710v3","created_at":"2026-07-05T11:30:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18710","created_at":"2026-07-05T11:30:02Z"},{"alias_kind":"pith_short_12","alias_value":"J6KJHU3E4OLG","created_at":"2026-07-05T11:30:02Z"},{"alias_kind":"pith_short_16","alias_value":"J6KJHU3E4OLGHC3G","created_at":"2026-07-05T11:30:02Z"},{"alias_kind":"pith_short_8","alias_value":"J6KJHU3E","created_at":"2026-07-05T11:30:02Z"}],"graph_snapshots":[{"event_id":"sha256:f107d5fd1acf4a78942243633719adce63cad43ff9624d7550777f761a1ae70d","target":"graph","created_at":"2026-07-05T11:30:02Z","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/2506.18710/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Benchmarks like Massive Multitask Language Understanding (MMLU) have played a pivotal role in evaluating AI's knowledge and abilities across diverse domains. However, existing benchmarks predominantly focus on content knowledge, leaving a critical gap in assessing models' understanding of pedagogy - the method and practice of teaching. This paper introduces The Pedagogy Benchmark, a novel dataset designed to evaluate large language models on their Cross-Domain Pedagogical Knowledge (CDPK) and Special Education Needs and Disability (SEND) pedagogical knowledge. These benchmarks are built on a c","authors_text":"Alasdair Mackintosh, Amy Waldock, Jared Lee, Mar\\'ia Jos\\'e Ogando Portela, Maxime Leli\\`evre, Meng Liu, Natalia Vald\\'es Aspillaga, Oliver G. B. Garrod, Paul Atherton, Robin A. A. Ince","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T14:49:01Z","title":"Benchmarking the Pedagogical Knowledge of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18710","kind":"arxiv","version":3},"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":"sha256:74afd41bd47e89e01cef92c87b2cc3a389252f8c369383e3ade0828596805b0a","target":"record","created_at":"2026-07-05T11:30:02Z","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":"b60a29fc92e9b0a3b50fb012ee7f3be3e262b418cbf65744b9123f7244aedec9","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T14:49:01Z","title_canon_sha256":"b6f62a8d6ba67e7ff319aa10a189a8db55bf7c3e4fb5c73bf1af416d5b762b85"},"schema_version":"1.0","source":{"id":"2506.18710","kind":"arxiv","version":3}},"canonical_sha256":"4f9493d364e396638b663322e236293410d43aa53c0b20d13710fa2d5f98122c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4f9493d364e396638b663322e236293410d43aa53c0b20d13710fa2d5f98122c","first_computed_at":"2026-07-05T11:30:02.410083Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:30:02.410083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1BUzMLX5LkCY1gu+pNbI3MnM8tf6bqEMbRJxiAWoKwAsBnwZ2w7L94HvQM8B9Qc4kizwonx3LuZZU89eIpLLAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:30:02.410590Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.18710","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:74afd41bd47e89e01cef92c87b2cc3a389252f8c369383e3ade0828596805b0a","sha256:f107d5fd1acf4a78942243633719adce63cad43ff9624d7550777f761a1ae70d"],"state_sha256":"566ad7e8078c4196796411fc98dafc03bcd9d47096234377656c9f4d021c903a"}