{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZLDKNJSDVP5OKZWZUMEDUTR7VZ","short_pith_number":"pith:ZLDKNJSD","canonical_record":{"source":{"id":"2507.22947","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-07-27T15:20:19Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"05419bba3bf42dbd8e0c520a63d16dc0a19091237ab5aa8dd04ed3dafe74ca17","abstract_canon_sha256":"8dd350d7db8f5df60230113e73ee1fa69799b362e61116084531c32e9a6187d5"},"schema_version":"1.0"},"canonical_sha256":"cac6a6a643abfae566d9a3083a4e3fae52f37212fe6a663514e2084ff1bab9db","source":{"kind":"arxiv","id":"2507.22947","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.22947","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"arxiv_version","alias_value":"2507.22947v1","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.22947","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"pith_short_12","alias_value":"ZLDKNJSDVP5O","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"pith_short_16","alias_value":"ZLDKNJSDVP5OKZWZ","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"pith_short_8","alias_value":"ZLDKNJSD","created_at":"2026-07-05T11:46:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZLDKNJSDVP5OKZWZUMEDUTR7VZ","target":"record","payload":{"canonical_record":{"source":{"id":"2507.22947","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-07-27T15:20:19Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"05419bba3bf42dbd8e0c520a63d16dc0a19091237ab5aa8dd04ed3dafe74ca17","abstract_canon_sha256":"8dd350d7db8f5df60230113e73ee1fa69799b362e61116084531c32e9a6187d5"},"schema_version":"1.0"},"canonical_sha256":"cac6a6a643abfae566d9a3083a4e3fae52f37212fe6a663514e2084ff1bab9db","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:46:02.654421Z","signature_b64":"TzEivsmlQZN4Js+mndy3NyLeYHc/uhk3yhHKbfOVdkUA/j5W34QQsSn/htvNt2WSMWK3OxUM8F939g/laCt/Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cac6a6a643abfae566d9a3083a4e3fae52f37212fe6a663514e2084ff1bab9db","last_reissued_at":"2026-07-05T11:46:02.653874Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:46:02.653874Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.22947","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:46:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AWb8zB/kPhZBgA9cr1/QcRCkA3/sHKn4oqYfa21zswtaSC387eHwMc0u1qyXs8qVBup5A6vNpwdppcN9lbvwCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T19:05:04.363681Z"},"content_sha256":"5ceb04c11d359bc115dc3b2ccb75b549d57c2bdfcae552433d7d72574921b826","schema_version":"1.0","event_id":"sha256:5ceb04c11d359bc115dc3b2ccb75b549d57c2bdfcae552433d7d72574921b826"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZLDKNJSDVP5OKZWZUMEDUTR7VZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ELMES: An Automated Framework for Evaluating Large Language Models in Educational Scenarios","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CY","authors_text":"Aimin Zhou, Bingdong Li, Bo Jiang, Hao Hao, Jian Chen, Min Zhang, Ruijia Li, Shou'ang Wei, Shuzhen Bi, Xin Lin, Xinyun Wang, Yu Song","submitted_at":"2025-07-27T15:20:19Z","abstract_excerpt":"The emergence of Large Language Models (LLMs) presents transformative opportunities for education, generating numerous novel application scenarios. However, significant challenges remain: evaluation metrics vary substantially across different educational scenarios, while many emerging scenarios lack appropriate assessment metrics. Current benchmarks predominantly measure general intelligence rather than pedagogical capabilities. To address this gap, we introduce ELMES, an open-source automated evaluation framework specifically designed for assessing LLMs in educational settings. ELMES features"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.22947","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/2507.22947/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:46:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MYSqqzDsTy/Yaj0XbGdo7TBetc0Kavp1FwOrldihYGnZusC55UljG1HLW210pBofKWRTbiC9sVwhN6IdvZ8yAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T19:05:04.364182Z"},"content_sha256":"1bb436751885f613cd9ccabd0ace41ab61e7a9b18e78d98ad146d69b77f3e943","schema_version":"1.0","event_id":"sha256:1bb436751885f613cd9ccabd0ace41ab61e7a9b18e78d98ad146d69b77f3e943"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZLDKNJSDVP5OKZWZUMEDUTR7VZ/bundle.json","state_url":"https://pith.science/pith/ZLDKNJSDVP5OKZWZUMEDUTR7VZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZLDKNJSDVP5OKZWZUMEDUTR7VZ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-10T19:05:04Z","links":{"resolver":"https://pith.science/pith/ZLDKNJSDVP5OKZWZUMEDUTR7VZ","bundle":"https://pith.science/pith/ZLDKNJSDVP5OKZWZUMEDUTR7VZ/bundle.json","state":"https://pith.science/pith/ZLDKNJSDVP5OKZWZUMEDUTR7VZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZLDKNJSDVP5OKZWZUMEDUTR7VZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZLDKNJSDVP5OKZWZUMEDUTR7VZ","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":"8dd350d7db8f5df60230113e73ee1fa69799b362e61116084531c32e9a6187d5","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-07-27T15:20:19Z","title_canon_sha256":"05419bba3bf42dbd8e0c520a63d16dc0a19091237ab5aa8dd04ed3dafe74ca17"},"schema_version":"1.0","source":{"id":"2507.22947","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.22947","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"arxiv_version","alias_value":"2507.22947v1","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.22947","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"pith_short_12","alias_value":"ZLDKNJSDVP5O","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"pith_short_16","alias_value":"ZLDKNJSDVP5OKZWZ","created_at":"2026-07-05T11:46:02Z"},{"alias_kind":"pith_short_8","alias_value":"ZLDKNJSD","created_at":"2026-07-05T11:46:02Z"}],"graph_snapshots":[{"event_id":"sha256:1bb436751885f613cd9ccabd0ace41ab61e7a9b18e78d98ad146d69b77f3e943","target":"graph","created_at":"2026-07-05T11:46:02Z","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/2507.22947/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The emergence of Large Language Models (LLMs) presents transformative opportunities for education, generating numerous novel application scenarios. However, significant challenges remain: evaluation metrics vary substantially across different educational scenarios, while many emerging scenarios lack appropriate assessment metrics. Current benchmarks predominantly measure general intelligence rather than pedagogical capabilities. To address this gap, we introduce ELMES, an open-source automated evaluation framework specifically designed for assessing LLMs in educational settings. ELMES features","authors_text":"Aimin Zhou, Bingdong Li, Bo Jiang, Hao Hao, Jian Chen, Min Zhang, Ruijia Li, Shou'ang Wei, Shuzhen Bi, Xin Lin, Xinyun Wang, Yu Song","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-07-27T15:20:19Z","title":"ELMES: An Automated Framework for Evaluating Large Language Models in Educational Scenarios"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.22947","kind":"arxiv","version":1},"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":"sha256:5ceb04c11d359bc115dc3b2ccb75b549d57c2bdfcae552433d7d72574921b826","target":"record","created_at":"2026-07-05T11:46:02Z","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":"8dd350d7db8f5df60230113e73ee1fa69799b362e61116084531c32e9a6187d5","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-07-27T15:20:19Z","title_canon_sha256":"05419bba3bf42dbd8e0c520a63d16dc0a19091237ab5aa8dd04ed3dafe74ca17"},"schema_version":"1.0","source":{"id":"2507.22947","kind":"arxiv","version":1}},"canonical_sha256":"cac6a6a643abfae566d9a3083a4e3fae52f37212fe6a663514e2084ff1bab9db","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cac6a6a643abfae566d9a3083a4e3fae52f37212fe6a663514e2084ff1bab9db","first_computed_at":"2026-07-05T11:46:02.653874Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:46:02.653874Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TzEivsmlQZN4Js+mndy3NyLeYHc/uhk3yhHKbfOVdkUA/j5W34QQsSn/htvNt2WSMWK3OxUM8F939g/laCt/Cg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:46:02.654421Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.22947","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5ceb04c11d359bc115dc3b2ccb75b549d57c2bdfcae552433d7d72574921b826","sha256:1bb436751885f613cd9ccabd0ace41ab61e7a9b18e78d98ad146d69b77f3e943"],"state_sha256":"7207031639a41322f748cc2071c38a0dc6ac1259c5b6db4fd1f03b1c0445c7d4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3O7hPK26mFXbhgA+QebxWn7vc7fQ+h5pcUKnJJNDqUtbqhjq2ptm6WVihDsHnOQvF2jL30gnqw/JFDY3ze2nDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T19:05:04.368525Z","bundle_sha256":"5cb5b5b79c5f97825ebdcce8572b5eaec28b1117584884406ad24a048c26ecdf"}}