{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SQEYJYR3MPDGWKDCBAXWU5I5PN","short_pith_number":"pith:SQEYJYR3","canonical_record":{"source":{"id":"2507.08232","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T00:36:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0a60a7ef73c2400082429837f429d4ea7e7786434780f9c9d8c38483c043d5a9","abstract_canon_sha256":"c85095149b7db23ba1eca388a696dea307e007c7ca762ec6f744fbaf963f77cd"},"schema_version":"1.0"},"canonical_sha256":"940984e23b63c66b2862082f6a751d7b74255a64abd1a380f2d0f0f0f92ecb0c","source":{"kind":"arxiv","id":"2507.08232","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.08232","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"arxiv_version","alias_value":"2507.08232v1","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08232","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"pith_short_12","alias_value":"SQEYJYR3MPDG","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"pith_short_16","alias_value":"SQEYJYR3MPDGWKDC","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"pith_short_8","alias_value":"SQEYJYR3","created_at":"2026-07-05T11:35:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SQEYJYR3MPDGWKDCBAXWU5I5PN","target":"record","payload":{"canonical_record":{"source":{"id":"2507.08232","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T00:36:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0a60a7ef73c2400082429837f429d4ea7e7786434780f9c9d8c38483c043d5a9","abstract_canon_sha256":"c85095149b7db23ba1eca388a696dea307e007c7ca762ec6f744fbaf963f77cd"},"schema_version":"1.0"},"canonical_sha256":"940984e23b63c66b2862082f6a751d7b74255a64abd1a380f2d0f0f0f92ecb0c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:19.680566Z","signature_b64":"MN3TFCFMtf996j4BL6x7f+rAPaHpOE5ChxJpxZHG7TbPJ46Wt0fPxxhNJnWygIYPixOvh/7e0VyJ+RgNVbEnBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"940984e23b63c66b2862082f6a751d7b74255a64abd1a380f2d0f0f0f92ecb0c","last_reissued_at":"2026-07-05T11:35:19.680064Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:19.680064Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.08232","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":"2026-07-05T11:35:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S8XEDX0EQwxiuyR9nxj1u0a+/2xcU+XCSyxwKgzSRxzLJ/MjAfe7KP2ERpKElax4p6OkqQ4eZ5O9hWu7eO0eBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T22:40:15.597038Z"},"content_sha256":"c2bea1df2a83a56a43dfbcfaa0e5a3261ba8772b12e2c0bef955e6ba78402cfe","schema_version":"1.0","event_id":"sha256:c2bea1df2a83a56a43dfbcfaa0e5a3261ba8772b12e2c0bef955e6ba78402cfe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SQEYJYR3MPDGWKDCBAXWU5I5PN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Can LLMs Reliably Simulate Real Students' Abilities in Mathematics and Reading Comprehension?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ekaterina Kochmar, Kaushal Kumar Maurya, KV Aditya Srivatsa","submitted_at":"2025-07-11T00:36:57Z","abstract_excerpt":"Large Language Models (LLMs) are increasingly used as proxy students in the development of Intelligent Tutoring Systems (ITSs) and in piloting test questions. However, to what extent these proxy students accurately emulate the behavior and characteristics of real students remains an open question. To investigate this, we collected a dataset of 489 items from the National Assessment of Educational Progress (NAEP), covering mathematics and reading comprehension in grades 4, 8, and 12. We then apply an Item Response Theory (IRT) model to position 11 diverse and state-of-the-art LLMs on the same a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08232","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/2507.08232/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":"2026-07-05T11:35:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L6R4xdfoF/E8/3No6sxURJSKESsmPvIiZFBzzCcFgT1rsOu4/PRLQ2patnklQ+0H+ZgrbILy8qkzWHxoUsXsDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T22:40:15.597558Z"},"content_sha256":"b2b31291f83744eaafbe04079bd025903f59093ce3aed0b18bfc2b302f0f30f8","schema_version":"1.0","event_id":"sha256:b2b31291f83744eaafbe04079bd025903f59093ce3aed0b18bfc2b302f0f30f8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SQEYJYR3MPDGWKDCBAXWU5I5PN/bundle.json","state_url":"https://pith.science/pith/SQEYJYR3MPDGWKDCBAXWU5I5PN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SQEYJYR3MPDGWKDCBAXWU5I5PN/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":"2026-08-09T22:40:15Z","links":{"resolver":"https://pith.science/pith/SQEYJYR3MPDGWKDCBAXWU5I5PN","bundle":"https://pith.science/pith/SQEYJYR3MPDGWKDCBAXWU5I5PN/bundle.json","state":"https://pith.science/pith/SQEYJYR3MPDGWKDCBAXWU5I5PN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SQEYJYR3MPDGWKDCBAXWU5I5PN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SQEYJYR3MPDGWKDCBAXWU5I5PN","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":"c85095149b7db23ba1eca388a696dea307e007c7ca762ec6f744fbaf963f77cd","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T00:36:57Z","title_canon_sha256":"0a60a7ef73c2400082429837f429d4ea7e7786434780f9c9d8c38483c043d5a9"},"schema_version":"1.0","source":{"id":"2507.08232","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.08232","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"arxiv_version","alias_value":"2507.08232v1","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08232","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"pith_short_12","alias_value":"SQEYJYR3MPDG","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"pith_short_16","alias_value":"SQEYJYR3MPDGWKDC","created_at":"2026-07-05T11:35:19Z"},{"alias_kind":"pith_short_8","alias_value":"SQEYJYR3","created_at":"2026-07-05T11:35:19Z"}],"graph_snapshots":[{"event_id":"sha256:b2b31291f83744eaafbe04079bd025903f59093ce3aed0b18bfc2b302f0f30f8","target":"graph","created_at":"2026-07-05T11:35:19Z","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/2507.08232/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) are increasingly used as proxy students in the development of Intelligent Tutoring Systems (ITSs) and in piloting test questions. However, to what extent these proxy students accurately emulate the behavior and characteristics of real students remains an open question. To investigate this, we collected a dataset of 489 items from the National Assessment of Educational Progress (NAEP), covering mathematics and reading comprehension in grades 4, 8, and 12. We then apply an Item Response Theory (IRT) model to position 11 diverse and state-of-the-art LLMs on the same a","authors_text":"Ekaterina Kochmar, Kaushal Kumar Maurya, KV Aditya Srivatsa","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T00:36:57Z","title":"Can LLMs Reliably Simulate Real Students' Abilities in Mathematics and Reading Comprehension?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08232","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:c2bea1df2a83a56a43dfbcfaa0e5a3261ba8772b12e2c0bef955e6ba78402cfe","target":"record","created_at":"2026-07-05T11:35:19Z","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":"c85095149b7db23ba1eca388a696dea307e007c7ca762ec6f744fbaf963f77cd","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T00:36:57Z","title_canon_sha256":"0a60a7ef73c2400082429837f429d4ea7e7786434780f9c9d8c38483c043d5a9"},"schema_version":"1.0","source":{"id":"2507.08232","kind":"arxiv","version":1}},"canonical_sha256":"940984e23b63c66b2862082f6a751d7b74255a64abd1a380f2d0f0f0f92ecb0c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"940984e23b63c66b2862082f6a751d7b74255a64abd1a380f2d0f0f0f92ecb0c","first_computed_at":"2026-07-05T11:35:19.680064Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:35:19.680064Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MN3TFCFMtf996j4BL6x7f+rAPaHpOE5ChxJpxZHG7TbPJ46Wt0fPxxhNJnWygIYPixOvh/7e0VyJ+RgNVbEnBg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:35:19.680566Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.08232","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c2bea1df2a83a56a43dfbcfaa0e5a3261ba8772b12e2c0bef955e6ba78402cfe","sha256:b2b31291f83744eaafbe04079bd025903f59093ce3aed0b18bfc2b302f0f30f8"],"state_sha256":"0ea823587ad17c9f888b49b88e6c369003b65774d09eacf7dc50f5ba5db27c3c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tfiMLStHkOD4WMY46Uvyybk9R7nqqypa40NY2PiQPXNJjS/SggHgphZ9LooPrzSr/dNYz3mceg/ts3lL4S0NAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T22:40:15.604408Z","bundle_sha256":"7b18a3c51f8a3e12a8b3ab2c55b81dfde680242fb7388deef0f07caca609832a"}}