{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PCRQIQEX3DBIF3E4YUWM2W4QDZ","short_pith_number":"pith:PCRQIQEX","schema_version":"1.0","canonical_sha256":"78a3044097d8c282ec9cc52ccd5b901e68855780591df029db5e4758ea32bd9a","source":{"kind":"arxiv","id":"2506.21596","version":2},"attestation_state":"computed","paper":{"title":"Evaluating Multimodal Large Language Models on Educational Textbook Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IR"],"primary_cat":"cs.CL","authors_text":"Ali Alkhathlan, Amani Jamal, Anas Zafar, Areej Alhothali, Hessa A. Alawwad, Usman Naseem","submitted_at":"2025-06-18T19:31:35Z","abstract_excerpt":"Multimodal large language models (MLLMs) have shown success in vision-language tasks, but their ability to reason over complex educational materials remains largely untested. This work presents the first evaluation of state-of-the-art MLLMs, including LLaVA-1.5 and LLaMA 3.2-Vision, on the textbook question answering (TQA) task using the CK12-QA dataset. We introduce a multimodal retrieval-augmented generation (RAG) pipeline to simulate real-world learning by providing relevant lesson paragraphs and diagrams as context. Our zero-shot experiments reveal a critical trade-off: while retrieved con"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2506.21596","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-18T19:31:35Z","cross_cats_sorted":["cs.AI","cs.IR"],"title_canon_sha256":"b3ff352d27a8addc4a3a77d1824c32efec001249533c44a7b2afb53a61545adc","abstract_canon_sha256":"42ecfe5fd66800bd66cc2ad6e417f08ae64b460ad7cae9b726830dd2673c5b4a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:37:24.525244Z","signature_b64":"nl7xNjDRFp9fnJLHYDkevtV5U/F0KGovHloik+tnVBbXIX7RyqFPzxbASyJWwJgO1hl/FD2BRSKLEelFZTPIBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"78a3044097d8c282ec9cc52ccd5b901e68855780591df029db5e4758ea32bd9a","last_reissued_at":"2026-07-05T11:37:24.524728Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:37:24.524728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating Multimodal Large Language Models on Educational Textbook Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IR"],"primary_cat":"cs.CL","authors_text":"Ali Alkhathlan, Amani Jamal, Anas Zafar, Areej Alhothali, Hessa A. Alawwad, Usman Naseem","submitted_at":"2025-06-18T19:31:35Z","abstract_excerpt":"Multimodal large language models (MLLMs) have shown success in vision-language tasks, but their ability to reason over complex educational materials remains largely untested. This work presents the first evaluation of state-of-the-art MLLMs, including LLaVA-1.5 and LLaMA 3.2-Vision, on the textbook question answering (TQA) task using the CK12-QA dataset. We introduce a multimodal retrieval-augmented generation (RAG) pipeline to simulate real-world learning by providing relevant lesson paragraphs and diagrams as context. Our zero-shot experiments reveal a critical trade-off: while retrieved con"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.21596","kind":"arxiv","version":2},"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/2506.21596/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2506.21596","created_at":"2026-07-05T11:37:24.524792+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.21596v2","created_at":"2026-07-05T11:37:24.524792+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.21596","created_at":"2026-07-05T11:37:24.524792+00:00"},{"alias_kind":"pith_short_12","alias_value":"PCRQIQEX3DBI","created_at":"2026-07-05T11:37:24.524792+00:00"},{"alias_kind":"pith_short_16","alias_value":"PCRQIQEX3DBIF3E4","created_at":"2026-07-05T11:37:24.524792+00:00"},{"alias_kind":"pith_short_8","alias_value":"PCRQIQEX","created_at":"2026-07-05T11:37:24.524792+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ","json":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ.json","graph_json":"https://pith.science/api/pith-number/PCRQIQEX3DBIF3E4YUWM2W4QDZ/graph.json","events_json":"https://pith.science/api/pith-number/PCRQIQEX3DBIF3E4YUWM2W4QDZ/events.json","paper":"https://pith.science/paper/PCRQIQEX"},"agent_actions":{"view_html":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ","download_json":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ.json","view_paper":"https://pith.science/paper/PCRQIQEX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.21596&json=true","fetch_graph":"https://pith.science/api/pith-number/PCRQIQEX3DBIF3E4YUWM2W4QDZ/graph.json","fetch_events":"https://pith.science/api/pith-number/PCRQIQEX3DBIF3E4YUWM2W4QDZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ/action/storage_attestation","attest_author":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ/action/author_attestation","sign_citation":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ/action/citation_signature","submit_replication":"https://pith.science/pith/PCRQIQEX3DBIF3E4YUWM2W4QDZ/action/replication_record"}},"created_at":"2026-07-05T11:37:24.524792+00:00","updated_at":"2026-07-05T11:37:24.524792+00:00"}