{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:X3XFOMA3RG5JE7G57MTWAOSQSQ","short_pith_number":"pith:X3XFOMA3","schema_version":"1.0","canonical_sha256":"beee57301b89ba927cddfb27603a50943e9bb0c97a032071d12cd127b252fd60","source":{"kind":"arxiv","id":"2509.16780","version":3},"attestation_state":"computed","paper":{"title":"Comparing RAG and GraphRAG for Page-Level Retrieval Question Answering on a Math Textbook","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC"],"primary_cat":"cs.IR","authors_text":"Chuangji Li, Eason Chen, Eric Li, Jionghao Lin, Kenneth R. Koedinger, Zimo Xiao","submitted_at":"2025-09-20T19:06:49Z","abstract_excerpt":"Large language models (LLMs) show promise as educational aids but often lack alignment with specific course materials. We investigate Retrieval-Augmented Generation (RAG) and GraphRAG for page-level question answering on an undergraduate mathematics textbook. Using a curated dataset of 477 question-answer pairs, each tied to a specific textbook page, we compare five embedding-based RAG models, a BM25 baseline, and GraphRAG across two metrics: retrieval accuracy (whether the correct page is retrieved) and answer quality (F1 score). Our results show that embedding-based RAG outperforms GraphRAG "},"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":"2509.16780","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-09-20T19:06:49Z","cross_cats_sorted":["cs.AI","cs.HC"],"title_canon_sha256":"6b389ad01efc478a8fa80a0bee0e94727061ef14a837954e6da881c59935bff6","abstract_canon_sha256":"44e17e30b9ef5097a718d7b6adfeac98352bf4c082527e36fce8cb9dca9577d4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-29T01:25:29.376719Z","signature_b64":"C9blcrp8GqxiRzalFaQ6+VeqCFw6aSK+OB08i2V6JnHR25HyqeSu+sfc9L0BeJxFHpI2pEItETF51PSD2TRZBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"beee57301b89ba927cddfb27603a50943e9bb0c97a032071d12cd127b252fd60","last_reissued_at":"2026-07-29T01:25:29.375764Z","signature_status":"signed_v1","first_computed_at":"2026-07-29T01:25:29.375764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Comparing RAG and GraphRAG for Page-Level Retrieval Question Answering on a Math Textbook","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC"],"primary_cat":"cs.IR","authors_text":"Chuangji Li, Eason Chen, Eric Li, Jionghao Lin, Kenneth R. Koedinger, Zimo Xiao","submitted_at":"2025-09-20T19:06:49Z","abstract_excerpt":"Large language models (LLMs) show promise as educational aids but often lack alignment with specific course materials. We investigate Retrieval-Augmented Generation (RAG) and GraphRAG for page-level question answering on an undergraduate mathematics textbook. Using a curated dataset of 477 question-answer pairs, each tied to a specific textbook page, we compare five embedding-based RAG models, a BM25 baseline, and GraphRAG across two metrics: retrieval accuracy (whether the correct page is retrieved) and answer quality (F1 score). Our results show that embedding-based RAG outperforms GraphRAG "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.16780","kind":"arxiv","version":3},"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/2509.16780/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":"2509.16780","created_at":"2026-07-29T01:25:29.376201+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.16780v3","created_at":"2026-07-29T01:25:29.376201+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.16780","created_at":"2026-07-29T01:25:29.376201+00:00"},{"alias_kind":"pith_short_12","alias_value":"X3XFOMA3RG5J","created_at":"2026-07-29T01:25:29.376201+00:00"},{"alias_kind":"pith_short_16","alias_value":"X3XFOMA3RG5JE7G5","created_at":"2026-07-29T01:25:29.376201+00:00"},{"alias_kind":"pith_short_8","alias_value":"X3XFOMA3","created_at":"2026-07-29T01:25:29.376201+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.22728","citing_title":"When Confidence Takes the Wrong Path: Diagnosing Retrieval-State Lock-In in RAG","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20041","citing_title":"AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ","json":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ.json","graph_json":"https://pith.science/api/pith-number/X3XFOMA3RG5JE7G57MTWAOSQSQ/graph.json","events_json":"https://pith.science/api/pith-number/X3XFOMA3RG5JE7G57MTWAOSQSQ/events.json","paper":"https://pith.science/paper/X3XFOMA3"},"agent_actions":{"view_html":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ","download_json":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ.json","view_paper":"https://pith.science/paper/X3XFOMA3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.16780&json=true","fetch_graph":"https://pith.science/api/pith-number/X3XFOMA3RG5JE7G57MTWAOSQSQ/graph.json","fetch_events":"https://pith.science/api/pith-number/X3XFOMA3RG5JE7G57MTWAOSQSQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ/action/storage_attestation","attest_author":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ/action/author_attestation","sign_citation":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ/action/citation_signature","submit_replication":"https://pith.science/pith/X3XFOMA3RG5JE7G57MTWAOSQSQ/action/replication_record"}},"created_at":"2026-07-29T01:25:29.376201+00:00","updated_at":"2026-07-29T01:25:29.376201+00:00"}