{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RRU5E6UH4GX5ZIH5F26MCROPS3","short_pith_number":"pith:RRU5E6UH","schema_version":"1.0","canonical_sha256":"8c69d27a87e1afdca0fd2ebcc145cf96fc7d46669cf69b51e0d33e3b2fb669a9","source":{"kind":"arxiv","id":"2310.03184","version":2},"attestation_state":"computed","paper":{"title":"Retrieval-augmented Generation to Improve Math Question-Answering: Trade-offs Between Groundedness and Human Preference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.HC"],"primary_cat":"cs.CL","authors_text":"Anoushka Gade, Chenglu Li, Millie-Ellen Postle, Owen Henkel, Wangda Zhu, Wanli Xing, Zachary Levonian","submitted_at":"2023-10-04T22:09:28Z","abstract_excerpt":"For middle-school math students, interactive question-answering (QA) with tutors is an effective way to learn. The flexibility and emergent capabilities of generative large language models (LLMs) has led to a surge of interest in automating portions of the tutoring process - including interactive QA to support conceptual discussion of mathematical concepts. However, LLM responses to math questions can be incorrect or mismatched to the educational context - such as being misaligned with a school's curriculum. One potential solution is retrieval-augmented generation (RAG), which involves incorpo"},"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":"2310.03184","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-04T22:09:28Z","cross_cats_sorted":["cs.HC"],"title_canon_sha256":"123b17cd4109dd8a6941bbdce298a88c9293dd03f97a645474a54b0e07a6c3e5","abstract_canon_sha256":"1d5fd30205c79aee1a1f3fd95b566e4e70e65936f040ef2f7b3224b57ba388bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:11:36.659231Z","signature_b64":"rlvXTe54klDmmwpUuVcST6rk2SSAT0Vvc6FwQVobm34Od2yyfxvu0aQOwtRVGx5Wmfp88/goqcCF9bzVBkwrBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8c69d27a87e1afdca0fd2ebcc145cf96fc7d46669cf69b51e0d33e3b2fb669a9","last_reissued_at":"2026-07-05T07:11:36.658677Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:11:36.658677Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Retrieval-augmented Generation to Improve Math Question-Answering: Trade-offs Between Groundedness and Human Preference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.HC"],"primary_cat":"cs.CL","authors_text":"Anoushka Gade, Chenglu Li, Millie-Ellen Postle, Owen Henkel, Wangda Zhu, Wanli Xing, Zachary Levonian","submitted_at":"2023-10-04T22:09:28Z","abstract_excerpt":"For middle-school math students, interactive question-answering (QA) with tutors is an effective way to learn. The flexibility and emergent capabilities of generative large language models (LLMs) has led to a surge of interest in automating portions of the tutoring process - including interactive QA to support conceptual discussion of mathematical concepts. However, LLM responses to math questions can be incorrect or mismatched to the educational context - such as being misaligned with a school's curriculum. One potential solution is retrieval-augmented generation (RAG), which involves incorpo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.03184","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/2310.03184/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":"2310.03184","created_at":"2026-07-05T07:11:36.658745+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.03184v2","created_at":"2026-07-05T07:11:36.658745+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.03184","created_at":"2026-07-05T07:11:36.658745+00:00"},{"alias_kind":"pith_short_12","alias_value":"RRU5E6UH4GX5","created_at":"2026-07-05T07:11:36.658745+00:00"},{"alias_kind":"pith_short_16","alias_value":"RRU5E6UH4GX5ZIH5","created_at":"2026-07-05T07:11:36.658745+00:00"},{"alias_kind":"pith_short_8","alias_value":"RRU5E6UH","created_at":"2026-07-05T07:11:36.658745+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.12610","citing_title":"Transforming External Knowledge into Triplets for Enhanced Retrieval in RAG of LLMs","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17843","citing_title":"Learning from AVA: Early Lessons from a Curated and Trustworthy Generative AI for Policy and Development Research","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15270","citing_title":"Enhancing Large Language Models with Retrieval Augmented Generation for Software Testing and Inspection Automation","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3","json":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3.json","graph_json":"https://pith.science/api/pith-number/RRU5E6UH4GX5ZIH5F26MCROPS3/graph.json","events_json":"https://pith.science/api/pith-number/RRU5E6UH4GX5ZIH5F26MCROPS3/events.json","paper":"https://pith.science/paper/RRU5E6UH"},"agent_actions":{"view_html":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3","download_json":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3.json","view_paper":"https://pith.science/paper/RRU5E6UH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.03184&json=true","fetch_graph":"https://pith.science/api/pith-number/RRU5E6UH4GX5ZIH5F26MCROPS3/graph.json","fetch_events":"https://pith.science/api/pith-number/RRU5E6UH4GX5ZIH5F26MCROPS3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3/action/storage_attestation","attest_author":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3/action/author_attestation","sign_citation":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3/action/citation_signature","submit_replication":"https://pith.science/pith/RRU5E6UH4GX5ZIH5F26MCROPS3/action/replication_record"}},"created_at":"2026-07-05T07:11:36.658745+00:00","updated_at":"2026-07-05T07:11:36.658745+00:00"}