{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AWXAAJWQZWRJAMZUDV4KDMKTXJ","short_pith_number":"pith:AWXAAJWQ","schema_version":"1.0","canonical_sha256":"05ae0026d0cda29033341d78a1b153ba74f321b94ee1476719b95a123a663ca1","source":{"kind":"arxiv","id":"2403.01432","version":5},"attestation_state":"computed","paper":{"title":"Fine Tuning vs. Retrieval Augmented Generation for Less Popular Knowledge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Evangelos Kanoulas, Faegheh Hasibi, Heydar Soudani","submitted_at":"2024-03-03T08:07:55Z","abstract_excerpt":"Language Models (LMs) memorize a vast amount of factual knowledge, exhibiting strong performance across diverse tasks and domains. However, it has been observed that the performance diminishes when dealing with less-popular or low-frequency concepts and entities, for example in domain specific applications. The two prominent approaches to enhance the performance of LMs on low-frequent topics are: Retrieval Augmented Generation (RAG) and fine-tuning (FT) over synthetic data. This paper explores and evaluates the impact of RAG and FT on customizing LMs in handling low-frequency entities on quest"},"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":"2403.01432","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-03T08:07:55Z","cross_cats_sorted":[],"title_canon_sha256":"00cb1a662fb734a09c5c1640667bcdd83d56e05f7307884a7d7d670fddd679fb","abstract_canon_sha256":"f69fb09e1ff2faa889eceda4acb16a0cbaec1e1ec9694413fd0c8e311cef35f7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:42:23.320751Z","signature_b64":"MCw/hJnJIt2gufpYnMme1+UrBmY43TW/ca2Mao+a02ZOp8cESQKQNuzrv7PkuAVQ+UZociSoB0bV7nuzTFjOBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"05ae0026d0cda29033341d78a1b153ba74f321b94ee1476719b95a123a663ca1","last_reissued_at":"2026-07-05T09:42:23.320256Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:42:23.320256Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fine Tuning vs. Retrieval Augmented Generation for Less Popular Knowledge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Evangelos Kanoulas, Faegheh Hasibi, Heydar Soudani","submitted_at":"2024-03-03T08:07:55Z","abstract_excerpt":"Language Models (LMs) memorize a vast amount of factual knowledge, exhibiting strong performance across diverse tasks and domains. However, it has been observed that the performance diminishes when dealing with less-popular or low-frequency concepts and entities, for example in domain specific applications. The two prominent approaches to enhance the performance of LMs on low-frequent topics are: Retrieval Augmented Generation (RAG) and fine-tuning (FT) over synthetic data. This paper explores and evaluates the impact of RAG and FT on customizing LMs in handling low-frequency entities on quest"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.01432","kind":"arxiv","version":5},"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/2403.01432/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":"2403.01432","created_at":"2026-07-05T09:42:23.320322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.01432v5","created_at":"2026-07-05T09:42:23.320322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.01432","created_at":"2026-07-05T09:42:23.320322+00:00"},{"alias_kind":"pith_short_12","alias_value":"AWXAAJWQZWRJ","created_at":"2026-07-05T09:42:23.320322+00:00"},{"alias_kind":"pith_short_16","alias_value":"AWXAAJWQZWRJAMZU","created_at":"2026-07-05T09:42:23.320322+00:00"},{"alias_kind":"pith_short_8","alias_value":"AWXAAJWQ","created_at":"2026-07-05T09:42:23.320322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.18027","citing_title":"PDF Retrieval Augmented Question Answering","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27415","citing_title":"ChipLingo: A Systematic Training Framework for Large Language Models in EDA","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09533","citing_title":"Assessment of RAG and Fine-Tuning for Industrial Question-Answering-Applications","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ","json":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ.json","graph_json":"https://pith.science/api/pith-number/AWXAAJWQZWRJAMZUDV4KDMKTXJ/graph.json","events_json":"https://pith.science/api/pith-number/AWXAAJWQZWRJAMZUDV4KDMKTXJ/events.json","paper":"https://pith.science/paper/AWXAAJWQ"},"agent_actions":{"view_html":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ","download_json":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ.json","view_paper":"https://pith.science/paper/AWXAAJWQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.01432&json=true","fetch_graph":"https://pith.science/api/pith-number/AWXAAJWQZWRJAMZUDV4KDMKTXJ/graph.json","fetch_events":"https://pith.science/api/pith-number/AWXAAJWQZWRJAMZUDV4KDMKTXJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ/action/storage_attestation","attest_author":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ/action/author_attestation","sign_citation":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ/action/citation_signature","submit_replication":"https://pith.science/pith/AWXAAJWQZWRJAMZUDV4KDMKTXJ/action/replication_record"}},"created_at":"2026-07-05T09:42:23.320322+00:00","updated_at":"2026-07-05T09:42:23.320322+00:00"}