{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PWJ4D3EENKPFBKGEO26U2XMZJH","short_pith_number":"pith:PWJ4D3EE","schema_version":"1.0","canonical_sha256":"7d93c1ec846a9e50a8c476bd4d5d9949c7d72f75eea2778f39a8f05ed82c8dde","source":{"kind":"arxiv","id":"2607.26497","version":1},"attestation_state":"computed","paper":{"title":"Which RAG Paradigm Wins at Scale? A Scaling Study of Retrieval-Augmented Generation Paradigms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benfeng Xu, Huarui Wu, Lei Zhang, Licheng Zhang, Pengyu Wang, Shaohan Wang, Xin Zeng","submitted_at":"2026-07-29T05:46:11Z","abstract_excerpt":"Retrieval-augmented generation (RAG) methods range from lexical and dense retrieval to graph-based indexing and agentic search. They are usually evaluated on different benchmarks at one corpus size, leaving their accuracy-cost scaling unclear. To bridge this gap, we present a controlled corpus-scaling study of these four paradigms. A ladder of 28 strictly nested tiers grows from roughly 1,000 to 512,000 documents while questions and a fixed bedrock of relevant and adversarial documents remain unchanged. Under one reader and judging protocol, we measure official accuracy, construction and query"},"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":"2607.26497","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T05:46:11Z","cross_cats_sorted":[],"title_canon_sha256":"61e4e85d6f5d88d03f9dd32632f85565688c0f83349d7ede45cd85a1b483e3fe","abstract_canon_sha256":"1d533b7b6f942f71f05c945514e550522c79988f0c68a2bba98a5cbdcdfd6a2e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d93c1ec846a9e50a8c476bd4d5d9949c7d72f75eea2778f39a8f05ed82c8dde","last_reissued_at":"2026-07-30T01:20:40.580760Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:20:40.580760Z"},"graph_snapshot":{"paper":{"title":"Which RAG Paradigm Wins at Scale? A Scaling Study of Retrieval-Augmented Generation Paradigms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benfeng Xu, Huarui Wu, Lei Zhang, Licheng Zhang, Pengyu Wang, Shaohan Wang, Xin Zeng","submitted_at":"2026-07-29T05:46:11Z","abstract_excerpt":"Retrieval-augmented generation (RAG) methods range from lexical and dense retrieval to graph-based indexing and agentic search. They are usually evaluated on different benchmarks at one corpus size, leaving their accuracy-cost scaling unclear. To bridge this gap, we present a controlled corpus-scaling study of these four paradigms. A ladder of 28 strictly nested tiers grows from roughly 1,000 to 512,000 documents while questions and a fixed bedrock of relevant and adversarial documents remain unchanged. Under one reader and judging protocol, we measure official accuracy, construction and query"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26497","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/2607.26497/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":"2607.26497","created_at":"2026-07-30T01:20:40.585793+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.26497v1","created_at":"2026-07-30T01:20:40.585793+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26497","created_at":"2026-07-30T01:20:40.585793+00:00"},{"alias_kind":"pith_short_12","alias_value":"PWJ4D3EENKPF","created_at":"2026-07-30T01:20:40.585793+00:00"},{"alias_kind":"pith_short_16","alias_value":"PWJ4D3EENKPFBKGE","created_at":"2026-07-30T01:20:40.585793+00:00"},{"alias_kind":"pith_short_8","alias_value":"PWJ4D3EE","created_at":"2026-07-30T01:20:40.585793+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/PWJ4D3EENKPFBKGEO26U2XMZJH","json":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH.json","graph_json":"https://pith.science/api/pith-number/PWJ4D3EENKPFBKGEO26U2XMZJH/graph.json","events_json":"https://pith.science/api/pith-number/PWJ4D3EENKPFBKGEO26U2XMZJH/events.json","paper":"https://pith.science/paper/PWJ4D3EE"},"agent_actions":{"view_html":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH","download_json":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH.json","view_paper":"https://pith.science/paper/PWJ4D3EE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.26497&json=true","fetch_graph":"https://pith.science/api/pith-number/PWJ4D3EENKPFBKGEO26U2XMZJH/graph.json","fetch_events":"https://pith.science/api/pith-number/PWJ4D3EENKPFBKGEO26U2XMZJH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH/action/storage_attestation","attest_author":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH/action/author_attestation","sign_citation":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH/action/citation_signature","submit_replication":"https://pith.science/pith/PWJ4D3EENKPFBKGEO26U2XMZJH/action/replication_record"}},"created_at":"2026-07-30T01:20:40.585793+00:00","updated_at":"2026-07-30T01:20:40.585793+00:00"}