{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LAGCOGH4FAEX4SI5PK2YNF3I24","short_pith_number":"pith:LAGCOGH4","schema_version":"1.0","canonical_sha256":"580c2718fc28097e491d7ab5869768d71867755becfc2b311e2bef61aa53ac39","source":{"kind":"arxiv","id":"2606.29377","version":1},"attestation_state":"computed","paper":{"title":"Diagnosing and Repairing Factual Errors in RAG under Budget Constraints","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ali Dehghantanha, Fattane Zarrinkalam, Havva Alizadeh Noughabi, Soroush Hashemifar","submitted_at":"2026-06-28T13:01:35Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) improves the factuality of large language models by grounding responses in external evidence, yet real-world deployments remain fragile. Failures often stem from missing or weakly relevant evidence, as well as from generation that does not faithfully reflect the retrieved context. Many existing approaches rely on fine-tuning, privileged access to internal model signals, or resource-insensitive escalation strategies, which limits their practicality in black-box and budget-constrained settings. We propose D2R-RAG (Diagnose-to-Repair RAG), a model-agnostic and"},"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":"2606.29377","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-28T13:01:35Z","cross_cats_sorted":[],"title_canon_sha256":"960934cd2796a242ff6cc74338fe9c0e08791dcc750224ac60e82b07e4147328","abstract_canon_sha256":"df4040b71b35174fb2fa8fd8ffee4cac589ea1b12e89d4d670d3258543cf83a6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:18:03.501231Z","signature_b64":"yA8qvMUxDJHrarAJ6by/OBbWksq+lhnpdvkQqdOQBQc0MOEmc4ZUbPMmOXU11Eu7eSdzh17rr7FqVUcxReO1DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"580c2718fc28097e491d7ab5869768d71867755becfc2b311e2bef61aa53ac39","last_reissued_at":"2026-06-30T01:18:03.500782Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:18:03.500782Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diagnosing and Repairing Factual Errors in RAG under Budget Constraints","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ali Dehghantanha, Fattane Zarrinkalam, Havva Alizadeh Noughabi, Soroush Hashemifar","submitted_at":"2026-06-28T13:01:35Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) improves the factuality of large language models by grounding responses in external evidence, yet real-world deployments remain fragile. Failures often stem from missing or weakly relevant evidence, as well as from generation that does not faithfully reflect the retrieved context. Many existing approaches rely on fine-tuning, privileged access to internal model signals, or resource-insensitive escalation strategies, which limits their practicality in black-box and budget-constrained settings. We propose D2R-RAG (Diagnose-to-Repair RAG), a model-agnostic and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29377","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/2606.29377/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":"2606.29377","created_at":"2026-06-30T01:18:03.500844+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.29377v1","created_at":"2026-06-30T01:18:03.500844+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29377","created_at":"2026-06-30T01:18:03.500844+00:00"},{"alias_kind":"pith_short_12","alias_value":"LAGCOGH4FAEX","created_at":"2026-06-30T01:18:03.500844+00:00"},{"alias_kind":"pith_short_16","alias_value":"LAGCOGH4FAEX4SI5","created_at":"2026-06-30T01:18:03.500844+00:00"},{"alias_kind":"pith_short_8","alias_value":"LAGCOGH4","created_at":"2026-06-30T01:18:03.500844+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.08944","citing_title":"What Would Fix This RAG Failure? Auditing Counterfactual Response with Paired Evidence Interventions","ref_index":2026,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24","json":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24.json","graph_json":"https://pith.science/api/pith-number/LAGCOGH4FAEX4SI5PK2YNF3I24/graph.json","events_json":"https://pith.science/api/pith-number/LAGCOGH4FAEX4SI5PK2YNF3I24/events.json","paper":"https://pith.science/paper/LAGCOGH4"},"agent_actions":{"view_html":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24","download_json":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24.json","view_paper":"https://pith.science/paper/LAGCOGH4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.29377&json=true","fetch_graph":"https://pith.science/api/pith-number/LAGCOGH4FAEX4SI5PK2YNF3I24/graph.json","fetch_events":"https://pith.science/api/pith-number/LAGCOGH4FAEX4SI5PK2YNF3I24/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24/action/storage_attestation","attest_author":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24/action/author_attestation","sign_citation":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24/action/citation_signature","submit_replication":"https://pith.science/pith/LAGCOGH4FAEX4SI5PK2YNF3I24/action/replication_record"}},"created_at":"2026-06-30T01:18:03.500844+00:00","updated_at":"2026-06-30T01:18:03.500844+00:00"}