{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TQIOAYXAGBHCS256OOVOIHLCMO","short_pith_number":"pith:TQIOAYXA","schema_version":"1.0","canonical_sha256":"9c10e062e0304e296bbe73aae41d62639db4d8e7c32d8eabbbd0b6946fdb9025","source":{"kind":"arxiv","id":"2507.18857","version":1},"attestation_state":"computed","paper":{"title":"PrismRAG: Boosting RAG Factuality with Distractor Resilience and Strategized Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aaron Colak, Anuj Kumar, Jiaqi Wang, Kai Sun, Minseok Kim, Mohammad Kachuee, Nirav Shah, Teja Gollapudi, Wen-tau Yih, Xiao Yang, Xin Luna Dong, Yin Huang, Yue Liu","submitted_at":"2025-07-25T00:15:31Z","abstract_excerpt":"Retrieval-augmented generation (RAG) often falls short when retrieved context includes confusing semi-relevant passages, or when answering questions require deep contextual understanding and reasoning. We propose an efficient fine-tuning framework, called PrismRAG, that (i) trains the model with distractor-aware QA pairs mixing gold evidence with subtle distractor passages, and (ii) instills reasoning-centric habits that make the LLM plan, rationalize, and synthesize without relying on extensive human engineered instructions. Evaluated across 12 open-book RAG QA benchmarks spanning diverse app"},"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":"2507.18857","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-25T00:15:31Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a564fcdfbcb6f502d9e72ea44c26aaea0f9eefc5b7202bbbcde590c880d26d25","abstract_canon_sha256":"878e7d7074b4b20e631f7561b7451a83c52ef4a967f6175dd4f50771523a9c9a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:59.608293Z","signature_b64":"fr5RBO3AH3/pij8HqjRq0HcBg2S7IynEpooW0SNkehIXt5liO7t/OwY+PKdYpY1vKSvjdtKhwbqehOgzRSVBCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c10e062e0304e296bbe73aae41d62639db4d8e7c32d8eabbbd0b6946fdb9025","last_reissued_at":"2026-07-05T11:42:59.607724Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:59.607724Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PrismRAG: Boosting RAG Factuality with Distractor Resilience and Strategized Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aaron Colak, Anuj Kumar, Jiaqi Wang, Kai Sun, Minseok Kim, Mohammad Kachuee, Nirav Shah, Teja Gollapudi, Wen-tau Yih, Xiao Yang, Xin Luna Dong, Yin Huang, Yue Liu","submitted_at":"2025-07-25T00:15:31Z","abstract_excerpt":"Retrieval-augmented generation (RAG) often falls short when retrieved context includes confusing semi-relevant passages, or when answering questions require deep contextual understanding and reasoning. We propose an efficient fine-tuning framework, called PrismRAG, that (i) trains the model with distractor-aware QA pairs mixing gold evidence with subtle distractor passages, and (ii) instills reasoning-centric habits that make the LLM plan, rationalize, and synthesize without relying on extensive human engineered instructions. Evaluated across 12 open-book RAG QA benchmarks spanning diverse app"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18857","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/2507.18857/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":"2507.18857","created_at":"2026-07-05T11:42:59.607803+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.18857v1","created_at":"2026-07-05T11:42:59.607803+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18857","created_at":"2026-07-05T11:42:59.607803+00:00"},{"alias_kind":"pith_short_12","alias_value":"TQIOAYXAGBHC","created_at":"2026-07-05T11:42:59.607803+00:00"},{"alias_kind":"pith_short_16","alias_value":"TQIOAYXAGBHCS256","created_at":"2026-07-05T11:42:59.607803+00:00"},{"alias_kind":"pith_short_8","alias_value":"TQIOAYXA","created_at":"2026-07-05T11:42:59.607803+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/TQIOAYXAGBHCS256OOVOIHLCMO","json":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO.json","graph_json":"https://pith.science/api/pith-number/TQIOAYXAGBHCS256OOVOIHLCMO/graph.json","events_json":"https://pith.science/api/pith-number/TQIOAYXAGBHCS256OOVOIHLCMO/events.json","paper":"https://pith.science/paper/TQIOAYXA"},"agent_actions":{"view_html":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO","download_json":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO.json","view_paper":"https://pith.science/paper/TQIOAYXA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.18857&json=true","fetch_graph":"https://pith.science/api/pith-number/TQIOAYXAGBHCS256OOVOIHLCMO/graph.json","fetch_events":"https://pith.science/api/pith-number/TQIOAYXAGBHCS256OOVOIHLCMO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO/action/storage_attestation","attest_author":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO/action/author_attestation","sign_citation":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO/action/citation_signature","submit_replication":"https://pith.science/pith/TQIOAYXAGBHCS256OOVOIHLCMO/action/replication_record"}},"created_at":"2026-07-05T11:42:59.607803+00:00","updated_at":"2026-07-05T11:42:59.607803+00:00"}