{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NKDUKDBYJLR6WYISWTTXMXBHVI","short_pith_number":"pith:NKDUKDBY","schema_version":"1.0","canonical_sha256":"6a87450c384ae3eb6112b4e7765c27aa28356ca23f0c38f7f76c4b74eda65fc3","source":{"kind":"arxiv","id":"2412.10571","version":3},"attestation_state":"computed","paper":{"title":"Evidence Contextualization and Counterfactual Attribution for Conversational QA over Heterogeneous Data with RAG Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Andreas Foltyn, Chris Hinze, Fabian Kuech, Joel Schlotthauer, Luzian Hahn, Rishiraj Saha Roy","submitted_at":"2024-12-13T21:28:17Z","abstract_excerpt":"Retrieval Augmented Generation (RAG) works as a backbone for interacting with an enterprise's own data via Conversational Question Answering (ConvQA). In a RAG system, a retriever fetches passages from a collection in response to a question, which are then included in the prompt of a large language model (LLM) for generating a natural language (NL) answer. However, several RAG systems today suffer from two shortcomings: (i) retrieved passages usually contain their raw text and lack appropriate document context, negatively impacting both retrieval and answering quality; and (ii) attribution str"},"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":"2412.10571","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-13T21:28:17Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"635531ac190dc00c64dd80eaf449f60ae054e89fb9f7ded17114f49570670268","abstract_canon_sha256":"25d631250c1b743ef022be8ab058c069069b8f848b642c63dbdd1b0a90c7c16a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:53:08.814762Z","signature_b64":"BNXRwlPvObFEiNZgv9e/aPUBtCfXUrz1X9qbQqOlqV5rqQJDXXfdWtog9csDb9j8zAgyAeDCxtTZWbG7ktsLBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a87450c384ae3eb6112b4e7765c27aa28356ca23f0c38f7f76c4b74eda65fc3","last_reissued_at":"2026-07-05T09:53:08.814301Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:53:08.814301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evidence Contextualization and Counterfactual Attribution for Conversational QA over Heterogeneous Data with RAG Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Andreas Foltyn, Chris Hinze, Fabian Kuech, Joel Schlotthauer, Luzian Hahn, Rishiraj Saha Roy","submitted_at":"2024-12-13T21:28:17Z","abstract_excerpt":"Retrieval Augmented Generation (RAG) works as a backbone for interacting with an enterprise's own data via Conversational Question Answering (ConvQA). In a RAG system, a retriever fetches passages from a collection in response to a question, which are then included in the prompt of a large language model (LLM) for generating a natural language (NL) answer. However, several RAG systems today suffer from two shortcomings: (i) retrieved passages usually contain their raw text and lack appropriate document context, negatively impacting both retrieval and answering quality; and (ii) attribution str"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.10571","kind":"arxiv","version":3},"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/2412.10571/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":"2412.10571","created_at":"2026-07-05T09:53:08.814365+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.10571v3","created_at":"2026-07-05T09:53:08.814365+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.10571","created_at":"2026-07-05T09:53:08.814365+00:00"},{"alias_kind":"pith_short_12","alias_value":"NKDUKDBYJLR6","created_at":"2026-07-05T09:53:08.814365+00:00"},{"alias_kind":"pith_short_16","alias_value":"NKDUKDBYJLR6WYIS","created_at":"2026-07-05T09:53:08.814365+00:00"},{"alias_kind":"pith_short_8","alias_value":"NKDUKDBY","created_at":"2026-07-05T09:53:08.814365+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":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI","json":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI.json","graph_json":"https://pith.science/api/pith-number/NKDUKDBYJLR6WYISWTTXMXBHVI/graph.json","events_json":"https://pith.science/api/pith-number/NKDUKDBYJLR6WYISWTTXMXBHVI/events.json","paper":"https://pith.science/paper/NKDUKDBY"},"agent_actions":{"view_html":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI","download_json":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI.json","view_paper":"https://pith.science/paper/NKDUKDBY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.10571&json=true","fetch_graph":"https://pith.science/api/pith-number/NKDUKDBYJLR6WYISWTTXMXBHVI/graph.json","fetch_events":"https://pith.science/api/pith-number/NKDUKDBYJLR6WYISWTTXMXBHVI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI/action/storage_attestation","attest_author":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI/action/author_attestation","sign_citation":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI/action/citation_signature","submit_replication":"https://pith.science/pith/NKDUKDBYJLR6WYISWTTXMXBHVI/action/replication_record"}},"created_at":"2026-07-05T09:53:08.814365+00:00","updated_at":"2026-07-05T09:53:08.814365+00:00"}