{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SDBGDC24GPIIFXUJFQFNLCW5KT","short_pith_number":"pith:SDBGDC24","schema_version":"1.0","canonical_sha256":"90c2618b5c33d082de892c0ad58add54f2b98f0afbb3fa1a837a219b0d8d7dc2","source":{"kind":"arxiv","id":"2410.15531","version":1},"attestation_state":"computed","paper":{"title":"Do RAG Systems Cover What Matters? Evaluating and Optimizing Responses with Sub-Question Coverage","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Chien-Sheng Wu, Kaige Xie, Philippe Laban, Prafulla Kumar Choubey","submitted_at":"2024-10-20T22:59:34Z","abstract_excerpt":"Evaluating retrieval-augmented generation (RAG) systems remains challenging, particularly for open-ended questions that lack definitive answers and require coverage of multiple sub-topics. In this paper, we introduce a novel evaluation framework based on sub-question coverage, which measures how well a RAG system addresses different facets of a question. We propose decomposing questions into sub-questions and classifying them into three types -- core, background, and follow-up -- to reflect their roles and importance. Using this categorization, we introduce a fine-grained evaluation protocol t"},"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":"2410.15531","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-20T22:59:34Z","cross_cats_sorted":[],"title_canon_sha256":"6b90ff86cd176ee7f5d01cfa2f8214a66fa8cab46bfd3beb8b38e7536cc185a1","abstract_canon_sha256":"f6dac5c3532b4bf437dabf5fe0816dc1576f57e14192919f2ee308f5b81dcb37"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:23:20.993038Z","signature_b64":"lJ9qR7J0TvkD0hKV8GxAMulK3SNwi+H6u+5prgK6k6JSc+A/261UF2wPr29XtfvMt6r4F8Sa+WfjCShbQ/3kAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90c2618b5c33d082de892c0ad58add54f2b98f0afbb3fa1a837a219b0d8d7dc2","last_reissued_at":"2026-07-05T09:23:20.992576Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:23:20.992576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do RAG Systems Cover What Matters? Evaluating and Optimizing Responses with Sub-Question Coverage","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Chien-Sheng Wu, Kaige Xie, Philippe Laban, Prafulla Kumar Choubey","submitted_at":"2024-10-20T22:59:34Z","abstract_excerpt":"Evaluating retrieval-augmented generation (RAG) systems remains challenging, particularly for open-ended questions that lack definitive answers and require coverage of multiple sub-topics. In this paper, we introduce a novel evaluation framework based on sub-question coverage, which measures how well a RAG system addresses different facets of a question. We propose decomposing questions into sub-questions and classifying them into three types -- core, background, and follow-up -- to reflect their roles and importance. Using this categorization, we introduce a fine-grained evaluation protocol t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.15531","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/2410.15531/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":"2410.15531","created_at":"2026-07-05T09:23:20.992645+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.15531v1","created_at":"2026-07-05T09:23:20.992645+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.15531","created_at":"2026-07-05T09:23:20.992645+00:00"},{"alias_kind":"pith_short_12","alias_value":"SDBGDC24GPII","created_at":"2026-07-05T09:23:20.992645+00:00"},{"alias_kind":"pith_short_16","alias_value":"SDBGDC24GPIIFXUJ","created_at":"2026-07-05T09:23:20.992645+00:00"},{"alias_kind":"pith_short_8","alias_value":"SDBGDC24","created_at":"2026-07-05T09:23:20.992645+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.20119","citing_title":"Can LLMs Be Trusted for Evaluating RAG Systems? A Survey of Methods and Datasets","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT","json":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT.json","graph_json":"https://pith.science/api/pith-number/SDBGDC24GPIIFXUJFQFNLCW5KT/graph.json","events_json":"https://pith.science/api/pith-number/SDBGDC24GPIIFXUJFQFNLCW5KT/events.json","paper":"https://pith.science/paper/SDBGDC24"},"agent_actions":{"view_html":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT","download_json":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT.json","view_paper":"https://pith.science/paper/SDBGDC24","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.15531&json=true","fetch_graph":"https://pith.science/api/pith-number/SDBGDC24GPIIFXUJFQFNLCW5KT/graph.json","fetch_events":"https://pith.science/api/pith-number/SDBGDC24GPIIFXUJFQFNLCW5KT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT/action/storage_attestation","attest_author":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT/action/author_attestation","sign_citation":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT/action/citation_signature","submit_replication":"https://pith.science/pith/SDBGDC24GPIIFXUJFQFNLCW5KT/action/replication_record"}},"created_at":"2026-07-05T09:23:20.992645+00:00","updated_at":"2026-07-05T09:23:20.992645+00:00"}