{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GA6O5EO4L6AHBLRV52WWGRM657","short_pith_number":"pith:GA6O5EO4","schema_version":"1.0","canonical_sha256":"303cee91dc5f8070ae35eead63459eefecd5cb6c1b8a14d812b02d7ec13082a7","source":{"kind":"arxiv","id":"2510.13217","version":2},"attestation_state":"computed","paper":{"title":"LLM-guided Hierarchical Search for End-to-end Reasoning Intensive Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Cho-Jui Hsieh, Inderjit S. Dhillon, Ngot Bui, Nilesh Gupta, Wei-Cheng Chang","submitted_at":"2025-10-15T07:05:17Z","abstract_excerpt":"Search systems are increasingly used for reasoning-intensive queries, where what makes a document relevant requires understanding or reasoning over the query-document relation rather than relying on surface vocabulary or topical similarity. The standard recipe - a cheap embedding-based retriever followed by an LLM verifier - works only when the embedding model places the right documents in its top-k, an assumption that recent reasoning-intensive IR benchmarks show often fails to hold even for SOTA embedding models. Recent query-side fixes such as query rewriting and agentic loops keep the LLM "},"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":"2510.13217","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-10-15T07:05:17Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5284b8c775c03daab8b6e4f88face7e768b8a79069e7b265c2c29b0a91f18334","abstract_canon_sha256":"74cde650998304ec5a49d59cff925bc43517dae162c268f0798ff7968d1a68be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T00:04:24.290047Z","signature_b64":"IbzWdHwzMixARc96UFUduB1KNaWIgkvaQCakiSzBj52CWC85SdTLXnXSOkYrW1T7Iox5ioWhnetWLxziQe6SDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"303cee91dc5f8070ae35eead63459eefecd5cb6c1b8a14d812b02d7ec13082a7","last_reissued_at":"2026-05-27T00:04:24.289355Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T00:04:24.289355Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM-guided Hierarchical Search for End-to-end Reasoning Intensive Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Cho-Jui Hsieh, Inderjit S. Dhillon, Ngot Bui, Nilesh Gupta, Wei-Cheng Chang","submitted_at":"2025-10-15T07:05:17Z","abstract_excerpt":"Search systems are increasingly used for reasoning-intensive queries, where what makes a document relevant requires understanding or reasoning over the query-document relation rather than relying on surface vocabulary or topical similarity. The standard recipe - a cheap embedding-based retriever followed by an LLM verifier - works only when the embedding model places the right documents in its top-k, an assumption that recent reasoning-intensive IR benchmarks show often fails to hold even for SOTA embedding models. Recent query-side fixes such as query rewriting and agentic loops keep the LLM "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.13217","kind":"arxiv","version":2},"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/2510.13217/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":"2510.13217","created_at":"2026-05-27T00:04:24.289446+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.13217v2","created_at":"2026-05-27T00:04:24.289446+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.13217","created_at":"2026-05-27T00:04:24.289446+00:00"},{"alias_kind":"pith_short_12","alias_value":"GA6O5EO4L6AH","created_at":"2026-05-27T00:04:24.289446+00:00"},{"alias_kind":"pith_short_16","alias_value":"GA6O5EO4L6AHBLRV","created_at":"2026-05-27T00:04:24.289446+00:00"},{"alias_kind":"pith_short_8","alias_value":"GA6O5EO4","created_at":"2026-05-27T00:04:24.289446+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2605.05657","citing_title":"Retrieval-Conditioned Topology Selection with Provable Budget Conservation for Multi-Agent Code Generation","ref_index":64,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05538","citing_title":"AgenticRAG: Agentic Retrieval for Enterprise Knowledge Bases","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00529","citing_title":"Hierarchical Abstract Tree for Cross-Document Retrieval-Augmented Generation","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00063","citing_title":"A Survey of Reasoning-Intensive Retrieval: Progress and Challenges","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657","json":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657.json","graph_json":"https://pith.science/api/pith-number/GA6O5EO4L6AHBLRV52WWGRM657/graph.json","events_json":"https://pith.science/api/pith-number/GA6O5EO4L6AHBLRV52WWGRM657/events.json","paper":"https://pith.science/paper/GA6O5EO4"},"agent_actions":{"view_html":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657","download_json":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657.json","view_paper":"https://pith.science/paper/GA6O5EO4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.13217&json=true","fetch_graph":"https://pith.science/api/pith-number/GA6O5EO4L6AHBLRV52WWGRM657/graph.json","fetch_events":"https://pith.science/api/pith-number/GA6O5EO4L6AHBLRV52WWGRM657/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657/action/storage_attestation","attest_author":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657/action/author_attestation","sign_citation":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657/action/citation_signature","submit_replication":"https://pith.science/pith/GA6O5EO4L6AHBLRV52WWGRM657/action/replication_record"}},"created_at":"2026-05-27T00:04:24.289446+00:00","updated_at":"2026-05-27T00:04:24.289446+00:00"}