{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TUXKUPYYSDEUVMCTHB2FZK7JE7","short_pith_number":"pith:TUXKUPYY","schema_version":"1.0","canonical_sha256":"9d2eaa3f1890c94ab05338745cabe927e3e2fe99f785073622c1c6de6406ff5a","source":{"kind":"arxiv","id":"2504.05181","version":2},"attestation_state":"computed","paper":{"title":"Lightweight and Direct Document Relevance Optimization for Generative Information Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Kidist Amde Mekonnen, Maarten de Rijke, Yubao Tang","submitted_at":"2025-04-07T15:27:37Z","abstract_excerpt":"Generative information retrieval (GenIR) is a promising neural retrieval paradigm that formulates document retrieval as a document identifier (docid) generation task, allowing for end-to-end optimization toward a unified global retrieval objective. However, existing GenIR models suffer from token-level misalignment, where models trained to predict the next token often fail to capture document-level relevance effectively. While reinforcement learning-based methods, such as reinforcement learning from relevance feedback (RLRF), aim to address this misalignment through reward modeling, they intro"},"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":"2504.05181","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-07T15:27:37Z","cross_cats_sorted":["cs.AI","cs.DL","cs.LG"],"title_canon_sha256":"6d54846e22816249063a90d40367fa372355701ce1ec2a0af686c80bad193fac","abstract_canon_sha256":"892956df68aa2f6d3c8a4b426d9dc331e15d3c962b376cb24704588c27899360"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:12.016095Z","signature_b64":"YGKUUCPMck7cxcfFmi1+NgSsjVURSxPW7+BO/v1fmMVdpZugsSI8jbTAPzfP5sjNyEM4Txhf0J2qxsVb7gJoDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d2eaa3f1890c94ab05338745cabe927e3e2fe99f785073622c1c6de6406ff5a","last_reissued_at":"2026-05-26T01:03:12.015245Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:12.015245Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lightweight and Direct Document Relevance Optimization for Generative Information Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Kidist Amde Mekonnen, Maarten de Rijke, Yubao Tang","submitted_at":"2025-04-07T15:27:37Z","abstract_excerpt":"Generative information retrieval (GenIR) is a promising neural retrieval paradigm that formulates document retrieval as a document identifier (docid) generation task, allowing for end-to-end optimization toward a unified global retrieval objective. However, existing GenIR models suffer from token-level misalignment, where models trained to predict the next token often fail to capture document-level relevance effectively. While reinforcement learning-based methods, such as reinforcement learning from relevance feedback (RLRF), aim to address this misalignment through reward modeling, they intro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.05181","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/2504.05181/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":"2504.05181","created_at":"2026-05-26T01:03:12.015363+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.05181v2","created_at":"2026-05-26T01:03:12.015363+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.05181","created_at":"2026-05-26T01:03:12.015363+00:00"},{"alias_kind":"pith_short_12","alias_value":"TUXKUPYYSDEU","created_at":"2026-05-26T01:03:12.015363+00:00"},{"alias_kind":"pith_short_16","alias_value":"TUXKUPYYSDEUVMCT","created_at":"2026-05-26T01:03:12.015363+00:00"},{"alias_kind":"pith_short_8","alias_value":"TUXKUPYY","created_at":"2026-05-26T01:03:12.015363+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.05764","citing_title":"Generative Retrieval Overcomes Limitations of Dense Retrieval but Struggles with Identifier Ambiguity","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7","json":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7.json","graph_json":"https://pith.science/api/pith-number/TUXKUPYYSDEUVMCTHB2FZK7JE7/graph.json","events_json":"https://pith.science/api/pith-number/TUXKUPYYSDEUVMCTHB2FZK7JE7/events.json","paper":"https://pith.science/paper/TUXKUPYY"},"agent_actions":{"view_html":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7","download_json":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7.json","view_paper":"https://pith.science/paper/TUXKUPYY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.05181&json=true","fetch_graph":"https://pith.science/api/pith-number/TUXKUPYYSDEUVMCTHB2FZK7JE7/graph.json","fetch_events":"https://pith.science/api/pith-number/TUXKUPYYSDEUVMCTHB2FZK7JE7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7/action/storage_attestation","attest_author":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7/action/author_attestation","sign_citation":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7/action/citation_signature","submit_replication":"https://pith.science/pith/TUXKUPYYSDEUVMCTHB2FZK7JE7/action/replication_record"}},"created_at":"2026-05-26T01:03:12.015363+00:00","updated_at":"2026-05-26T01:03:12.015363+00:00"}