{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6J46SWNWPM73DNBNZNOWTZZICG","short_pith_number":"pith:6J46SWNW","schema_version":"1.0","canonical_sha256":"f279e959b67b3fb1b42dcb5d69e72811a3c0e66bdad611825d7ff25e4eaace0b","source":{"kind":"arxiv","id":"2607.23038","version":1},"attestation_state":"computed","paper":{"title":"EGR: Embedding-Native Generative Retrieval with a Shared LLM","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Congfei Zhang, Hsiang-wei Chao, Jinchao Li, Jingxiao Ma, Jitin Krishnan, Siman Wang, Tong Zhao, Vincent Zhang, Wenfeng Zhuo, Xiao Bai, XiaoDong Liu, Yajun Wang, Yunzhi Zhou, Yu Zhang, Zhe Liu, Zichu Li","submitted_at":"2026-07-25T04:45:04Z","abstract_excerpt":"Generative retrieval is increasingly popular in large-scale recommendation and advertising systems, yet current methods introduce practical complications. Semantic-ID methods rely on quantization, mutable identifier vocabularies, and token-to-item grounding; embedding-based pipelines train the item encoder separately from the query generator, which limits user-item alignment. We propose EGR, an Embedding-Native Generative Retrieval framework for recommendation and advertising. EGR uses a single shared LLM to learn item representations from item metadata and user representations from interactio"},"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":"2607.23038","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-07-25T04:45:04Z","cross_cats_sorted":[],"title_canon_sha256":"97da0737f85f300c0db4a014613573272fe69615f80439f5f91a2e042cf2375b","abstract_canon_sha256":"073b139011c73776aac4309e5d005ea566addad3da9729d0872bd28d68a71e15"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T00:22:08.937401Z","signature_b64":"vZK60qPWaOcfSakugU3pA41VbNUHo9k/Z26NWBAf9Y4Uppj6FqqIbjSj5IXd6pv+oOK97/cDmMiwrjx9huX1Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f279e959b67b3fb1b42dcb5d69e72811a3c0e66bdad611825d7ff25e4eaace0b","last_reissued_at":"2026-07-28T00:22:08.936592Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T00:22:08.936592Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EGR: Embedding-Native Generative Retrieval with a Shared LLM","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Congfei Zhang, Hsiang-wei Chao, Jinchao Li, Jingxiao Ma, Jitin Krishnan, Siman Wang, Tong Zhao, Vincent Zhang, Wenfeng Zhuo, Xiao Bai, XiaoDong Liu, Yajun Wang, Yunzhi Zhou, Yu Zhang, Zhe Liu, Zichu Li","submitted_at":"2026-07-25T04:45:04Z","abstract_excerpt":"Generative retrieval is increasingly popular in large-scale recommendation and advertising systems, yet current methods introduce practical complications. Semantic-ID methods rely on quantization, mutable identifier vocabularies, and token-to-item grounding; embedding-based pipelines train the item encoder separately from the query generator, which limits user-item alignment. We propose EGR, an Embedding-Native Generative Retrieval framework for recommendation and advertising. EGR uses a single shared LLM to learn item representations from item metadata and user representations from interactio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.23038","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/2607.23038/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":"2607.23038","created_at":"2026-07-28T00:22:08.937001+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.23038v1","created_at":"2026-07-28T00:22:08.937001+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.23038","created_at":"2026-07-28T00:22:08.937001+00:00"},{"alias_kind":"pith_short_12","alias_value":"6J46SWNWPM73","created_at":"2026-07-28T00:22:08.937001+00:00"},{"alias_kind":"pith_short_16","alias_value":"6J46SWNWPM73DNBN","created_at":"2026-07-28T00:22:08.937001+00:00"},{"alias_kind":"pith_short_8","alias_value":"6J46SWNW","created_at":"2026-07-28T00:22:08.937001+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/6J46SWNWPM73DNBNZNOWTZZICG","json":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG.json","graph_json":"https://pith.science/api/pith-number/6J46SWNWPM73DNBNZNOWTZZICG/graph.json","events_json":"https://pith.science/api/pith-number/6J46SWNWPM73DNBNZNOWTZZICG/events.json","paper":"https://pith.science/paper/6J46SWNW"},"agent_actions":{"view_html":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG","download_json":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG.json","view_paper":"https://pith.science/paper/6J46SWNW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.23038&json=true","fetch_graph":"https://pith.science/api/pith-number/6J46SWNWPM73DNBNZNOWTZZICG/graph.json","fetch_events":"https://pith.science/api/pith-number/6J46SWNWPM73DNBNZNOWTZZICG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG/action/storage_attestation","attest_author":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG/action/author_attestation","sign_citation":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG/action/citation_signature","submit_replication":"https://pith.science/pith/6J46SWNWPM73DNBNZNOWTZZICG/action/replication_record"}},"created_at":"2026-07-28T00:22:08.937001+00:00","updated_at":"2026-07-28T00:22:08.937001+00:00"}