{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NKKPQQ6GYZBCD4BNFQ3B24GZMS","short_pith_number":"pith:NKKPQQ6G","schema_version":"1.0","canonical_sha256":"6a94f843c6c64221f02d2c361d70d964bff10e298227d3ac7c2816ec7b06ea6b","source":{"kind":"arxiv","id":"2404.06078","version":1},"attestation_state":"computed","paper":{"title":"End-to-end training of Multimodal Model and ranking Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Di Zhang, Erpeng Xue, Guorui Zhou, Han Li, Jinkai Yu, Lu Xu, Na Mou, Shen Jiang, Xiuqi Deng, Xiyao Li, Yang Song, Zhaojie Liu, Zhongyuan Wang","submitted_at":"2024-04-09T07:32:55Z","abstract_excerpt":"Traditional recommender systems heavily rely on ID features, which often encounter challenges related to cold-start and generalization. Modeling pre-extracted content features can mitigate these issues, but is still a suboptimal solution due to the discrepancies between training tasks and model parameters. End-to-end training presents a promising solution for these problems, yet most of the existing works mainly focus on retrieval models, leaving the multimodal techniques under-utilized. In this paper, we propose an industrial multimodal recommendation framework named EM3: End-to-end training "},"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":"2404.06078","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2024-04-09T07:32:55Z","cross_cats_sorted":[],"title_canon_sha256":"a3fde7b752fcf4badf43e540c6b354481a6ba37eb60255be6a09ce59d51a6f4d","abstract_canon_sha256":"4f869115c016c11d305946c548d1d871a5a098cfebb2e6867bf7fa993650b461"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:06:03.397185Z","signature_b64":"DOsGIMOvXuk7wOPFQb4Qi5E3vtqAk7Dk+SpHPY9J53rPddNz5u5OVZ5siPWq6zHkBFQ9yAMimMTh8UJ7EoPBDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a94f843c6c64221f02d2c361d70d964bff10e298227d3ac7c2816ec7b06ea6b","last_reissued_at":"2026-07-05T08:06:03.396655Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:06:03.396655Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"End-to-end training of Multimodal Model and ranking Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Di Zhang, Erpeng Xue, Guorui Zhou, Han Li, Jinkai Yu, Lu Xu, Na Mou, Shen Jiang, Xiuqi Deng, Xiyao Li, Yang Song, Zhaojie Liu, Zhongyuan Wang","submitted_at":"2024-04-09T07:32:55Z","abstract_excerpt":"Traditional recommender systems heavily rely on ID features, which often encounter challenges related to cold-start and generalization. Modeling pre-extracted content features can mitigate these issues, but is still a suboptimal solution due to the discrepancies between training tasks and model parameters. End-to-end training presents a promising solution for these problems, yet most of the existing works mainly focus on retrieval models, leaving the multimodal techniques under-utilized. In this paper, we propose an industrial multimodal recommendation framework named EM3: End-to-end training "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.06078","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/2404.06078/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":"2404.06078","created_at":"2026-07-05T08:06:03.396712+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.06078v1","created_at":"2026-07-05T08:06:03.396712+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.06078","created_at":"2026-07-05T08:06:03.396712+00:00"},{"alias_kind":"pith_short_12","alias_value":"NKKPQQ6GYZBC","created_at":"2026-07-05T08:06:03.396712+00:00"},{"alias_kind":"pith_short_16","alias_value":"NKKPQQ6GYZBCD4BN","created_at":"2026-07-05T08:06:03.396712+00:00"},{"alias_kind":"pith_short_8","alias_value":"NKKPQQ6G","created_at":"2026-07-05T08:06:03.396712+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.21832","citing_title":"FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21832","citing_title":"FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17366","citing_title":"Text-Guided Visual Representation Learning for Robust Multimodal E-Commerce Recommendation","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS","json":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS.json","graph_json":"https://pith.science/api/pith-number/NKKPQQ6GYZBCD4BNFQ3B24GZMS/graph.json","events_json":"https://pith.science/api/pith-number/NKKPQQ6GYZBCD4BNFQ3B24GZMS/events.json","paper":"https://pith.science/paper/NKKPQQ6G"},"agent_actions":{"view_html":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS","download_json":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS.json","view_paper":"https://pith.science/paper/NKKPQQ6G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.06078&json=true","fetch_graph":"https://pith.science/api/pith-number/NKKPQQ6GYZBCD4BNFQ3B24GZMS/graph.json","fetch_events":"https://pith.science/api/pith-number/NKKPQQ6GYZBCD4BNFQ3B24GZMS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS/action/storage_attestation","attest_author":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS/action/author_attestation","sign_citation":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS/action/citation_signature","submit_replication":"https://pith.science/pith/NKKPQQ6GYZBCD4BNFQ3B24GZMS/action/replication_record"}},"created_at":"2026-07-05T08:06:03.396712+00:00","updated_at":"2026-07-05T08:06:03.396712+00:00"}