{"paper":{"title":"UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"UxSID uses semantic IDs and dual-level attention to model ultra-long user sequences with target-aware preferences.","cross_cats":["cs.IR","cs.LG"],"primary_cat":"cs.AI","authors_text":"Han Li, Hongwei Zhang, Huanjie Wang, Jiangxia Cao, Jing Yao, Junfeng Shu, Liwei Guan, Qiqiang Zhong, Yiyang Lv, Yiyu Wang, Zhaojie Liu","submitted_at":"2026-05-09T16:26:48Z","abstract_excerpt":"Modeling ultra-long user sequences involves a difficult trade-off between efficiency and effectiveness. While current paradigms rely on either item-specific search or item-agnostic compression, we propose UxSID, a framework exploring a third path: semantic-group shared interest memory. By utilizing Semantic IDs (SIDs) and a dual-level attention strategy, UxSID captures target-aware preferences without the heavy cost of item-specific models. This end-to-end architecture balances computational parsimony with semantic awareness, achieving state-of-the-art performance and a 0.337% revenue lift in "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By utilizing Semantic IDs (SIDs) and a dual-level attention strategy, UxSID captures target-aware preferences without the heavy cost of item-specific models... achieving state-of-the-art performance and a 0.337% revenue lift in large-scale advertising A/B test.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That semantic grouping via SIDs plus dual-level attention actually preserves target-aware preferences better than item-specific search or item-agnostic compression without introducing new biases or losing critical signals in the ultra-long sequences.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"UxSID introduces semantic-group shared interest memory with Semantic IDs and dual-level attention to model ultra-long user sequences, claiming state-of-the-art results and a 0.337% revenue lift in advertising A/B tests.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"UxSID uses semantic IDs and dual-level attention to model ultra-long user sequences with target-aware preferences.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"045795a06f7a7523f31cc5f9c7b581150a50a09ae7fc16bf6cd76ed707111bd3"},"source":{"id":"2605.09040","kind":"arxiv","version":3},"verdict":{"id":"ed7a99cc-489a-44b1-a21f-8d4753148338","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:01:42.096991Z","strongest_claim":"By utilizing Semantic IDs (SIDs) and a dual-level attention strategy, UxSID captures target-aware preferences without the heavy cost of item-specific models... achieving state-of-the-art performance and a 0.337% revenue lift in large-scale advertising A/B test.","one_line_summary":"UxSID introduces semantic-group shared interest memory with Semantic IDs and dual-level attention to model ultra-long user sequences, claiming state-of-the-art results and a 0.337% revenue lift in advertising A/B tests.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That semantic grouping via SIDs plus dual-level attention actually preserves target-aware preferences better than item-specific search or item-agnostic compression without introducing new biases or losing critical signals in the ultra-long sequences.","pith_extraction_headline":"UxSID uses semantic IDs and dual-level attention to model ultra-long user sequences with target-aware preferences."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.09040/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T20:38:45.705513Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T13:31:19.238511Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T10:32:41.531799Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"8eb21a7a0a7a4d9a785989a660fe9acf46bf1afdc980c1d6294b0f2e1f2ab975"},"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"}