{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XAUXTUJ566LCIT5BFOBZZZKSME","short_pith_number":"pith:XAUXTUJ5","schema_version":"1.0","canonical_sha256":"b82979d13df796244fa12b839ce5526117bd0f22b4857bc05fd32462c67b1958","source":{"kind":"arxiv","id":"2607.03948","version":1},"attestation_state":"computed","paper":{"title":"Online Linear Programming for Multi-Objective Routing in LLM Serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"cs.AI","authors_text":"Yinyu Ye, Zijie Zhou, Zixi Chen","submitted_at":"2026-07-04T16:48:37Z","abstract_excerpt":"We study the online routing problem in large language model serving, where requests arrive sequentially and must be dispatched to parallel decode workers under tight batch-size and KV-cache constraints. Unlike widely used routing heuristics that are not tied to explicit service-level objectives (SLOs) and offer limited control over latency-throughput trade-offs, we introduce a multi-objective optimization framework that formulates routing as an online linear programming with interpretable decision rewards. We apply an efficient bid-price control policy based on the online linear programming th"},"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":"2607.03948","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-04T16:48:37Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"d0efac91b4f6dec132261fa17bac1fb2c6d01bc61374c447a9add91ca41708ca","abstract_canon_sha256":"5fe17a91fa2aba57524a205372682e3fac027425da85e3170327051109ea6ecd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:18:49.137061Z","signature_b64":"kVYlxCd6cbMbl/06cTvfSMsxnx7lZu+KaMLZe9/w2xLIQuIHw+aTRa6jIk9AtP//ulWPw3+lOi6XI0Gthb4EAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b82979d13df796244fa12b839ce5526117bd0f22b4857bc05fd32462c67b1958","last_reissued_at":"2026-07-07T02:18:49.136141Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:18:49.136141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Online Linear Programming for Multi-Objective Routing in LLM Serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"cs.AI","authors_text":"Yinyu Ye, Zijie Zhou, Zixi Chen","submitted_at":"2026-07-04T16:48:37Z","abstract_excerpt":"We study the online routing problem in large language model serving, where requests arrive sequentially and must be dispatched to parallel decode workers under tight batch-size and KV-cache constraints. Unlike widely used routing heuristics that are not tied to explicit service-level objectives (SLOs) and offer limited control over latency-throughput trade-offs, we introduce a multi-objective optimization framework that formulates routing as an online linear programming with interpretable decision rewards. We apply an efficient bid-price control policy based on the online linear programming th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.03948","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/2607.03948/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":"2607.03948","created_at":"2026-07-07T02:18:49.136257+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.03948v1","created_at":"2026-07-07T02:18:49.136257+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.03948","created_at":"2026-07-07T02:18:49.136257+00:00"},{"alias_kind":"pith_short_12","alias_value":"XAUXTUJ566LC","created_at":"2026-07-07T02:18:49.136257+00:00"},{"alias_kind":"pith_short_16","alias_value":"XAUXTUJ566LCIT5B","created_at":"2026-07-07T02:18:49.136257+00:00"},{"alias_kind":"pith_short_8","alias_value":"XAUXTUJ5","created_at":"2026-07-07T02:18:49.136257+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/XAUXTUJ566LCIT5BFOBZZZKSME","json":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME.json","graph_json":"https://pith.science/api/pith-number/XAUXTUJ566LCIT5BFOBZZZKSME/graph.json","events_json":"https://pith.science/api/pith-number/XAUXTUJ566LCIT5BFOBZZZKSME/events.json","paper":"https://pith.science/paper/XAUXTUJ5"},"agent_actions":{"view_html":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME","download_json":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME.json","view_paper":"https://pith.science/paper/XAUXTUJ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.03948&json=true","fetch_graph":"https://pith.science/api/pith-number/XAUXTUJ566LCIT5BFOBZZZKSME/graph.json","fetch_events":"https://pith.science/api/pith-number/XAUXTUJ566LCIT5BFOBZZZKSME/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME/action/storage_attestation","attest_author":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME/action/author_attestation","sign_citation":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME/action/citation_signature","submit_replication":"https://pith.science/pith/XAUXTUJ566LCIT5BFOBZZZKSME/action/replication_record"}},"created_at":"2026-07-07T02:18:49.136257+00:00","updated_at":"2026-07-07T02:18:49.136257+00:00"}