{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:I5MQMOLERYWF7DW5X2NR47NNEO","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"cb9ee57f1d119ec88bef5d143a41bc2f62a06ec0df7907cd92fd9932d22a4606","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2025-04-05T17:41:26Z","title_canon_sha256":"3d2a4d1e1f3d20a154e5e2a3b135a86c5ee8b48673d5a71090c607da01f5c12a"},"schema_version":"1.0","source":{"id":"2504.08784","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.08784","created_at":"2026-07-05T10:47:59Z"},{"alias_kind":"arxiv_version","alias_value":"2504.08784v1","created_at":"2026-07-05T10:47:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.08784","created_at":"2026-07-05T10:47:59Z"},{"alias_kind":"pith_short_12","alias_value":"I5MQMOLERYWF","created_at":"2026-07-05T10:47:59Z"},{"alias_kind":"pith_short_16","alias_value":"I5MQMOLERYWF7DW5","created_at":"2026-07-05T10:47:59Z"},{"alias_kind":"pith_short_8","alias_value":"I5MQMOLE","created_at":"2026-07-05T10:47:59Z"}],"graph_snapshots":[{"event_id":"sha256:43ea1534c9d076ce0e86025e036fe29612da80770a015803a9642848467e5b78","target":"graph","created_at":"2026-07-05T10:47:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2504.08784/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces SLOs-Serve, a system designed for serving multi-stage large language model (LLM) requests with application- and stage-specific service level objectives (SLOs). The key idea behind SLOs-Serve is to customize the allocation of tokens to meet these SLO requirements. SLOs-Serve uses a multi-SLO dynamic programming-based algorithm to continuously optimize token allocations under SLO constraints by exploring the full design space of chunked prefill and (optional) speculative decoding. Leveraging this resource planning algorithm, SLOs-Serve effectively supports multi-SLOs and mu","authors_text":"Arvind Krishnamurthy, Phillip B. Gibbons, Samira Khan, Siyuan Chen, Zhipeng Jia","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2025-04-05T17:41:26Z","title":"SLOs-Serve: Optimized Serving of Multi-SLO LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.08784","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ffac1fd8d980f8d0f9c5776c32dfbef965278a8b8c8c1059bba309c95930d42e","target":"record","created_at":"2026-07-05T10:47:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"cb9ee57f1d119ec88bef5d143a41bc2f62a06ec0df7907cd92fd9932d22a4606","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2025-04-05T17:41:26Z","title_canon_sha256":"3d2a4d1e1f3d20a154e5e2a3b135a86c5ee8b48673d5a71090c607da01f5c12a"},"schema_version":"1.0","source":{"id":"2504.08784","kind":"arxiv","version":1}},"canonical_sha256":"47590639648e2c5f8eddbe9b1e7dad23bde21ee4679a62f09eef0327240aa998","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"47590639648e2c5f8eddbe9b1e7dad23bde21ee4679a62f09eef0327240aa998","first_computed_at":"2026-07-05T10:47:59.344344Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:47:59.344344Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xsRyFDuvHPhnYiQAWZNEQhGKZYHeehSS9+Fo4QK+Y5RzIKbJejx51TpbH4Flqb3l4fmRpbkQ/71uvVCHP1/sAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:47:59.344820Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.08784","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ffac1fd8d980f8d0f9c5776c32dfbef965278a8b8c8c1059bba309c95930d42e","sha256:43ea1534c9d076ce0e86025e036fe29612da80770a015803a9642848467e5b78"],"state_sha256":"9421f38ccb615a3b1423328bbefc9e44245b16c5de99cdf216da96921930b8d6"}