{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:3UNJHOKX3ZIP752Q66AOLWZAQP","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":"2f81bf5886885711cede8487d5194cbe0e1b50b1eb314e43eb0b87ea60eee50e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-04T08:29:47Z","title_canon_sha256":"a1c0c0c6b87927f1d32f0c9a2fad3a0392d4fb5f372a6d8dc5d74e4adc743c03"},"schema_version":"1.0","source":{"id":"2605.02329","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.02329","created_at":"2026-05-26T02:05:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.02329v2","created_at":"2026-05-26T02:05:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.02329","created_at":"2026-05-26T02:05:09Z"},{"alias_kind":"pith_short_12","alias_value":"3UNJHOKX3ZIP","created_at":"2026-05-26T02:05:09Z"},{"alias_kind":"pith_short_16","alias_value":"3UNJHOKX3ZIP752Q","created_at":"2026-05-26T02:05:09Z"},{"alias_kind":"pith_short_8","alias_value":"3UNJHOKX","created_at":"2026-05-26T02:05:09Z"}],"graph_snapshots":[{"event_id":"sha256:e64f9af9029f4a193874ba55ed2f705bfc272ca889c19799bdef946e8092226e","target":"graph","created_at":"2026-05-26T02:05:09Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experimental results demonstrate that, compared with state-of-the-art baselines, Kairos improves TTFT SLO attainment by up to 23.9%, TPOT SLO attainment by up to 27.1%, end-to-end SLO attainment by up to 33.8%, and decode throughput by up to 19.3%."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the prediction of prefill completion times is accurate enough to make good scheduling decisions and that the chosen online serving dataset accurately reflects production request patterns."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Kairos improves SLO attainment and throughput in LLM serving by adapting to request length imbalance with priority scheduling and adaptive batching."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Kairos improves TTFT SLO attainment by up to 24% and decode throughput by 19% in disaggregated LLM inference."}],"snapshot_sha256":"a73493e69cefcbf703d7d281e617922ed08d78aea1265ceeaada6da3ece0fd59"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"1d58b2b3e059ebb17d0c742a8e62e76387f59714ebeb2c9e6fff6c738b46cd50"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T16:33:55.695019Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-20T03:31:22.702230Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T16:28:57.228857Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.02329/integrity.json","findings":[],"snapshot_sha256":"bb49f5b19e08761ff1a5810d5c76659f838166c0318d0c20354189148d76263e","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In production environments, large language model (LLM) serving is required to meet stringent service-level objectives (SLOs) amid highly variable request patterns. In practice, request lengths follow a long-tail distribution, which gives rise to head-of-line blocking on the prefill side and underutilization caused by stragglers on the decode side in disaggregated serving architectures. Current systems, which adopt first-come-first-served (FCFS) scheduling for prefill and continuous batching for decode, lack the ability to adapt to this imbalance, resulting in compromised SLO attainment and red","authors_text":"Qipeng Wang, Zhendong Yang","cross_cats":[],"headline":"Kairos improves TTFT SLO attainment by up to 24% and decode throughput by 19% in disaggregated LLM inference.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-04T08:29:47Z","title":"Taming Request Imbalance: SLO-Aware Scheduling for Disaggregated LLM Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.02329","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T18:24:01.920421Z","id":"91ef7cad-1e1c-42b4-8762-a4ffd40ba1e4","model_set":{"reader":"grok-4.3"},"one_line_summary":"Kairos improves SLO attainment and throughput in LLM serving by adapting to request length imbalance with priority scheduling and adaptive batching.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Kairos improves TTFT SLO attainment by up to 24% and decode throughput by 19% in disaggregated LLM inference.","strongest_claim":"Experimental results demonstrate that, compared with state-of-the-art baselines, Kairos improves TTFT SLO attainment by up to 23.9%, TPOT SLO attainment by up to 27.1%, end-to-end SLO attainment by up to 33.8%, and decode throughput by up to 19.3%.","weakest_assumption":"That the prediction of prefill completion times is accurate enough to make good scheduling decisions and that the chosen online serving dataset accurately reflects production request patterns."}},"verdict_id":"91ef7cad-1e1c-42b4-8762-a4ffd40ba1e4"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:be48b516d986da29c9076864e67a41a2bca13baf1bb9ccab85f408685369ef44","target":"record","created_at":"2026-05-26T02:05:09Z","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":"2f81bf5886885711cede8487d5194cbe0e1b50b1eb314e43eb0b87ea60eee50e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-04T08:29:47Z","title_canon_sha256":"a1c0c0c6b87927f1d32f0c9a2fad3a0392d4fb5f372a6d8dc5d74e4adc743c03"},"schema_version":"1.0","source":{"id":"2605.02329","kind":"arxiv","version":2}},"canonical_sha256":"dd1a93b957de50fff750f780e5db2083dded13054e9a0bbd29450f62ee6d2bfe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"dd1a93b957de50fff750f780e5db2083dded13054e9a0bbd29450f62ee6d2bfe","first_computed_at":"2026-05-26T02:05:09.868224Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T02:05:09.868224Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"60v2zYSwm6HjK7ows+tmLN9MqoE8emxSTDUzbfDngt9i0zaChdhiD9cC+ICUCesTlzhGiuxAwPlRNv9ZCHTRDQ==","signature_status":"signed_v1","signed_at":"2026-05-26T02:05:09.869878Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.02329","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:be48b516d986da29c9076864e67a41a2bca13baf1bb9ccab85f408685369ef44","sha256:e64f9af9029f4a193874ba55ed2f705bfc272ca889c19799bdef946e8092226e"],"state_sha256":"df0bbb18fe2dbaba67a3d161849adbf8013e498feb1de42bb1f9abaa47379516"}