{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DPAEFPU3TBO6J6CR36PPRWP6C6","short_pith_number":"pith:DPAEFPU3","canonical_record":{"source":{"id":"2606.18431","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T19:25:37Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"78928151ce54a97aa5d681940077813751ac5a526d501ed196d0cedbe130d14d","abstract_canon_sha256":"6aa6352217b778e6cb66cb21e0e7a3d4f380abccb1e32413fb2f92853b5c0a59"},"schema_version":"1.0"},"canonical_sha256":"1bc042be9b985de4f851df9ef8d9fe17b22b49e02dc69da7fe786bad22f63c1b","source":{"kind":"arxiv","id":"2606.18431","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.18431","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"arxiv_version","alias_value":"2606.18431v1","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18431","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"pith_short_12","alias_value":"DPAEFPU3TBO6","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"pith_short_16","alias_value":"DPAEFPU3TBO6J6CR","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"pith_short_8","alias_value":"DPAEFPU3","created_at":"2026-06-19T16:11:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DPAEFPU3TBO6J6CR36PPRWP6C6","target":"record","payload":{"canonical_record":{"source":{"id":"2606.18431","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T19:25:37Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"78928151ce54a97aa5d681940077813751ac5a526d501ed196d0cedbe130d14d","abstract_canon_sha256":"6aa6352217b778e6cb66cb21e0e7a3d4f380abccb1e32413fb2f92853b5c0a59"},"schema_version":"1.0"},"canonical_sha256":"1bc042be9b985de4f851df9ef8d9fe17b22b49e02dc69da7fe786bad22f63c1b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:11:01.084444Z","signature_b64":"ACIFiLswUQdPDwNIg7ShrCEzYbBMJL+B+R0FgSkej2WRDISrQ7PF8PMP/wKEPEXJj5AwqY2LEw+ofx6dQ2naDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1bc042be9b985de4f851df9ef8d9fe17b22b49e02dc69da7fe786bad22f63c1b","last_reissued_at":"2026-06-19T16:11:01.084092Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:11:01.084092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.18431","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-19T16:11:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7PxyKeCUCPqwI/ZOdK0z/hBPj9sMXiU6yEsriwjc93bNzhllA0gOA1JIojjVW8A1q7BzvPilQ7U4PW4BqCCEDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T04:41:38.065568Z"},"content_sha256":"5b4ed99b9c57e246613a6e160bbdfacd4eab4edb08afec6b282b204c9445084f","schema_version":"1.0","event_id":"sha256:5b4ed99b9c57e246613a6e160bbdfacd4eab4edb08afec6b282b204c9445084f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DPAEFPU3TBO6J6CR36PPRWP6C6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Beyond Prediction: Tail-Aware Scheduling for LLM Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Esha Choukse, G. Edward Suh, Haoran Qiu, Jiayang Chen, Rodrigo Fonseca, Udit Gupta, Yuanfan Chen, Yueying Li, Ziv Scully","submitted_at":"2026-06-16T19:25:37Z","abstract_excerpt":"LLM serving exhibits extreme length variability, making size-based scheduling difficult in practice. Recent LLM schedulers approximate SJF/SRPT using predicted decode lengths or ranks and primarily report mean-centric metrics such as TTFT and TBT. We show that these prediction-driven policies can be fragile under distribution shifts, bursty arrivals, and GPU memory pressure, while offering limited control over the tail latency (P90-P99) that dominates user experience, even with perfect decode-length knowledge. We introduce a distribution-aware, prediction-free scheduling framework that replace"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18431","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/2606.18431/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-19T16:11:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H3RjpEGRq7TRMUtQiLm6auVHkzoiT0xg8y8QHxpvuHziPf4h92R7uGItYxqDacd0Y5owTRY7oCZD/M3snFx+Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T04:41:38.065953Z"},"content_sha256":"20df8a682275c8cc1b068feddef856ebb1c1a4feea2c967d6d608ddb1871c472","schema_version":"1.0","event_id":"sha256:20df8a682275c8cc1b068feddef856ebb1c1a4feea2c967d6d608ddb1871c472"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DPAEFPU3TBO6J6CR36PPRWP6C6/bundle.json","state_url":"https://pith.science/pith/DPAEFPU3TBO6J6CR36PPRWP6C6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DPAEFPU3TBO6J6CR36PPRWP6C6/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-04T04:41:38Z","links":{"resolver":"https://pith.science/pith/DPAEFPU3TBO6J6CR36PPRWP6C6","bundle":"https://pith.science/pith/DPAEFPU3TBO6J6CR36PPRWP6C6/bundle.json","state":"https://pith.science/pith/DPAEFPU3TBO6J6CR36PPRWP6C6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DPAEFPU3TBO6J6CR36PPRWP6C6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DPAEFPU3TBO6J6CR36PPRWP6C6","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":"6aa6352217b778e6cb66cb21e0e7a3d4f380abccb1e32413fb2f92853b5c0a59","cross_cats_sorted":["cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T19:25:37Z","title_canon_sha256":"78928151ce54a97aa5d681940077813751ac5a526d501ed196d0cedbe130d14d"},"schema_version":"1.0","source":{"id":"2606.18431","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.18431","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"arxiv_version","alias_value":"2606.18431v1","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18431","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"pith_short_12","alias_value":"DPAEFPU3TBO6","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"pith_short_16","alias_value":"DPAEFPU3TBO6J6CR","created_at":"2026-06-19T16:11:01Z"},{"alias_kind":"pith_short_8","alias_value":"DPAEFPU3","created_at":"2026-06-19T16:11:01Z"}],"graph_snapshots":[{"event_id":"sha256:20df8a682275c8cc1b068feddef856ebb1c1a4feea2c967d6d608ddb1871c472","target":"graph","created_at":"2026-06-19T16:11:01Z","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/2606.18431/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM serving exhibits extreme length variability, making size-based scheduling difficult in practice. Recent LLM schedulers approximate SJF/SRPT using predicted decode lengths or ranks and primarily report mean-centric metrics such as TTFT and TBT. We show that these prediction-driven policies can be fragile under distribution shifts, bursty arrivals, and GPU memory pressure, while offering limited control over the tail latency (P90-P99) that dominates user experience, even with perfect decode-length knowledge. We introduce a distribution-aware, prediction-free scheduling framework that replace","authors_text":"Esha Choukse, G. Edward Suh, Haoran Qiu, Jiayang Chen, Rodrigo Fonseca, Udit Gupta, Yuanfan Chen, Yueying Li, Ziv Scully","cross_cats":["cs.DC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T19:25:37Z","title":"Beyond Prediction: Tail-Aware Scheduling for LLM Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18431","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":"sha256:5b4ed99b9c57e246613a6e160bbdfacd4eab4edb08afec6b282b204c9445084f","target":"record","created_at":"2026-06-19T16:11:01Z","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":"6aa6352217b778e6cb66cb21e0e7a3d4f380abccb1e32413fb2f92853b5c0a59","cross_cats_sorted":["cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T19:25:37Z","title_canon_sha256":"78928151ce54a97aa5d681940077813751ac5a526d501ed196d0cedbe130d14d"},"schema_version":"1.0","source":{"id":"2606.18431","kind":"arxiv","version":1}},"canonical_sha256":"1bc042be9b985de4f851df9ef8d9fe17b22b49e02dc69da7fe786bad22f63c1b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1bc042be9b985de4f851df9ef8d9fe17b22b49e02dc69da7fe786bad22f63c1b","first_computed_at":"2026-06-19T16:11:01.084092Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:11:01.084092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ACIFiLswUQdPDwNIg7ShrCEzYbBMJL+B+R0FgSkej2WRDISrQ7PF8PMP/wKEPEXJj5AwqY2LEw+ofx6dQ2naDA==","signature_status":"signed_v1","signed_at":"2026-06-19T16:11:01.084444Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.18431","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5b4ed99b9c57e246613a6e160bbdfacd4eab4edb08afec6b282b204c9445084f","sha256:20df8a682275c8cc1b068feddef856ebb1c1a4feea2c967d6d608ddb1871c472"],"state_sha256":"d8b9db72fe3e9e926c13d8e9780bd224a1cf7f6414fb78770bb2866e81b72514"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JKhFWjOJCexbm1wP4iTLlHuP3t1a7IlIpDBsATtl7pc89qIsQ0IOSNG/OSJrri0OEaj1P4lhwCWvvaFOEOEpAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T04:41:38.068298Z","bundle_sha256":"7b9284a9690858d4b9e3ba36332c327e2b641dd99cd202967769be708ad22095"}}