{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PODFNQCBJSKNXODMSWDX2RMQTD","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":"82db838fb7503e473c77aa22badf13cc6e6e5eb628b6c06164529fd3f8d885f3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-01T05:31:21Z","title_canon_sha256":"e994b8631553cdf31074fe20d7851ca6127853a8dfef1e1c34a0e0eaaebf7d20"},"schema_version":"1.0","source":{"id":"2604.00499","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.00499","created_at":"2026-05-26T02:05:08Z"},{"alias_kind":"arxiv_version","alias_value":"2604.00499v2","created_at":"2026-05-26T02:05:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.00499","created_at":"2026-05-26T02:05:08Z"},{"alias_kind":"pith_short_12","alias_value":"PODFNQCBJSKN","created_at":"2026-05-26T02:05:08Z"},{"alias_kind":"pith_short_16","alias_value":"PODFNQCBJSKNXODM","created_at":"2026-05-26T02:05:08Z"},{"alias_kind":"pith_short_8","alias_value":"PODFNQCB","created_at":"2026-05-26T02:05:08Z"}],"graph_snapshots":[{"event_id":"sha256:e2f81f7802517b2eb0c91fdb9eaeb7d174a3e7aff431e1da11d129ae4567f00a","target":"graph","created_at":"2026-05-26T02:05:08Z","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/2604.00499/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"To schedule LLM inference, the \\textit{shortest job first} (SJF) principle is favorable by prioritizing requests with short output lengths to avoid head-of-line (HOL) blocking. Existing methods usually predict a single output length for each request to facilitate scheduling. We argue that such a \\textit{point estimate} does not match the \\textit{stochastic} decoding process of LLM inference, where output length is \\textit{uncertain} by nature and determined by when the end-of-sequence (EOS) token is sampled. Hence, the output length of each request should be fitted with a distribution rather t","authors_text":"Fangcheng Fu, Hao Luo, Hao Wang, Haoyu Zheng, Hongchao Zhu, Jiawei Jiang, Xiaokai Zhou, Xiao Yan, Yongqiang Zhang, Yuanyuan Zhu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-01T05:31:21Z","title":"Scheduling LLM Inference with Uncertainty-Aware Output Length Predictions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.00499","kind":"arxiv","version":2},"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":"sha256:6d24a5365e7185e50f42a984a82cd3960242a7417f41a3c78a884522495a4eec","target":"record","created_at":"2026-05-26T02:05:08Z","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":"82db838fb7503e473c77aa22badf13cc6e6e5eb628b6c06164529fd3f8d885f3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-01T05:31:21Z","title_canon_sha256":"e994b8631553cdf31074fe20d7851ca6127853a8dfef1e1c34a0e0eaaebf7d20"},"schema_version":"1.0","source":{"id":"2604.00499","kind":"arxiv","version":2}},"canonical_sha256":"7b8656c0414c94dbb86c95877d459098f7aa6d8e18d835fc609c2f9388500b3b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7b8656c0414c94dbb86c95877d459098f7aa6d8e18d835fc609c2f9388500b3b","first_computed_at":"2026-05-26T02:05:08.720791Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T02:05:08.720791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"B8lC+PdzufxhCcg3mNbxZC63lOr6UHMWQHqiwtuwFkZmfDkd1yzpP7pY/Cx9GP4s9w4RpftVchbzz+G0U94yDQ==","signature_status":"signed_v1","signed_at":"2026-05-26T02:05:08.721891Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.00499","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6d24a5365e7185e50f42a984a82cd3960242a7417f41a3c78a884522495a4eec","sha256:e2f81f7802517b2eb0c91fdb9eaeb7d174a3e7aff431e1da11d129ae4567f00a"],"state_sha256":"cf8392621f4e63051fc0d67019da44e66a9c5d8aef4c62ed61025aa9ddd961e1"}