{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:D2STXRV27UTVMXPA5XOM4I7AQB","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":"4e4fa65ae995333fa52f68db4683c53a351a868f7e248f7d30dd4329eefe5c0e","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.PF","submitted_at":"2024-10-31T00:58:11Z","title_canon_sha256":"e9fa2f597c875b7448a266f704959c61f3065a0d163d6fcce2bbc02abf5368fd"},"schema_version":"1.0","source":{"id":"2410.23537","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.23537","created_at":"2026-07-05T09:29:01Z"},{"alias_kind":"arxiv_version","alias_value":"2410.23537v1","created_at":"2026-07-05T09:29:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.23537","created_at":"2026-07-05T09:29:01Z"},{"alias_kind":"pith_short_12","alias_value":"D2STXRV27UTV","created_at":"2026-07-05T09:29:01Z"},{"alias_kind":"pith_short_16","alias_value":"D2STXRV27UTVMXPA","created_at":"2026-07-05T09:29:01Z"},{"alias_kind":"pith_short_8","alias_value":"D2STXRV2","created_at":"2026-07-05T09:29:01Z"}],"graph_snapshots":[{"event_id":"sha256:9999479cdab8d90360319014d085622662dc6378cd2141973eb13afe2c0790e1","target":"graph","created_at":"2026-07-05T09:29:01Z","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/2410.23537/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) represent a revolutionary advancement in the contemporary landscape of artificial general intelligence (AGI). As exemplified by ChatGPT, LLM-based applications necessitate minimal response latency and maximal throughput for inference serving. However, due to the unpredictability of LLM execution, the first-come-first-serve (FCFS) scheduling policy employed by current LLM serving systems suffers from head-of-line (HoL) blocking issues and long job response times.\n  In this paper, we propose a new efficient LLM inference serving framework, named ALISE. The key design","authors_text":"Jun Wang, Youpeng Zhao","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.PF","submitted_at":"2024-10-31T00:58:11Z","title":"ALISE: Accelerating Large Language Model Serving with Speculative Scheduling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.23537","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":"sha256:4a46d337ce399d046397899a2721292f84a490d04c8e2ce98883b75b8a03a0d5","target":"record","created_at":"2026-07-05T09:29:01Z","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":"4e4fa65ae995333fa52f68db4683c53a351a868f7e248f7d30dd4329eefe5c0e","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.PF","submitted_at":"2024-10-31T00:58:11Z","title_canon_sha256":"e9fa2f597c875b7448a266f704959c61f3065a0d163d6fcce2bbc02abf5368fd"},"schema_version":"1.0","source":{"id":"2410.23537","kind":"arxiv","version":1}},"canonical_sha256":"1ea53bc6bafd27565de0eddcce23e0807c22778270c0a30551b18316751a774c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1ea53bc6bafd27565de0eddcce23e0807c22778270c0a30551b18316751a774c","first_computed_at":"2026-07-05T09:29:01.772754Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:29:01.772754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"n9xCMEnsmc96qD13THQMPDvzSQV7t81TMrCOwSSJe4VqQJZOtjz2xenu4n227yChhFSGD5fBB7IZ8XzfkbmBBg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:29:01.773190Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.23537","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4a46d337ce399d046397899a2721292f84a490d04c8e2ce98883b75b8a03a0d5","sha256:9999479cdab8d90360319014d085622662dc6378cd2141973eb13afe2c0790e1"],"state_sha256":"e74e022ef75da76990e601d15a50fcf458d9beadce1d650f03c42c1e473f7c3c"}