{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LJSVDZREN5YL7ME3UL6MJLO67P","short_pith_number":"pith:LJSVDZRE","schema_version":"1.0","canonical_sha256":"5a6551e6246f70bfb09ba2fcc4addefbe771ed9f1ac85c4494b79105e116a1fd","source":{"kind":"arxiv","id":"2411.16102","version":2},"attestation_state":"computed","paper":{"title":"BlendServe: Optimizing Offline Inference for Auto-regressive Large Models with Resource-aware Batching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Baris Kasikci, Ion Stoica, Jiarong Xing, Kan Zhu, Lianmin Zheng, Shuo Yang, Yang Zhou, Yilong Zhao","submitted_at":"2024-11-25T05:24:53Z","abstract_excerpt":"Offline batch inference, which leverages the flexibility of request batching to achieve higher throughput and lower costs, is becoming more popular for latency-insensitive applications. Meanwhile, recent progress in model capability and modality makes requests more diverse in compute and memory demands, creating unique opportunities for throughput improvement by resource overlapping. However, a request schedule that maximizes resource overlapping can conflict with the schedule that maximizes prefix sharing, a widely-used performance optimization, causing sub-optimal inference throughput. We pr"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2411.16102","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-25T05:24:53Z","cross_cats_sorted":[],"title_canon_sha256":"3e27e0c577d27852ef193bb39dd95f797fb62dffc965eadf6a993ed0cf6053cb","abstract_canon_sha256":"5a8ff2ae7fb97bbeb32ed560825d62d058425765c2df7c2ca8344e17d545db03"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:04:37.648148Z","signature_b64":"89fDLYVGfsaHoFX5UkiJqeSfjoZQeI8PiaIZtzFN7HCIzdA6j6BErovDGTRsKB5zs+I9/qeGDcRa41pNx0NBBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a6551e6246f70bfb09ba2fcc4addefbe771ed9f1ac85c4494b79105e116a1fd","last_reissued_at":"2026-06-09T01:04:37.647628Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:04:37.647628Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BlendServe: Optimizing Offline Inference for Auto-regressive Large Models with Resource-aware Batching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Baris Kasikci, Ion Stoica, Jiarong Xing, Kan Zhu, Lianmin Zheng, Shuo Yang, Yang Zhou, Yilong Zhao","submitted_at":"2024-11-25T05:24:53Z","abstract_excerpt":"Offline batch inference, which leverages the flexibility of request batching to achieve higher throughput and lower costs, is becoming more popular for latency-insensitive applications. Meanwhile, recent progress in model capability and modality makes requests more diverse in compute and memory demands, creating unique opportunities for throughput improvement by resource overlapping. However, a request schedule that maximizes resource overlapping can conflict with the schedule that maximizes prefix sharing, a widely-used performance optimization, causing sub-optimal inference throughput. We pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.16102","kind":"arxiv","version":2},"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/2411.16102/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2411.16102","created_at":"2026-06-09T01:04:37.647693+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.16102v2","created_at":"2026-06-09T01:04:37.647693+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.16102","created_at":"2026-06-09T01:04:37.647693+00:00"},{"alias_kind":"pith_short_12","alias_value":"LJSVDZREN5YL","created_at":"2026-06-09T01:04:37.647693+00:00"},{"alias_kind":"pith_short_16","alias_value":"LJSVDZREN5YL7ME3","created_at":"2026-06-09T01:04:37.647693+00:00"},{"alias_kind":"pith_short_8","alias_value":"LJSVDZRE","created_at":"2026-06-09T01:04:37.647693+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.02960","citing_title":"MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving","ref_index":70,"is_internal_anchor":true},{"citing_arxiv_id":"2605.02189","citing_title":"PipeMax: Enhancing Offline LLM Inference on Commodity GPU Servers","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2605.02960","citing_title":"MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving","ref_index":70,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P","json":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P.json","graph_json":"https://pith.science/api/pith-number/LJSVDZREN5YL7ME3UL6MJLO67P/graph.json","events_json":"https://pith.science/api/pith-number/LJSVDZREN5YL7ME3UL6MJLO67P/events.json","paper":"https://pith.science/paper/LJSVDZRE"},"agent_actions":{"view_html":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P","download_json":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P.json","view_paper":"https://pith.science/paper/LJSVDZRE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.16102&json=true","fetch_graph":"https://pith.science/api/pith-number/LJSVDZREN5YL7ME3UL6MJLO67P/graph.json","fetch_events":"https://pith.science/api/pith-number/LJSVDZREN5YL7ME3UL6MJLO67P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P/action/storage_attestation","attest_author":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P/action/author_attestation","sign_citation":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P/action/citation_signature","submit_replication":"https://pith.science/pith/LJSVDZREN5YL7ME3UL6MJLO67P/action/replication_record"}},"created_at":"2026-06-09T01:04:37.647693+00:00","updated_at":"2026-06-09T01:04:37.647693+00:00"}