{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:C76TANP6Z6CRMYR3DSED5NTA3E","short_pith_number":"pith:C76TANP6","schema_version":"1.0","canonical_sha256":"17fd3035fecf8516623b1c883eb660d93ccfa7117411cde87b084a8434f92192","source":{"kind":"arxiv","id":"2306.02003","version":2},"attestation_state":"computed","paper":{"title":"On Optimal Caching and Model Multiplexing for Large Model Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.PF","cs.SY","eess.SY","stat.ML"],"primary_cat":"cs.LG","authors_text":"Banghua Zhu, Clark Barrett, Jiantao Jiao, Lianmin Zheng, Michael I. Jordan, Ying Sheng","submitted_at":"2023-06-03T05:01:51Z","abstract_excerpt":"Large Language Models (LLMs) and other large foundation models have achieved noteworthy success, but their size exacerbates existing resource consumption and latency challenges. In particular, the large-scale deployment of these models is hindered by the significant resource requirements during inference. In this paper, we study two approaches for mitigating these challenges: employing a cache to store previous queries and learning a model multiplexer to choose from an ensemble of models for query processing.\n  Theoretically, we provide an optimal algorithm for jointly optimizing both approach"},"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":"2306.02003","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-03T05:01:51Z","cross_cats_sorted":["cs.AI","cs.PF","cs.SY","eess.SY","stat.ML"],"title_canon_sha256":"ff8fe195eaf03fbbfa7c241f42fb5014057415356f1991c4fca590a448e3bc2f","abstract_canon_sha256":"ae10b4aa7e3acc25b4889e4053b85973006c941224ddbc1e7bfbba0edbd453b5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:45:36.564225Z","signature_b64":"gUGM1NJ/sKRf1mZFu/V7u9i8DpFYDtnF9SHSLxLXPvqLTxCnCBNht/CJUolVXLQ4ZuUDIS/scPPhyDM75WeNAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"17fd3035fecf8516623b1c883eb660d93ccfa7117411cde87b084a8434f92192","last_reissued_at":"2026-07-05T06:45:36.563725Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:45:36.563725Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Optimal Caching and Model Multiplexing for Large Model Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.PF","cs.SY","eess.SY","stat.ML"],"primary_cat":"cs.LG","authors_text":"Banghua Zhu, Clark Barrett, Jiantao Jiao, Lianmin Zheng, Michael I. Jordan, Ying Sheng","submitted_at":"2023-06-03T05:01:51Z","abstract_excerpt":"Large Language Models (LLMs) and other large foundation models have achieved noteworthy success, but their size exacerbates existing resource consumption and latency challenges. In particular, the large-scale deployment of these models is hindered by the significant resource requirements during inference. In this paper, we study two approaches for mitigating these challenges: employing a cache to store previous queries and learning a model multiplexer to choose from an ensemble of models for query processing.\n  Theoretically, we provide an optimal algorithm for jointly optimizing both approach"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.02003","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/2306.02003/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":"2306.02003","created_at":"2026-07-05T06:45:36.563786+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.02003v2","created_at":"2026-07-05T06:45:36.563786+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.02003","created_at":"2026-07-05T06:45:36.563786+00:00"},{"alias_kind":"pith_short_12","alias_value":"C76TANP6Z6CR","created_at":"2026-07-05T06:45:36.563786+00:00"},{"alias_kind":"pith_short_16","alias_value":"C76TANP6Z6CRMYR3","created_at":"2026-07-05T06:45:36.563786+00:00"},{"alias_kind":"pith_short_8","alias_value":"C76TANP6","created_at":"2026-07-05T06:45:36.563786+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.17288","citing_title":"When Efficiency Backfires: Cascading LLMs Trigger Cascade Failure under Adversarial Attack","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E","json":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E.json","graph_json":"https://pith.science/api/pith-number/C76TANP6Z6CRMYR3DSED5NTA3E/graph.json","events_json":"https://pith.science/api/pith-number/C76TANP6Z6CRMYR3DSED5NTA3E/events.json","paper":"https://pith.science/paper/C76TANP6"},"agent_actions":{"view_html":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E","download_json":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E.json","view_paper":"https://pith.science/paper/C76TANP6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.02003&json=true","fetch_graph":"https://pith.science/api/pith-number/C76TANP6Z6CRMYR3DSED5NTA3E/graph.json","fetch_events":"https://pith.science/api/pith-number/C76TANP6Z6CRMYR3DSED5NTA3E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E/action/storage_attestation","attest_author":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E/action/author_attestation","sign_citation":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E/action/citation_signature","submit_replication":"https://pith.science/pith/C76TANP6Z6CRMYR3DSED5NTA3E/action/replication_record"}},"created_at":"2026-07-05T06:45:36.563786+00:00","updated_at":"2026-07-05T06:45:36.563786+00:00"}