{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RURAYOQFLRAZFIIONOJQTHW7MC","short_pith_number":"pith:RURAYOQF","schema_version":"1.0","canonical_sha256":"8d220c3a055c4192a10e6b93099edf6090d1a7776a419ed05898b17a3cd77d57","source":{"kind":"arxiv","id":"2504.17449","version":1},"attestation_state":"computed","paper":{"title":"HMI: Hierarchical Knowledge Management for Efficient Multi-Tenant Inference in Pretrained Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Gang Chen, Guiming Xie, Huan Li, Jue Wang, Jun Zhang, Ke Chen, Lidan Shou, Qin Xie, Xuejian Gong","submitted_at":"2025-04-24T11:28:40Z","abstract_excerpt":"The significant computational demands of pretrained language models (PLMs), which often require dedicated hardware, present a substantial challenge in serving them efficiently, especially in multi-tenant environments. To address this, we introduce HMI, a Hierarchical knowledge management-based Multi-tenant Inference system, designed to manage tenants with distinct PLMs resource-efficiently. Our approach is three-fold: Firstly, we categorize PLM knowledge into general, domain-specific, and task-specific. Leveraging insights on knowledge acquisition across different model layers, we construct hi"},"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":"2504.17449","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-24T11:28:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"273a3c0b4a14f86059faf85c9fe4e11eaf8607b9864f8590de6aed2164b1779c","abstract_canon_sha256":"bf3bb75bbe12c2afd0093fb7ba4a642a721a5598d6a2a1224773dbba8f23b600"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:29.004220Z","signature_b64":"iZ70OKodrlFh6eOxJKyw1DM9/NipQ9jHwRRfN8PSBl382x4S9OngDH+5kMXfhDLMJHlFWpwCqiCMSor+YHVSBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d220c3a055c4192a10e6b93099edf6090d1a7776a419ed05898b17a3cd77d57","last_reissued_at":"2026-07-05T10:53:29.003697Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:29.003697Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HMI: Hierarchical Knowledge Management for Efficient Multi-Tenant Inference in Pretrained Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Gang Chen, Guiming Xie, Huan Li, Jue Wang, Jun Zhang, Ke Chen, Lidan Shou, Qin Xie, Xuejian Gong","submitted_at":"2025-04-24T11:28:40Z","abstract_excerpt":"The significant computational demands of pretrained language models (PLMs), which often require dedicated hardware, present a substantial challenge in serving them efficiently, especially in multi-tenant environments. To address this, we introduce HMI, a Hierarchical knowledge management-based Multi-tenant Inference system, designed to manage tenants with distinct PLMs resource-efficiently. Our approach is three-fold: Firstly, we categorize PLM knowledge into general, domain-specific, and task-specific. Leveraging insights on knowledge acquisition across different model layers, we construct hi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.17449","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/2504.17449/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":"2504.17449","created_at":"2026-07-05T10:53:29.003765+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.17449v1","created_at":"2026-07-05T10:53:29.003765+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.17449","created_at":"2026-07-05T10:53:29.003765+00:00"},{"alias_kind":"pith_short_12","alias_value":"RURAYOQFLRAZ","created_at":"2026-07-05T10:53:29.003765+00:00"},{"alias_kind":"pith_short_16","alias_value":"RURAYOQFLRAZFIIO","created_at":"2026-07-05T10:53:29.003765+00:00"},{"alias_kind":"pith_short_8","alias_value":"RURAYOQF","created_at":"2026-07-05T10:53:29.003765+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC","json":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC.json","graph_json":"https://pith.science/api/pith-number/RURAYOQFLRAZFIIONOJQTHW7MC/graph.json","events_json":"https://pith.science/api/pith-number/RURAYOQFLRAZFIIONOJQTHW7MC/events.json","paper":"https://pith.science/paper/RURAYOQF"},"agent_actions":{"view_html":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC","download_json":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC.json","view_paper":"https://pith.science/paper/RURAYOQF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.17449&json=true","fetch_graph":"https://pith.science/api/pith-number/RURAYOQFLRAZFIIONOJQTHW7MC/graph.json","fetch_events":"https://pith.science/api/pith-number/RURAYOQFLRAZFIIONOJQTHW7MC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC/action/storage_attestation","attest_author":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC/action/author_attestation","sign_citation":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC/action/citation_signature","submit_replication":"https://pith.science/pith/RURAYOQFLRAZFIIONOJQTHW7MC/action/replication_record"}},"created_at":"2026-07-05T10:53:29.003765+00:00","updated_at":"2026-07-05T10:53:29.003765+00:00"}