{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4D2L4HYONPZUYQBURRVKYHRUYS","short_pith_number":"pith:4D2L4HYO","schema_version":"1.0","canonical_sha256":"e0f4be1f0e6bf34c40348c6aac1e34c496e2db0a93dd2daa090c0a33dc8e90c6","source":{"kind":"arxiv","id":"2305.14066","version":2},"attestation_state":"computed","paper":{"title":"One-stop Training of Multiple Capacity Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dongdong Zhang, Furu Wei, Haoyang Huang, Lan Jiang, Rui Jiang","submitted_at":"2023-05-23T13:44:09Z","abstract_excerpt":"Training models with varying capacities can be advantageous for deploying them in different scenarios. While high-capacity models offer better performance, low-capacity models require fewer computing resources for training and inference. In this work, we propose a novel one-stop training framework to jointly train high-capacity and low-capactiy models. This framework consists of two composite model architectures and a joint training algorithm called Two-Stage Joint-Training (TSJT). Unlike knowledge distillation, where multiple capacity models are trained from scratch separately, our approach i"},"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":"2305.14066","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T13:44:09Z","cross_cats_sorted":[],"title_canon_sha256":"4e24d2ce5edf6b05328cd0aa59ac2786706736bffbb301aa13ab05c2151e5c5e","abstract_canon_sha256":"18077c860411a839d5eecf353facf639b48ffb646baa413e3e3ed128a4519bf7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:21.874862Z","signature_b64":"DxOxS17U5XQeVJN6ofZIKMAXA20AXcS3q2mTQ1ncvymJJzUuZ9z3Ln/DvXvsEgHB/eBESqpEXjSekEEuxAy+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e0f4be1f0e6bf34c40348c6aac1e34c496e2db0a93dd2daa090c0a33dc8e90c6","last_reissued_at":"2026-07-05T06:13:21.874429Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:21.874429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"One-stop Training of Multiple Capacity Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dongdong Zhang, Furu Wei, Haoyang Huang, Lan Jiang, Rui Jiang","submitted_at":"2023-05-23T13:44:09Z","abstract_excerpt":"Training models with varying capacities can be advantageous for deploying them in different scenarios. While high-capacity models offer better performance, low-capacity models require fewer computing resources for training and inference. In this work, we propose a novel one-stop training framework to jointly train high-capacity and low-capactiy models. This framework consists of two composite model architectures and a joint training algorithm called Two-Stage Joint-Training (TSJT). Unlike knowledge distillation, where multiple capacity models are trained from scratch separately, our approach i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14066","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/2305.14066/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":"2305.14066","created_at":"2026-07-05T06:13:21.874486+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14066v2","created_at":"2026-07-05T06:13:21.874486+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14066","created_at":"2026-07-05T06:13:21.874486+00:00"},{"alias_kind":"pith_short_12","alias_value":"4D2L4HYONPZU","created_at":"2026-07-05T06:13:21.874486+00:00"},{"alias_kind":"pith_short_16","alias_value":"4D2L4HYONPZUYQBU","created_at":"2026-07-05T06:13:21.874486+00:00"},{"alias_kind":"pith_short_8","alias_value":"4D2L4HYO","created_at":"2026-07-05T06:13:21.874486+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/4D2L4HYONPZUYQBURRVKYHRUYS","json":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS.json","graph_json":"https://pith.science/api/pith-number/4D2L4HYONPZUYQBURRVKYHRUYS/graph.json","events_json":"https://pith.science/api/pith-number/4D2L4HYONPZUYQBURRVKYHRUYS/events.json","paper":"https://pith.science/paper/4D2L4HYO"},"agent_actions":{"view_html":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS","download_json":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS.json","view_paper":"https://pith.science/paper/4D2L4HYO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14066&json=true","fetch_graph":"https://pith.science/api/pith-number/4D2L4HYONPZUYQBURRVKYHRUYS/graph.json","fetch_events":"https://pith.science/api/pith-number/4D2L4HYONPZUYQBURRVKYHRUYS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/action/storage_attestation","attest_author":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/action/author_attestation","sign_citation":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/action/citation_signature","submit_replication":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/action/replication_record"}},"created_at":"2026-07-05T06:13:21.874486+00:00","updated_at":"2026-07-05T06:13:21.874486+00:00"}