{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:4D2L4HYONPZUYQBURRVKYHRUYS","short_pith_number":"pith:4D2L4HYO","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"},"canonical_sha256":"e0f4be1f0e6bf34c40348c6aac1e34c496e2db0a93dd2daa090c0a33dc8e90c6","source":{"kind":"arxiv","id":"2305.14066","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14066","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14066v2","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14066","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"pith_short_12","alias_value":"4D2L4HYONPZU","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"pith_short_16","alias_value":"4D2L4HYONPZUYQBU","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"pith_short_8","alias_value":"4D2L4HYO","created_at":"2026-07-05T06:13:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:4D2L4HYONPZUYQBURRVKYHRUYS","target":"record","payload":{"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"},"canonical_sha256":"e0f4be1f0e6bf34c40348c6aac1e34c496e2db0a93dd2daa090c0a33dc8e90c6","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"},"source_kind":"arxiv","source_id":"2305.14066","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:13:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KfqdP90k5q/ovSH/S4ukqDd+SVCLSDpa4yyH3pEGH79D7IVvi+PopbNubSmkWsLhbse5Xl0HHBYxsTWcw6wRCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:50:02.195001Z"},"content_sha256":"c74c345a205ec7cf0dbfc1067b639dc84e911d0664acd776e084e071b764c579","schema_version":"1.0","event_id":"sha256:c74c345a205ec7cf0dbfc1067b639dc84e911d0664acd776e084e071b764c579"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:4D2L4HYONPZUYQBURRVKYHRUYS","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:13:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pGRZlVeHzhU7nD/ttHYXGrwSPhKKlRugI05qf6O+YeFf+HBs266J2bEhQB84470CqhwMntnpvaHIoMeaUe/rDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:50:02.195497Z"},"content_sha256":"f84121480aa3d90465d152f5ed84f4163e6b5e8b8f458cebc8679210ced4de7b","schema_version":"1.0","event_id":"sha256:f84121480aa3d90465d152f5ed84f4163e6b5e8b8f458cebc8679210ced4de7b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/bundle.json","state_url":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4D2L4HYONPZUYQBURRVKYHRUYS/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T21:50:02Z","links":{"resolver":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS","bundle":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/bundle.json","state":"https://pith.science/pith/4D2L4HYONPZUYQBURRVKYHRUYS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4D2L4HYONPZUYQBURRVKYHRUYS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:4D2L4HYONPZUYQBURRVKYHRUYS","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":"18077c860411a839d5eecf353facf639b48ffb646baa413e3e3ed128a4519bf7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T13:44:09Z","title_canon_sha256":"4e24d2ce5edf6b05328cd0aa59ac2786706736bffbb301aa13ab05c2151e5c5e"},"schema_version":"1.0","source":{"id":"2305.14066","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14066","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14066v2","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14066","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"pith_short_12","alias_value":"4D2L4HYONPZU","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"pith_short_16","alias_value":"4D2L4HYONPZUYQBU","created_at":"2026-07-05T06:13:21Z"},{"alias_kind":"pith_short_8","alias_value":"4D2L4HYO","created_at":"2026-07-05T06:13:21Z"}],"graph_snapshots":[{"event_id":"sha256:f84121480aa3d90465d152f5ed84f4163e6b5e8b8f458cebc8679210ced4de7b","target":"graph","created_at":"2026-07-05T06:13:21Z","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/2305.14066/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Dongdong Zhang, Furu Wei, Haoyang Huang, Lan Jiang, Rui Jiang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T13:44:09Z","title":"One-stop Training of Multiple Capacity Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14066","kind":"arxiv","version":2},"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:c74c345a205ec7cf0dbfc1067b639dc84e911d0664acd776e084e071b764c579","target":"record","created_at":"2026-07-05T06:13:21Z","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":"18077c860411a839d5eecf353facf639b48ffb646baa413e3e3ed128a4519bf7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T13:44:09Z","title_canon_sha256":"4e24d2ce5edf6b05328cd0aa59ac2786706736bffbb301aa13ab05c2151e5c5e"},"schema_version":"1.0","source":{"id":"2305.14066","kind":"arxiv","version":2}},"canonical_sha256":"e0f4be1f0e6bf34c40348c6aac1e34c496e2db0a93dd2daa090c0a33dc8e90c6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e0f4be1f0e6bf34c40348c6aac1e34c496e2db0a93dd2daa090c0a33dc8e90c6","first_computed_at":"2026-07-05T06:13:21.874429Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:13:21.874429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DxOxS17U5XQeVJN6ofZIKMAXA20AXcS3q2mTQ1ncvymJJzUuZ9z3Ln/DvXvsEgHB/eBESqpEXjSekEEuxAy+Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T06:13:21.874862Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.14066","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c74c345a205ec7cf0dbfc1067b639dc84e911d0664acd776e084e071b764c579","sha256:f84121480aa3d90465d152f5ed84f4163e6b5e8b8f458cebc8679210ced4de7b"],"state_sha256":"75656bdd21d37e7061bd662845cf5b7751eea3ee1bd59d01204f9dca47cf73b4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/6wARurt6EjXufoubql74xDGEBrNnbndUsGrgrb6lk21PdjG+AczcE0qpC3AbEL52SFHLymlhHLJH+Gh9yTGDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T21:50:02.207206Z","bundle_sha256":"25492e242c1f74e06fa3a9a318b64c9231fa7bdd1d1204abbcb6c2da26d422f3"}}