{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3GKBKLGQO43F2VS4YEH3M32IQH","short_pith_number":"pith:3GKBKLGQ","schema_version":"1.0","canonical_sha256":"d994152cd077365d565cc10fb66f4881c8ab97211fe6859c51b6d7b39f3a76ab","source":{"kind":"arxiv","id":"2306.06629","version":1},"attestation_state":"computed","paper":{"title":"GKD: A General Knowledge Distillation Framework for Large-scale Pre-trained Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hongyin Tang, Jiahao Liu, Jie Tang, Jingang Wang, Keqing He, Peng Zhang, Shicheng Tan, Shu Zhao, Weng Lam Tam, Wenwen Gong, Yang Yang, Yuanchun Wang","submitted_at":"2023-06-11T09:17:21Z","abstract_excerpt":"Currently, the reduction in the parameter scale of large-scale pre-trained language models (PLMs) through knowledge distillation has greatly facilitated their widespread deployment on various devices. However, the deployment of knowledge distillation systems faces great challenges in real-world industrial-strength applications, which require the use of complex distillation methods on even larger-scale PLMs (over 10B), limited by memory on GPUs and the switching of methods. To overcome these challenges, we propose GKD, a general knowledge distillation framework that supports distillation on lar"},"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":"2306.06629","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-11T09:17:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2306e41db846efc7a099573bbf91d6673855b23368e8416e6d17a810b301e0ac","abstract_canon_sha256":"dbfb96fce14a7e315118e57e4fbe9bf111e4345a5f4ccbdf157ec4be0a6981c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:19:27.064907Z","signature_b64":"Xf6nB1hfYFsdQy85jemqrvSG9/hEoibhYZmN6rFGEqOFWqKMHotxjwMGJhZDHUCEyZ3x0SIwH+15XkSMFB7iBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d994152cd077365d565cc10fb66f4881c8ab97211fe6859c51b6d7b39f3a76ab","last_reissued_at":"2026-07-05T06:19:27.064416Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:19:27.064416Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GKD: A General Knowledge Distillation Framework for Large-scale Pre-trained Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hongyin Tang, Jiahao Liu, Jie Tang, Jingang Wang, Keqing He, Peng Zhang, Shicheng Tan, Shu Zhao, Weng Lam Tam, Wenwen Gong, Yang Yang, Yuanchun Wang","submitted_at":"2023-06-11T09:17:21Z","abstract_excerpt":"Currently, the reduction in the parameter scale of large-scale pre-trained language models (PLMs) through knowledge distillation has greatly facilitated their widespread deployment on various devices. However, the deployment of knowledge distillation systems faces great challenges in real-world industrial-strength applications, which require the use of complex distillation methods on even larger-scale PLMs (over 10B), limited by memory on GPUs and the switching of methods. To overcome these challenges, we propose GKD, a general knowledge distillation framework that supports distillation on lar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.06629","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/2306.06629/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":"2306.06629","created_at":"2026-07-05T06:19:27.064482+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.06629v1","created_at":"2026-07-05T06:19:27.064482+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.06629","created_at":"2026-07-05T06:19:27.064482+00:00"},{"alias_kind":"pith_short_12","alias_value":"3GKBKLGQO43F","created_at":"2026-07-05T06:19:27.064482+00:00"},{"alias_kind":"pith_short_16","alias_value":"3GKBKLGQO43F2VS4","created_at":"2026-07-05T06:19:27.064482+00:00"},{"alias_kind":"pith_short_8","alias_value":"3GKBKLGQ","created_at":"2026-07-05T06:19:27.064482+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27797","citing_title":"Optimizing Teacher-Student Partitioning for Scalable Knowledge Distillation on HPC Systems","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01347","citing_title":"MAD-OPD: Breaking the Ceiling in On-Policy Distillation via Multi-Agent Debate","ref_index":34,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH","json":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH.json","graph_json":"https://pith.science/api/pith-number/3GKBKLGQO43F2VS4YEH3M32IQH/graph.json","events_json":"https://pith.science/api/pith-number/3GKBKLGQO43F2VS4YEH3M32IQH/events.json","paper":"https://pith.science/paper/3GKBKLGQ"},"agent_actions":{"view_html":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH","download_json":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH.json","view_paper":"https://pith.science/paper/3GKBKLGQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.06629&json=true","fetch_graph":"https://pith.science/api/pith-number/3GKBKLGQO43F2VS4YEH3M32IQH/graph.json","fetch_events":"https://pith.science/api/pith-number/3GKBKLGQO43F2VS4YEH3M32IQH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH/action/storage_attestation","attest_author":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH/action/author_attestation","sign_citation":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH/action/citation_signature","submit_replication":"https://pith.science/pith/3GKBKLGQO43F2VS4YEH3M32IQH/action/replication_record"}},"created_at":"2026-07-05T06:19:27.064482+00:00","updated_at":"2026-07-05T06:19:27.064482+00:00"}