{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:VIGTEFZYCBRYXHLZYUXKISJCOC","short_pith_number":"pith:VIGTEFZY","schema_version":"1.0","canonical_sha256":"aa0d32173810638b9d79c52ea4492270b62ffa5e569fd257d9e0782f8344025c","source":{"kind":"arxiv","id":"2303.13005","version":2},"attestation_state":"computed","paper":{"title":"From Knowledge Distillation to Self-Knowledge Distillation: A Unified Approach with Normalized Loss and Customized Soft Labels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ailing Zeng, Chun Yuan, Tianke Zhang, Yu Li, Zhe Li, Zhendong Yang","submitted_at":"2023-03-23T02:59:36Z","abstract_excerpt":"Knowledge Distillation (KD) uses the teacher's prediction logits as soft labels to guide the student, while self-KD does not need a real teacher to require the soft labels. This work unifies the formulations of the two tasks by decomposing and reorganizing the generic KD loss into a Normalized KD (NKD) loss and customized soft labels for both target class (image's category) and non-target classes named Universal Self-Knowledge Distillation (USKD). We decompose the KD loss and find the non-target loss from it forces the student's non-target logits to match the teacher's, but the sum of the two "},"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":"2303.13005","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-03-23T02:59:36Z","cross_cats_sorted":[],"title_canon_sha256":"e8a3321ce063355a1e1411b6ee1f40e7d99de9d84c982150c2f48dc6cffc8686","abstract_canon_sha256":"3513e0fe2e7e454a2d224c01124e486deaea308443be46565e2f5cfe6a0a4e63"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:31:10.468216Z","signature_b64":"Wjm0xKA3y/B+HOzfSIr0X6w6J3OcikBPO3gvROOZHPBOUFe1k54GUekpwPstkWS2tXmiDNlFITTzkZ2WPY5oBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa0d32173810638b9d79c52ea4492270b62ffa5e569fd257d9e0782f8344025c","last_reissued_at":"2026-07-05T06:31:10.467740Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:31:10.467740Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Knowledge Distillation to Self-Knowledge Distillation: A Unified Approach with Normalized Loss and Customized Soft Labels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ailing Zeng, Chun Yuan, Tianke Zhang, Yu Li, Zhe Li, Zhendong Yang","submitted_at":"2023-03-23T02:59:36Z","abstract_excerpt":"Knowledge Distillation (KD) uses the teacher's prediction logits as soft labels to guide the student, while self-KD does not need a real teacher to require the soft labels. This work unifies the formulations of the two tasks by decomposing and reorganizing the generic KD loss into a Normalized KD (NKD) loss and customized soft labels for both target class (image's category) and non-target classes named Universal Self-Knowledge Distillation (USKD). We decompose the KD loss and find the non-target loss from it forces the student's non-target logits to match the teacher's, but the sum of the two "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.13005","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/2303.13005/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":"2303.13005","created_at":"2026-07-05T06:31:10.467806+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.13005v2","created_at":"2026-07-05T06:31:10.467806+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.13005","created_at":"2026-07-05T06:31:10.467806+00:00"},{"alias_kind":"pith_short_12","alias_value":"VIGTEFZYCBRY","created_at":"2026-07-05T06:31:10.467806+00:00"},{"alias_kind":"pith_short_16","alias_value":"VIGTEFZYCBRYXHLZ","created_at":"2026-07-05T06:31:10.467806+00:00"},{"alias_kind":"pith_short_8","alias_value":"VIGTEFZY","created_at":"2026-07-05T06:31:10.467806+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/VIGTEFZYCBRYXHLZYUXKISJCOC","json":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC.json","graph_json":"https://pith.science/api/pith-number/VIGTEFZYCBRYXHLZYUXKISJCOC/graph.json","events_json":"https://pith.science/api/pith-number/VIGTEFZYCBRYXHLZYUXKISJCOC/events.json","paper":"https://pith.science/paper/VIGTEFZY"},"agent_actions":{"view_html":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC","download_json":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC.json","view_paper":"https://pith.science/paper/VIGTEFZY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.13005&json=true","fetch_graph":"https://pith.science/api/pith-number/VIGTEFZYCBRYXHLZYUXKISJCOC/graph.json","fetch_events":"https://pith.science/api/pith-number/VIGTEFZYCBRYXHLZYUXKISJCOC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC/action/storage_attestation","attest_author":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC/action/author_attestation","sign_citation":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC/action/citation_signature","submit_replication":"https://pith.science/pith/VIGTEFZYCBRYXHLZYUXKISJCOC/action/replication_record"}},"created_at":"2026-07-05T06:31:10.467806+00:00","updated_at":"2026-07-05T06:31:10.467806+00:00"}