{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:46TCVU4GLGKUZMQ52FCSQY27OK","short_pith_number":"pith:46TCVU4G","schema_version":"1.0","canonical_sha256":"e7a62ad38659954cb21dd14528635f7295d72e09e1bbe8c0ca4e1aba7d374e94","source":{"kind":"arxiv","id":"2202.07940","version":1},"attestation_state":"computed","paper":{"title":"Meta Knowledge Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Boxiao Liu, Hongsheng Li, Jihao Liu, Yu Liu","submitted_at":"2022-02-16T09:09:51Z","abstract_excerpt":"Recent studies pointed out that knowledge distillation (KD) suffers from two degradation problems, the teacher-student gap and the incompatibility with strong data augmentations, making it not applicable to training state-of-the-art models, which are trained with advanced augmentations. However, we observe that a key factor, i.e., the temperatures in the softmax functions for generating probabilities of both the teacher and student models, was mostly overlooked in previous methods. With properly tuned temperatures, such degradation problems of KD can be much mitigated. However, instead of rely"},"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":"2202.07940","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-02-16T09:09:51Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"0cd92b52321c7fa0851a614faadd8e5db6d623e1598e67ad060dd3196cd4781a","abstract_canon_sha256":"5462b8dbfc9098688c817a4d01d1a2aa46e4c4107cd5ccbb0dffcf36868ef8d3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:57:35.393560Z","signature_b64":"jWLdzyJh10FwZCOMfkVM4A3JdxQUta2/hiYtYHEXl2X6pspVl7n73MlCFX9Yiga8Ggv5908TOtmoaZfmA3XwDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7a62ad38659954cb21dd14528635f7295d72e09e1bbe8c0ca4e1aba7d374e94","last_reissued_at":"2026-07-05T03:57:35.393073Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:57:35.393073Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Meta Knowledge Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Boxiao Liu, Hongsheng Li, Jihao Liu, Yu Liu","submitted_at":"2022-02-16T09:09:51Z","abstract_excerpt":"Recent studies pointed out that knowledge distillation (KD) suffers from two degradation problems, the teacher-student gap and the incompatibility with strong data augmentations, making it not applicable to training state-of-the-art models, which are trained with advanced augmentations. However, we observe that a key factor, i.e., the temperatures in the softmax functions for generating probabilities of both the teacher and student models, was mostly overlooked in previous methods. With properly tuned temperatures, such degradation problems of KD can be much mitigated. However, instead of rely"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.07940","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/2202.07940/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":"2202.07940","created_at":"2026-07-05T03:57:35.393130+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.07940v1","created_at":"2026-07-05T03:57:35.393130+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.07940","created_at":"2026-07-05T03:57:35.393130+00:00"},{"alias_kind":"pith_short_12","alias_value":"46TCVU4GLGKU","created_at":"2026-07-05T03:57:35.393130+00:00"},{"alias_kind":"pith_short_16","alias_value":"46TCVU4GLGKUZMQ5","created_at":"2026-07-05T03:57:35.393130+00:00"},{"alias_kind":"pith_short_8","alias_value":"46TCVU4G","created_at":"2026-07-05T03:57:35.393130+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.20357","citing_title":"Consistently Informative Soft-Label Temperature for Knowledge Distillation","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK","json":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK.json","graph_json":"https://pith.science/api/pith-number/46TCVU4GLGKUZMQ52FCSQY27OK/graph.json","events_json":"https://pith.science/api/pith-number/46TCVU4GLGKUZMQ52FCSQY27OK/events.json","paper":"https://pith.science/paper/46TCVU4G"},"agent_actions":{"view_html":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK","download_json":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK.json","view_paper":"https://pith.science/paper/46TCVU4G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.07940&json=true","fetch_graph":"https://pith.science/api/pith-number/46TCVU4GLGKUZMQ52FCSQY27OK/graph.json","fetch_events":"https://pith.science/api/pith-number/46TCVU4GLGKUZMQ52FCSQY27OK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK/action/storage_attestation","attest_author":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK/action/author_attestation","sign_citation":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK/action/citation_signature","submit_replication":"https://pith.science/pith/46TCVU4GLGKUZMQ52FCSQY27OK/action/replication_record"}},"created_at":"2026-07-05T03:57:35.393130+00:00","updated_at":"2026-07-05T03:57:35.393130+00:00"}