{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:5GAF3A6ZHKNCXTXFBR5IOZW4W6","short_pith_number":"pith:5GAF3A6Z","schema_version":"1.0","canonical_sha256":"e9805d83d93a9a2bcee50c7a8766dcb7a0b5d750cce9c5c548cf5a3c647a412f","source":{"kind":"arxiv","id":"2305.15781","version":1},"attestation_state":"computed","paper":{"title":"VanillaKD: Revisit the Power of Vanilla Knowledge Distillation from Small Scale to Large Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chang Xu, Han Hu, Jianyuan Guo, Kai Han, Yunhe Wang, Zhiwei Hao","submitted_at":"2023-05-25T06:50:08Z","abstract_excerpt":"The tremendous success of large models trained on extensive datasets demonstrates that scale is a key ingredient in achieving superior results. Therefore, the reflection on the rationality of designing knowledge distillation (KD) approaches for limited-capacity architectures solely based on small-scale datasets is now deemed imperative. In this paper, we identify the \\emph{small data pitfall} that presents in previous KD methods, which results in the underestimation of the power of vanilla KD framework on large-scale datasets such as ImageNet-1K. Specifically, we show that employing stronger d"},"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":"2305.15781","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-25T06:50:08Z","cross_cats_sorted":[],"title_canon_sha256":"1d5d910bdaffcb561391a10da8a411cbbc10f1e001a3f32b9ad65607d959be74","abstract_canon_sha256":"7a5e2772b2697aa436b945d0278002d01b954bf2eb1a057befb6f27736d95ab4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:44.757185Z","signature_b64":"ZBBA/XTbiwCH3UATVZaywY1kvKknYhYLCfk8sKRiuejJtVOBh6kKew77sg6OyWQM/LLe1m4LNW7PJKBKSp+9Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e9805d83d93a9a2bcee50c7a8766dcb7a0b5d750cce9c5c548cf5a3c647a412f","last_reissued_at":"2026-07-05T06:13:44.756747Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:44.756747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VanillaKD: Revisit the Power of Vanilla Knowledge Distillation from Small Scale to Large Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chang Xu, Han Hu, Jianyuan Guo, Kai Han, Yunhe Wang, Zhiwei Hao","submitted_at":"2023-05-25T06:50:08Z","abstract_excerpt":"The tremendous success of large models trained on extensive datasets demonstrates that scale is a key ingredient in achieving superior results. Therefore, the reflection on the rationality of designing knowledge distillation (KD) approaches for limited-capacity architectures solely based on small-scale datasets is now deemed imperative. In this paper, we identify the \\emph{small data pitfall} that presents in previous KD methods, which results in the underestimation of the power of vanilla KD framework on large-scale datasets such as ImageNet-1K. Specifically, we show that employing stronger d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.15781","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/2305.15781/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":"2305.15781","created_at":"2026-07-05T06:13:44.756802+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.15781v1","created_at":"2026-07-05T06:13:44.756802+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.15781","created_at":"2026-07-05T06:13:44.756802+00:00"},{"alias_kind":"pith_short_12","alias_value":"5GAF3A6ZHKNC","created_at":"2026-07-05T06:13:44.756802+00:00"},{"alias_kind":"pith_short_16","alias_value":"5GAF3A6ZHKNCXTXF","created_at":"2026-07-05T06:13:44.756802+00:00"},{"alias_kind":"pith_short_8","alias_value":"5GAF3A6Z","created_at":"2026-07-05T06:13:44.756802+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.26857","citing_title":"Edge AI for Automotive Vulnerable Road User Safety: Deployable Detection via Knowledge Distillation","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6","json":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6.json","graph_json":"https://pith.science/api/pith-number/5GAF3A6ZHKNCXTXFBR5IOZW4W6/graph.json","events_json":"https://pith.science/api/pith-number/5GAF3A6ZHKNCXTXFBR5IOZW4W6/events.json","paper":"https://pith.science/paper/5GAF3A6Z"},"agent_actions":{"view_html":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6","download_json":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6.json","view_paper":"https://pith.science/paper/5GAF3A6Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.15781&json=true","fetch_graph":"https://pith.science/api/pith-number/5GAF3A6ZHKNCXTXFBR5IOZW4W6/graph.json","fetch_events":"https://pith.science/api/pith-number/5GAF3A6ZHKNCXTXFBR5IOZW4W6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6/action/storage_attestation","attest_author":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6/action/author_attestation","sign_citation":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6/action/citation_signature","submit_replication":"https://pith.science/pith/5GAF3A6ZHKNCXTXFBR5IOZW4W6/action/replication_record"}},"created_at":"2026-07-05T06:13:44.756802+00:00","updated_at":"2026-07-05T06:13:44.756802+00:00"}