{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VTIM57CFMGQ5Z3VFAN3ONO22KR","short_pith_number":"pith:VTIM57CF","schema_version":"1.0","canonical_sha256":"acd0cefc4561a1dceea50376e6bb5a547dddbe47393bd4641ebae8d5bc27f1ae","source":{"kind":"arxiv","id":"2205.14443","version":2},"attestation_state":"computed","paper":{"title":"A Closer Look at Self-Supervised Lightweight Vision Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jin Gao, Shaoru Wang, Weiming Hu, Xiaoqin Zhang, Zeming Li","submitted_at":"2022-05-28T14:14:57Z","abstract_excerpt":"Self-supervised learning on large-scale Vision Transformers (ViTs) as pre-training methods has achieved promising downstream performance. Yet, how much these pre-training paradigms promote lightweight ViTs' performance is considerably less studied. In this work, we develop and benchmark several self-supervised pre-training methods on image classification tasks and some downstream dense prediction tasks. We surprisingly find that if proper pre-training is adopted, even vanilla lightweight ViTs show comparable performance to previous SOTA networks with delicate architecture design. It breaks the"},"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":"2205.14443","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-28T14:14:57Z","cross_cats_sorted":[],"title_canon_sha256":"39a7434650dd3c57ada6e6313366a106fe82de9db9b5f18aba24a7a9c8accbe2","abstract_canon_sha256":"965021c07174903f667248f6bd6e78b31918f943137ea0527bdd6e987c27cfe7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:06:40.034200Z","signature_b64":"coVkADOCr1Rue4wXchV/Hc28ICwyoijLcPyzKxkfU4WkgsRYmiSZdDHhOtSogW6z668QnlOLpzSw2/MeGeuTCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acd0cefc4561a1dceea50376e6bb5a547dddbe47393bd4641ebae8d5bc27f1ae","last_reissued_at":"2026-07-05T06:06:40.033802Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:06:40.033802Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Closer Look at Self-Supervised Lightweight Vision Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jin Gao, Shaoru Wang, Weiming Hu, Xiaoqin Zhang, Zeming Li","submitted_at":"2022-05-28T14:14:57Z","abstract_excerpt":"Self-supervised learning on large-scale Vision Transformers (ViTs) as pre-training methods has achieved promising downstream performance. Yet, how much these pre-training paradigms promote lightweight ViTs' performance is considerably less studied. In this work, we develop and benchmark several self-supervised pre-training methods on image classification tasks and some downstream dense prediction tasks. We surprisingly find that if proper pre-training is adopted, even vanilla lightweight ViTs show comparable performance to previous SOTA networks with delicate architecture design. It breaks the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.14443","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/2205.14443/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":"2205.14443","created_at":"2026-07-05T06:06:40.033860+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.14443v2","created_at":"2026-07-05T06:06:40.033860+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.14443","created_at":"2026-07-05T06:06:40.033860+00:00"},{"alias_kind":"pith_short_12","alias_value":"VTIM57CFMGQ5","created_at":"2026-07-05T06:06:40.033860+00:00"},{"alias_kind":"pith_short_16","alias_value":"VTIM57CFMGQ5Z3VF","created_at":"2026-07-05T06:06:40.033860+00:00"},{"alias_kind":"pith_short_8","alias_value":"VTIM57CF","created_at":"2026-07-05T06:06:40.033860+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/VTIM57CFMGQ5Z3VFAN3ONO22KR","json":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR.json","graph_json":"https://pith.science/api/pith-number/VTIM57CFMGQ5Z3VFAN3ONO22KR/graph.json","events_json":"https://pith.science/api/pith-number/VTIM57CFMGQ5Z3VFAN3ONO22KR/events.json","paper":"https://pith.science/paper/VTIM57CF"},"agent_actions":{"view_html":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR","download_json":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR.json","view_paper":"https://pith.science/paper/VTIM57CF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.14443&json=true","fetch_graph":"https://pith.science/api/pith-number/VTIM57CFMGQ5Z3VFAN3ONO22KR/graph.json","fetch_events":"https://pith.science/api/pith-number/VTIM57CFMGQ5Z3VFAN3ONO22KR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR/action/storage_attestation","attest_author":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR/action/author_attestation","sign_citation":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR/action/citation_signature","submit_replication":"https://pith.science/pith/VTIM57CFMGQ5Z3VFAN3ONO22KR/action/replication_record"}},"created_at":"2026-07-05T06:06:40.033860+00:00","updated_at":"2026-07-05T06:06:40.033860+00:00"}