{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:73UQAGQ2VCRRZAR6OJRNQ2RL6V","short_pith_number":"pith:73UQAGQ2","schema_version":"1.0","canonical_sha256":"fee9001a1aa8a31c823e7262d86a2bf54711b69150680d4fc6849e4ac33d131f","source":{"kind":"arxiv","id":"2203.07057","version":2},"attestation_state":"computed","paper":{"title":"Self-Promoted Supervision for Few-Shot Transformer","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bowen Dong, Pan Zhou, Shuicheng Yan, Wangmeng Zuo","submitted_at":"2022-03-14T12:53:27Z","abstract_excerpt":"The few-shot learning ability of vision transformers (ViTs) is rarely investigated though heavily desired. In this work, we empirically find that with the same few-shot learning frameworks, \\eg~Meta-Baseline, replacing the widely used CNN feature extractor with a ViT model often severely impairs few-shot classification performance. Moreover, our empirical study shows that in the absence of inductive bias, ViTs often learn the low-qualified token dependencies under few-shot learning regime where only a few labeled training data are available, which largely contributes to the above performance 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":"2203.07057","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2022-03-14T12:53:27Z","cross_cats_sorted":[],"title_canon_sha256":"7e35f9a0b35b945ea473cd4c1a5ef0bbbbae34af578ca2ede1d3861f65f0eec9","abstract_canon_sha256":"240115fa9eabd0e855d4af4e7d69d9b6b1aaf1f8a30e049d814f1ca11e38ec24"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:30:21.630552Z","signature_b64":"37CJ8Uk3U9CzitpYg9pndbB3t4MqLkQE/XXhEZBQbDcWP8sJJFCaH/navIXuDH7fuf1jQRuAoC5/HD91PZO4CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fee9001a1aa8a31c823e7262d86a2bf54711b69150680d4fc6849e4ac33d131f","last_reissued_at":"2026-07-05T04:30:21.630062Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:30:21.630062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Promoted Supervision for Few-Shot Transformer","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bowen Dong, Pan Zhou, Shuicheng Yan, Wangmeng Zuo","submitted_at":"2022-03-14T12:53:27Z","abstract_excerpt":"The few-shot learning ability of vision transformers (ViTs) is rarely investigated though heavily desired. In this work, we empirically find that with the same few-shot learning frameworks, \\eg~Meta-Baseline, replacing the widely used CNN feature extractor with a ViT model often severely impairs few-shot classification performance. Moreover, our empirical study shows that in the absence of inductive bias, ViTs often learn the low-qualified token dependencies under few-shot learning regime where only a few labeled training data are available, which largely contributes to the above performance d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.07057","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/2203.07057/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":"2203.07057","created_at":"2026-07-05T04:30:21.630118+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.07057v2","created_at":"2026-07-05T04:30:21.630118+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.07057","created_at":"2026-07-05T04:30:21.630118+00:00"},{"alias_kind":"pith_short_12","alias_value":"73UQAGQ2VCRR","created_at":"2026-07-05T04:30:21.630118+00:00"},{"alias_kind":"pith_short_16","alias_value":"73UQAGQ2VCRRZAR6","created_at":"2026-07-05T04:30:21.630118+00:00"},{"alias_kind":"pith_short_8","alias_value":"73UQAGQ2","created_at":"2026-07-05T04:30:21.630118+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.20872","citing_title":"In Context Learning with Vision Transformers: Case Study","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V","json":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V.json","graph_json":"https://pith.science/api/pith-number/73UQAGQ2VCRRZAR6OJRNQ2RL6V/graph.json","events_json":"https://pith.science/api/pith-number/73UQAGQ2VCRRZAR6OJRNQ2RL6V/events.json","paper":"https://pith.science/paper/73UQAGQ2"},"agent_actions":{"view_html":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V","download_json":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V.json","view_paper":"https://pith.science/paper/73UQAGQ2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.07057&json=true","fetch_graph":"https://pith.science/api/pith-number/73UQAGQ2VCRRZAR6OJRNQ2RL6V/graph.json","fetch_events":"https://pith.science/api/pith-number/73UQAGQ2VCRRZAR6OJRNQ2RL6V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V/action/storage_attestation","attest_author":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V/action/author_attestation","sign_citation":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V/action/citation_signature","submit_replication":"https://pith.science/pith/73UQAGQ2VCRRZAR6OJRNQ2RL6V/action/replication_record"}},"created_at":"2026-07-05T04:30:21.630118+00:00","updated_at":"2026-07-05T04:30:21.630118+00:00"}