{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TGEPQTNFN3SHQNRQIS7MIHVIBU","short_pith_number":"pith:TGEPQTNF","schema_version":"1.0","canonical_sha256":"9988f84da56ee478363044bec41ea80d09ae57a8980f55388c1f236a05829293","source":{"kind":"arxiv","id":"2205.01580","version":1},"attestation_state":"computed","paper":{"title":"Better plain ViT baselines for ImageNet-1k","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai","submitted_at":"2022-05-03T15:54:44Z","abstract_excerpt":"It is commonly accepted that the Vision Transformer model requires sophisticated regularization techniques to excel at ImageNet-1k scale data. Surprisingly, we find this is not the case and standard data augmentation is sufficient. This note presents a few minor modifications to the original Vision Transformer (ViT) vanilla training setting that dramatically improve the performance of plain ViT models. Notably, 90 epochs of training surpass 76% top-1 accuracy in under seven hours on a TPUv3-8, similar to the classic ResNet50 baseline, and 300 epochs of training reach 80% in less than one day."},"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":"2205.01580","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-03T15:54:44Z","cross_cats_sorted":[],"title_canon_sha256":"e7ea3fa2905fd6234a5c319067fb4042c0d6833fa42ce1a69a55a2e44895fa48","abstract_canon_sha256":"8892a08920fb50791993b441d4fc3be2612440233bd214c371c1ac7de4aef4e6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:20:06.549555Z","signature_b64":"ZeCfTitjEWS7q2N+GIbXEJv9GBJN+J3TK1ZTFIykkFdyXhQFm9pV7W3+CFrFev71U7uhPFoiCBmcXttgPpP6Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9988f84da56ee478363044bec41ea80d09ae57a8980f55388c1f236a05829293","last_reissued_at":"2026-07-05T04:20:06.549130Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:20:06.549130Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Better plain ViT baselines for ImageNet-1k","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai","submitted_at":"2022-05-03T15:54:44Z","abstract_excerpt":"It is commonly accepted that the Vision Transformer model requires sophisticated regularization techniques to excel at ImageNet-1k scale data. Surprisingly, we find this is not the case and standard data augmentation is sufficient. This note presents a few minor modifications to the original Vision Transformer (ViT) vanilla training setting that dramatically improve the performance of plain ViT models. Notably, 90 epochs of training surpass 76% top-1 accuracy in under seven hours on a TPUv3-8, similar to the classic ResNet50 baseline, and 300 epochs of training reach 80% in less than one day."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.01580","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/2205.01580/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":"2205.01580","created_at":"2026-07-05T04:20:06.549196+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.01580v1","created_at":"2026-07-05T04:20:06.549196+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.01580","created_at":"2026-07-05T04:20:06.549196+00:00"},{"alias_kind":"pith_short_12","alias_value":"TGEPQTNFN3SH","created_at":"2026-07-05T04:20:06.549196+00:00"},{"alias_kind":"pith_short_16","alias_value":"TGEPQTNFN3SHQNRQ","created_at":"2026-07-05T04:20:06.549196+00:00"},{"alias_kind":"pith_short_8","alias_value":"TGEPQTNF","created_at":"2026-07-05T04:20:06.549196+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09646","citing_title":"Do Video Foundation Models Understand Intuitive Physics? A Layerwise Probing Analysis","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24789","citing_title":"Self-Supervised Contrastive Learning for Cardiac MR Sequence Classification","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07558","citing_title":"Page image classifier fine-tuned on century-spanning archives of scanned documents for further content-specific processing","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27144","citing_title":"Is an Image Also Worth 16x16=256 Superpixels? A Framework for Attentional Image Classification","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30126","citing_title":"PARCEL: Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29823","citing_title":"Quantifying and Optimizing Simplicity via Polynomial Representations","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21114","citing_title":"Page image classification for content-specific data processing","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2512.08217","citing_title":"Correction of Decoupled Weight Decay","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05052","citing_title":"Predicting spin-orbit coupling in hole spin qubit arrays with vision-transformer-based neural networks on a generalized Hubbard model","ref_index":49,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU","json":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU.json","graph_json":"https://pith.science/api/pith-number/TGEPQTNFN3SHQNRQIS7MIHVIBU/graph.json","events_json":"https://pith.science/api/pith-number/TGEPQTNFN3SHQNRQIS7MIHVIBU/events.json","paper":"https://pith.science/paper/TGEPQTNF"},"agent_actions":{"view_html":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU","download_json":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU.json","view_paper":"https://pith.science/paper/TGEPQTNF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.01580&json=true","fetch_graph":"https://pith.science/api/pith-number/TGEPQTNFN3SHQNRQIS7MIHVIBU/graph.json","fetch_events":"https://pith.science/api/pith-number/TGEPQTNFN3SHQNRQIS7MIHVIBU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU/action/storage_attestation","attest_author":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU/action/author_attestation","sign_citation":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU/action/citation_signature","submit_replication":"https://pith.science/pith/TGEPQTNFN3SHQNRQIS7MIHVIBU/action/replication_record"}},"created_at":"2026-07-05T04:20:06.549196+00:00","updated_at":"2026-07-05T04:20:06.549196+00:00"}