{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2V7NTIGQV5T2JI54OBRDMWUIEW","short_pith_number":"pith:2V7NTIGQ","schema_version":"1.0","canonical_sha256":"d57ed9a0d0af67a4a3bc7062365a882594d63cb67532849a037e01df1223062e","source":{"kind":"arxiv","id":"2207.08409","version":3},"attestation_state":"computed","paper":{"title":"TokenMix: Rethinking Image Mixing for Data Augmentation in Vision Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Boxiao Liu, Hang Zhou, Hongsheng Li, Jihao Liu, Yu Liu","submitted_at":"2022-07-18T07:08:29Z","abstract_excerpt":"CutMix is a popular augmentation technique commonly used for training modern convolutional and transformer vision networks. It was originally designed to encourage Convolution Neural Networks (CNNs) to focus more on an image's global context instead of local information, which greatly improves the performance of CNNs. However, we found it to have limited benefits for transformer-based architectures that naturally have a global receptive field. In this paper, we propose a novel data augmentation technique TokenMix to improve the performance of vision transformers. TokenMix mixes two images at t"},"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":"2207.08409","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-07-18T07:08:29Z","cross_cats_sorted":[],"title_canon_sha256":"b93333d53c1b162dc37a6b1522a23e035df30a06211f4c868a9952a83c40f318","abstract_canon_sha256":"e9ccb439030211deac5286c102c36141d0114ee47e99cf4b7e64a4b91aa72803"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:02:28.752862Z","signature_b64":"C3ltQ+7sVtNBsKqGBg2yilrP++FG9OZuA8yME3HulPyl/51ZZxOY4GcEOebiIi7q3zU/D3UjgCqH0/Vm/+61Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d57ed9a0d0af67a4a3bc7062365a882594d63cb67532849a037e01df1223062e","last_reissued_at":"2026-07-05T06:02:28.752373Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:02:28.752373Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TokenMix: Rethinking Image Mixing for Data Augmentation in Vision Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Boxiao Liu, Hang Zhou, Hongsheng Li, Jihao Liu, Yu Liu","submitted_at":"2022-07-18T07:08:29Z","abstract_excerpt":"CutMix is a popular augmentation technique commonly used for training modern convolutional and transformer vision networks. It was originally designed to encourage Convolution Neural Networks (CNNs) to focus more on an image's global context instead of local information, which greatly improves the performance of CNNs. However, we found it to have limited benefits for transformer-based architectures that naturally have a global receptive field. In this paper, we propose a novel data augmentation technique TokenMix to improve the performance of vision transformers. TokenMix mixes two images at t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.08409","kind":"arxiv","version":3},"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/2207.08409/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":"2207.08409","created_at":"2026-07-05T06:02:28.752444+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.08409v3","created_at":"2026-07-05T06:02:28.752444+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.08409","created_at":"2026-07-05T06:02:28.752444+00:00"},{"alias_kind":"pith_short_12","alias_value":"2V7NTIGQV5T2","created_at":"2026-07-05T06:02:28.752444+00:00"},{"alias_kind":"pith_short_16","alias_value":"2V7NTIGQV5T2JI54","created_at":"2026-07-05T06:02:28.752444+00:00"},{"alias_kind":"pith_short_8","alias_value":"2V7NTIGQ","created_at":"2026-07-05T06:02:28.752444+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/2V7NTIGQV5T2JI54OBRDMWUIEW","json":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW.json","graph_json":"https://pith.science/api/pith-number/2V7NTIGQV5T2JI54OBRDMWUIEW/graph.json","events_json":"https://pith.science/api/pith-number/2V7NTIGQV5T2JI54OBRDMWUIEW/events.json","paper":"https://pith.science/paper/2V7NTIGQ"},"agent_actions":{"view_html":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW","download_json":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW.json","view_paper":"https://pith.science/paper/2V7NTIGQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.08409&json=true","fetch_graph":"https://pith.science/api/pith-number/2V7NTIGQV5T2JI54OBRDMWUIEW/graph.json","fetch_events":"https://pith.science/api/pith-number/2V7NTIGQV5T2JI54OBRDMWUIEW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW/action/storage_attestation","attest_author":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW/action/author_attestation","sign_citation":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW/action/citation_signature","submit_replication":"https://pith.science/pith/2V7NTIGQV5T2JI54OBRDMWUIEW/action/replication_record"}},"created_at":"2026-07-05T06:02:28.752444+00:00","updated_at":"2026-07-05T06:02:28.752444+00:00"}