{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:JTSDYR3KPZBVY77PZ3GKAPJSNP","short_pith_number":"pith:JTSDYR3K","schema_version":"1.0","canonical_sha256":"4ce43c476a7e435c7fefcecca03d326be6b2e5fe4b4cefaa4d85fba86a369844","source":{"kind":"arxiv","id":"2104.11227","version":1},"attestation_state":"computed","paper":{"title":"Multiscale Vision Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Xiong, Christoph Feichtenhofer, Haoqi Fan, Jitendra Malik, Karttikeya Mangalam, Yanghao Li, Zhicheng Yan","submitted_at":"2021-04-22T17:59:45Z","abstract_excerpt":"We present Multiscale Vision Transformers (MViT) for video and image recognition, by connecting the seminal idea of multiscale feature hierarchies with transformer models. Multiscale Transformers have several channel-resolution scale stages. Starting from the input resolution and a small channel dimension, the stages hierarchically expand the channel capacity while reducing the spatial resolution. This creates a multiscale pyramid of features with early layers operating at high spatial resolution to model simple low-level visual information, and deeper layers at spatially coarse, but complex, "},"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":"2104.11227","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-04-22T17:59:45Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"cdf38f1a11e85be40e3aaf532b2aba030aabf62512dfa2f9be9ba4f1c7e9dd1e","abstract_canon_sha256":"51c718ab3873d6f0a25ba2d15d9791365ed85fc3116e52b35238067f9b6b58a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:34:19.009170Z","signature_b64":"aDCDOphtMvr8wAbxn4xuVsW6KphwZKDJUCzjazvcuK0JiiKYaynNkjkzPNj50on7KB+7d0KqOUMuk/RP784LBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ce43c476a7e435c7fefcecca03d326be6b2e5fe4b4cefaa4d85fba86a369844","last_reissued_at":"2026-07-05T02:34:19.008628Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:34:19.008628Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multiscale Vision Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Xiong, Christoph Feichtenhofer, Haoqi Fan, Jitendra Malik, Karttikeya Mangalam, Yanghao Li, Zhicheng Yan","submitted_at":"2021-04-22T17:59:45Z","abstract_excerpt":"We present Multiscale Vision Transformers (MViT) for video and image recognition, by connecting the seminal idea of multiscale feature hierarchies with transformer models. Multiscale Transformers have several channel-resolution scale stages. Starting from the input resolution and a small channel dimension, the stages hierarchically expand the channel capacity while reducing the spatial resolution. This creates a multiscale pyramid of features with early layers operating at high spatial resolution to model simple low-level visual information, and deeper layers at spatially coarse, but complex, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.11227","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/2104.11227/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":"2104.11227","created_at":"2026-07-05T02:34:19.008687+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.11227v1","created_at":"2026-07-05T02:34:19.008687+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.11227","created_at":"2026-07-05T02:34:19.008687+00:00"},{"alias_kind":"pith_short_12","alias_value":"JTSDYR3KPZBV","created_at":"2026-07-05T02:34:19.008687+00:00"},{"alias_kind":"pith_short_16","alias_value":"JTSDYR3KPZBVY77P","created_at":"2026-07-05T02:34:19.008687+00:00"},{"alias_kind":"pith_short_8","alias_value":"JTSDYR3K","created_at":"2026-07-05T02:34:19.008687+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23118","citing_title":"LUMINA-26: Low-Light Understanding for Modeling and Interpreting Night-time Actions","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2502.05564","citing_title":"TabICL: A Tabular Foundation Model for In-Context Learning on Large Data","ref_index":146,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20760","citing_title":"Exploring High-Order Self-Similarity for Video Understanding","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP","json":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP.json","graph_json":"https://pith.science/api/pith-number/JTSDYR3KPZBVY77PZ3GKAPJSNP/graph.json","events_json":"https://pith.science/api/pith-number/JTSDYR3KPZBVY77PZ3GKAPJSNP/events.json","paper":"https://pith.science/paper/JTSDYR3K"},"agent_actions":{"view_html":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP","download_json":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP.json","view_paper":"https://pith.science/paper/JTSDYR3K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.11227&json=true","fetch_graph":"https://pith.science/api/pith-number/JTSDYR3KPZBVY77PZ3GKAPJSNP/graph.json","fetch_events":"https://pith.science/api/pith-number/JTSDYR3KPZBVY77PZ3GKAPJSNP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP/action/storage_attestation","attest_author":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP/action/author_attestation","sign_citation":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP/action/citation_signature","submit_replication":"https://pith.science/pith/JTSDYR3KPZBVY77PZ3GKAPJSNP/action/replication_record"}},"created_at":"2026-07-05T02:34:19.008687+00:00","updated_at":"2026-07-05T02:34:19.008687+00:00"}