{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:DIKUXXJKGPAIFVWA4FVJLBR5JI","short_pith_number":"pith:DIKUXXJK","schema_version":"1.0","canonical_sha256":"1a154bdd2a33c082d6c0e16a95863d4a24d5772e28cd1fcdb9b8a3f86a67f969","source":{"kind":"arxiv","id":"2110.09408","version":3},"attestation_state":"computed","paper":{"title":"HRFormer: High-Resolution Transformer for Dense Prediction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chao Zhang, Jingdong Wang, Lang Huang, Rao Fu, Weihong Lin, Xilin Chen, Yuhui Yuan","submitted_at":"2021-10-18T15:37:58Z","abstract_excerpt":"We present a High-Resolution Transformer (HRFormer) that learns high-resolution representations for dense prediction tasks, in contrast to the original Vision Transformer that produces low-resolution representations and has high memory and computational cost. We take advantage of the multi-resolution parallel design introduced in high-resolution convolutional networks (HRNet), along with local-window self-attention that performs self-attention over small non-overlapping image windows, for improving the memory and computation efficiency. In addition, we introduce a convolution into the FFN to e"},"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":"2110.09408","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-18T15:37:58Z","cross_cats_sorted":[],"title_canon_sha256":"36424d400b0543f57c035d7689b09dbf5bf40a8b6b9f487d88630d635f249b5c","abstract_canon_sha256":"7a810498a31874d833ad779db1e673d453835c9ef21c6fc617815b4cff8042b0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:29:52.810390Z","signature_b64":"wCluTJCYRfnEE9Ru3TX4ZyJZb7W2pNrZGW7ImJ+93veO0JrJX0r4pyJ/vD4F83ykZNu7xNF4sxSRQ5OiOJgiCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a154bdd2a33c082d6c0e16a95863d4a24d5772e28cd1fcdb9b8a3f86a67f969","last_reissued_at":"2026-07-05T03:29:52.809887Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:29:52.809887Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HRFormer: High-Resolution Transformer for Dense Prediction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chao Zhang, Jingdong Wang, Lang Huang, Rao Fu, Weihong Lin, Xilin Chen, Yuhui Yuan","submitted_at":"2021-10-18T15:37:58Z","abstract_excerpt":"We present a High-Resolution Transformer (HRFormer) that learns high-resolution representations for dense prediction tasks, in contrast to the original Vision Transformer that produces low-resolution representations and has high memory and computational cost. We take advantage of the multi-resolution parallel design introduced in high-resolution convolutional networks (HRNet), along with local-window self-attention that performs self-attention over small non-overlapping image windows, for improving the memory and computation efficiency. In addition, we introduce a convolution into the FFN to e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.09408","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/2110.09408/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":"2110.09408","created_at":"2026-07-05T03:29:52.809948+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.09408v3","created_at":"2026-07-05T03:29:52.809948+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.09408","created_at":"2026-07-05T03:29:52.809948+00:00"},{"alias_kind":"pith_short_12","alias_value":"DIKUXXJKGPAI","created_at":"2026-07-05T03:29:52.809948+00:00"},{"alias_kind":"pith_short_16","alias_value":"DIKUXXJKGPAIFVWA","created_at":"2026-07-05T03:29:52.809948+00:00"},{"alias_kind":"pith_short_8","alias_value":"DIKUXXJK","created_at":"2026-07-05T03:29:52.809948+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.19527","citing_title":"Dual-Prompt CLIP with Hybrid Visual Encoders for Occluded Person Re-Identification","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18091","citing_title":"Accelerating Vision Transformers with Adaptive Patch Sizes","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI","json":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI.json","graph_json":"https://pith.science/api/pith-number/DIKUXXJKGPAIFVWA4FVJLBR5JI/graph.json","events_json":"https://pith.science/api/pith-number/DIKUXXJKGPAIFVWA4FVJLBR5JI/events.json","paper":"https://pith.science/paper/DIKUXXJK"},"agent_actions":{"view_html":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI","download_json":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI.json","view_paper":"https://pith.science/paper/DIKUXXJK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.09408&json=true","fetch_graph":"https://pith.science/api/pith-number/DIKUXXJKGPAIFVWA4FVJLBR5JI/graph.json","fetch_events":"https://pith.science/api/pith-number/DIKUXXJKGPAIFVWA4FVJLBR5JI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI/action/storage_attestation","attest_author":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI/action/author_attestation","sign_citation":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI/action/citation_signature","submit_replication":"https://pith.science/pith/DIKUXXJKGPAIFVWA4FVJLBR5JI/action/replication_record"}},"created_at":"2026-07-05T03:29:52.809948+00:00","updated_at":"2026-07-05T03:29:52.809948+00:00"}