{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:A6W5WHPX3MDLBPDOTULMTSV2FR","short_pith_number":"pith:A6W5WHPX","schema_version":"1.0","canonical_sha256":"07addb1df7db06b0bc6e9d16c9caba2c7382d7c861dedd45d7c9199d8766c734","source":{"kind":"arxiv","id":"2410.14874","version":2},"attestation_state":"computed","paper":{"title":"Improving Vision Transformers by Overlapping Heads in Multi-Head Self-Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bo Luo, Guanghui Wang, Tianxiao Zhang","submitted_at":"2024-10-18T21:41:55Z","abstract_excerpt":"Vision Transformers have made remarkable progress in recent years, achieving state-of-the-art performance in most vision tasks. A key component of this success is due to the introduction of the Multi-Head Self-Attention (MHSA) module, which enables each head to learn different representations by applying the attention mechanism independently. In this paper, we empirically demonstrate that Vision Transformers can be further enhanced by overlapping the heads in MHSA. We introduce Multi-Overlapped-Head Self-Attention (MOHSA), where heads are overlapped with their two adjacent heads for queries, k"},"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":"2410.14874","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-10-18T21:41:55Z","cross_cats_sorted":[],"title_canon_sha256":"406f1a9022d5b9ef374f856589edfc52379919162b50b328715d5118a045b4e0","abstract_canon_sha256":"554e883652c64a5159a1063c25ae0c02515899653fbb118c92face337e0949be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:23.076541Z","signature_b64":"DsYmnBD3JwyUciA+8Y0slD4WAJcN2h52moRIernC1FRpF/UImPN8FIIckN6YxbfI3X2Yx860fole55Al6qEzCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"07addb1df7db06b0bc6e9d16c9caba2c7382d7c861dedd45d7c9199d8766c734","last_reissued_at":"2026-07-05T10:08:23.076119Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:23.076119Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Vision Transformers by Overlapping Heads in Multi-Head Self-Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bo Luo, Guanghui Wang, Tianxiao Zhang","submitted_at":"2024-10-18T21:41:55Z","abstract_excerpt":"Vision Transformers have made remarkable progress in recent years, achieving state-of-the-art performance in most vision tasks. A key component of this success is due to the introduction of the Multi-Head Self-Attention (MHSA) module, which enables each head to learn different representations by applying the attention mechanism independently. In this paper, we empirically demonstrate that Vision Transformers can be further enhanced by overlapping the heads in MHSA. We introduce Multi-Overlapped-Head Self-Attention (MOHSA), where heads are overlapped with their two adjacent heads for queries, k"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.14874","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/2410.14874/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":"2410.14874","created_at":"2026-07-05T10:08:23.076174+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.14874v2","created_at":"2026-07-05T10:08:23.076174+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.14874","created_at":"2026-07-05T10:08:23.076174+00:00"},{"alias_kind":"pith_short_12","alias_value":"A6W5WHPX3MDL","created_at":"2026-07-05T10:08:23.076174+00:00"},{"alias_kind":"pith_short_16","alias_value":"A6W5WHPX3MDLBPDO","created_at":"2026-07-05T10:08:23.076174+00:00"},{"alias_kind":"pith_short_8","alias_value":"A6W5WHPX","created_at":"2026-07-05T10:08:23.076174+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/A6W5WHPX3MDLBPDOTULMTSV2FR","json":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR.json","graph_json":"https://pith.science/api/pith-number/A6W5WHPX3MDLBPDOTULMTSV2FR/graph.json","events_json":"https://pith.science/api/pith-number/A6W5WHPX3MDLBPDOTULMTSV2FR/events.json","paper":"https://pith.science/paper/A6W5WHPX"},"agent_actions":{"view_html":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR","download_json":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR.json","view_paper":"https://pith.science/paper/A6W5WHPX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.14874&json=true","fetch_graph":"https://pith.science/api/pith-number/A6W5WHPX3MDLBPDOTULMTSV2FR/graph.json","fetch_events":"https://pith.science/api/pith-number/A6W5WHPX3MDLBPDOTULMTSV2FR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR/action/storage_attestation","attest_author":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR/action/author_attestation","sign_citation":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR/action/citation_signature","submit_replication":"https://pith.science/pith/A6W5WHPX3MDLBPDOTULMTSV2FR/action/replication_record"}},"created_at":"2026-07-05T10:08:23.076174+00:00","updated_at":"2026-07-05T10:08:23.076174+00:00"}