{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:HCDDGH7TDLTOTQQUYIEPQC2NBV","short_pith_number":"pith:HCDDGH7T","schema_version":"1.0","canonical_sha256":"3886331ff31ae6e9c214c208f80b4d0d451d0629116fdf4226c3d52708d68938","source":{"kind":"arxiv","id":"2306.10875","version":1},"attestation_state":"computed","paper":{"title":"Vision Transformer with Attention Map Hallucination and FFN Compaction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongliang He, Fu Li, Haiyang Xu, Jingdong Wang, Zhichao Zhou","submitted_at":"2023-06-19T12:08:55Z","abstract_excerpt":"Vision Transformer(ViT) is now dominating many vision tasks. The drawback of quadratic complexity of its token-wise multi-head self-attention (MHSA), is extensively addressed via either token sparsification or dimension reduction (in spatial or channel). However, the therein redundancy of MHSA is usually overlooked and so is the feed-forward network (FFN). To this end, we propose attention map hallucination and FFN compaction to fill in the blank. Specifically, we observe similar attention maps exist in vanilla ViT and propose to hallucinate half of the attention maps from the rest with much c"},"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":"2306.10875","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-06-19T12:08:55Z","cross_cats_sorted":[],"title_canon_sha256":"90a59aa1771f519412a0ba0ef3619b0ed2c40fbb6027701f0d723d0cb10c31e9","abstract_canon_sha256":"143f607b12f86eab43332a4db7ceea8aa440135785ee0ceaf6d13292f5397672"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:22:12.907588Z","signature_b64":"zL8zDzPko5u8Z/37hoV1Iujc6tU5rIMV/YHtgTZjk/gFDB/sNdr2n3vNZuFUqxe14y0/v+kyQEgR8p60O8kwBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3886331ff31ae6e9c214c208f80b4d0d451d0629116fdf4226c3d52708d68938","last_reissued_at":"2026-07-05T06:22:12.907133Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:22:12.907133Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Vision Transformer with Attention Map Hallucination and FFN Compaction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongliang He, Fu Li, Haiyang Xu, Jingdong Wang, Zhichao Zhou","submitted_at":"2023-06-19T12:08:55Z","abstract_excerpt":"Vision Transformer(ViT) is now dominating many vision tasks. The drawback of quadratic complexity of its token-wise multi-head self-attention (MHSA), is extensively addressed via either token sparsification or dimension reduction (in spatial or channel). However, the therein redundancy of MHSA is usually overlooked and so is the feed-forward network (FFN). To this end, we propose attention map hallucination and FFN compaction to fill in the blank. Specifically, we observe similar attention maps exist in vanilla ViT and propose to hallucinate half of the attention maps from the rest with much c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.10875","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/2306.10875/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":"2306.10875","created_at":"2026-07-05T06:22:12.907198+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.10875v1","created_at":"2026-07-05T06:22:12.907198+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.10875","created_at":"2026-07-05T06:22:12.907198+00:00"},{"alias_kind":"pith_short_12","alias_value":"HCDDGH7TDLTO","created_at":"2026-07-05T06:22:12.907198+00:00"},{"alias_kind":"pith_short_16","alias_value":"HCDDGH7TDLTOTQQU","created_at":"2026-07-05T06:22:12.907198+00:00"},{"alias_kind":"pith_short_8","alias_value":"HCDDGH7T","created_at":"2026-07-05T06:22:12.907198+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22901","citing_title":"Explainable AI in Speaker Recognition -- Attention Map Visualisation and Evaluation","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV","json":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV.json","graph_json":"https://pith.science/api/pith-number/HCDDGH7TDLTOTQQUYIEPQC2NBV/graph.json","events_json":"https://pith.science/api/pith-number/HCDDGH7TDLTOTQQUYIEPQC2NBV/events.json","paper":"https://pith.science/paper/HCDDGH7T"},"agent_actions":{"view_html":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV","download_json":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV.json","view_paper":"https://pith.science/paper/HCDDGH7T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.10875&json=true","fetch_graph":"https://pith.science/api/pith-number/HCDDGH7TDLTOTQQUYIEPQC2NBV/graph.json","fetch_events":"https://pith.science/api/pith-number/HCDDGH7TDLTOTQQUYIEPQC2NBV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV/action/storage_attestation","attest_author":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV/action/author_attestation","sign_citation":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV/action/citation_signature","submit_replication":"https://pith.science/pith/HCDDGH7TDLTOTQQUYIEPQC2NBV/action/replication_record"}},"created_at":"2026-07-05T06:22:12.907198+00:00","updated_at":"2026-07-05T06:22:12.907198+00:00"}