{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TT77NA3GLRPHIREYWBGVNBIHZ5","short_pith_number":"pith:TT77NA3G","schema_version":"1.0","canonical_sha256":"9cfff683665c5e744498b04d568507cf7e589b6a560b2c49017395d9a456add7","source":{"kind":"arxiv","id":"2304.05659","version":1},"attestation_state":"computed","paper":{"title":"RIFormer: Keep Your Vision Backbone Effective While Removing Token Mixer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Jiahao Wang, Kai Chen, Ping Luo, Songyang Zhang, Taiqiang Wu, Xihui Liu, Yong Liu, Yujiu Yang","submitted_at":"2023-04-12T07:34:13Z","abstract_excerpt":"This paper studies how to keep a vision backbone effective while removing token mixers in its basic building blocks. Token mixers, as self-attention for vision transformers (ViTs), are intended to perform information communication between different spatial tokens but suffer from considerable computational cost and latency. However, directly removing them will lead to an incomplete model structure prior, and thus brings a significant accuracy drop. To this end, we first develop an RepIdentityFormer base on the re-parameterizing idea, to study the token mixer free model architecture. And we then"},"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":"2304.05659","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-04-12T07:34:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"93afd7704bd0fa017e09aaf66b3961094054fe6ba6eb066e022d803ff361c7dc","abstract_canon_sha256":"4278f54d4458bf3c8f18a24f1b5a8b61cb44be5317e5897ea70971aabbf7e722"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:00:21.926638Z","signature_b64":"Y/Y5fvkbBZBPpbu09FdT/d14kqzNAbTu8HLfd0VqvqK/Ya3rGo6vOotX02w46q8M4C5PTMYIXj47TewKUiSJBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9cfff683665c5e744498b04d568507cf7e589b6a560b2c49017395d9a456add7","last_reissued_at":"2026-07-05T06:00:21.926201Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:00:21.926201Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RIFormer: Keep Your Vision Backbone Effective While Removing Token Mixer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Jiahao Wang, Kai Chen, Ping Luo, Songyang Zhang, Taiqiang Wu, Xihui Liu, Yong Liu, Yujiu Yang","submitted_at":"2023-04-12T07:34:13Z","abstract_excerpt":"This paper studies how to keep a vision backbone effective while removing token mixers in its basic building blocks. Token mixers, as self-attention for vision transformers (ViTs), are intended to perform information communication between different spatial tokens but suffer from considerable computational cost and latency. However, directly removing them will lead to an incomplete model structure prior, and thus brings a significant accuracy drop. To this end, we first develop an RepIdentityFormer base on the re-parameterizing idea, to study the token mixer free model architecture. And we then"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.05659","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/2304.05659/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":"2304.05659","created_at":"2026-07-05T06:00:21.926265+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.05659v1","created_at":"2026-07-05T06:00:21.926265+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.05659","created_at":"2026-07-05T06:00:21.926265+00:00"},{"alias_kind":"pith_short_12","alias_value":"TT77NA3GLRPH","created_at":"2026-07-05T06:00:21.926265+00:00"},{"alias_kind":"pith_short_16","alias_value":"TT77NA3GLRPHIREY","created_at":"2026-07-05T06:00:21.926265+00:00"},{"alias_kind":"pith_short_8","alias_value":"TT77NA3G","created_at":"2026-07-05T06:00:21.926265+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/TT77NA3GLRPHIREYWBGVNBIHZ5","json":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5.json","graph_json":"https://pith.science/api/pith-number/TT77NA3GLRPHIREYWBGVNBIHZ5/graph.json","events_json":"https://pith.science/api/pith-number/TT77NA3GLRPHIREYWBGVNBIHZ5/events.json","paper":"https://pith.science/paper/TT77NA3G"},"agent_actions":{"view_html":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5","download_json":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5.json","view_paper":"https://pith.science/paper/TT77NA3G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.05659&json=true","fetch_graph":"https://pith.science/api/pith-number/TT77NA3GLRPHIREYWBGVNBIHZ5/graph.json","fetch_events":"https://pith.science/api/pith-number/TT77NA3GLRPHIREYWBGVNBIHZ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5/action/storage_attestation","attest_author":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5/action/author_attestation","sign_citation":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5/action/citation_signature","submit_replication":"https://pith.science/pith/TT77NA3GLRPHIREYWBGVNBIHZ5/action/replication_record"}},"created_at":"2026-07-05T06:00:21.926265+00:00","updated_at":"2026-07-05T06:00:21.926265+00:00"}