{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:CBEVNJOO3C7H2JK2VVWMHVKH2S","short_pith_number":"pith:CBEVNJOO","schema_version":"1.0","canonical_sha256":"104956a5ced8be7d255aad6cc3d547d4b5ce9923b8316e9fa76fdf9fe4e3d4d2","source":{"kind":"arxiv","id":"2203.06108","version":2},"attestation_state":"computed","paper":{"title":"Active Token Mixer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cuiling Lan, Guoqiang Wei, Yan Lu, Zhibo Chen, Zhizheng Zhang","submitted_at":"2022-03-11T17:29:54Z","abstract_excerpt":"The three existing dominant network families, i.e., CNNs, Transformers, and MLPs, differ from each other mainly in the ways of fusing spatial contextual information, leaving designing more effective token-mixing mechanisms at the core of backbone architecture development. In this work, we propose an innovative token-mixer, dubbed Active Token Mixer (ATM), to actively incorporate flexible contextual information distributed across different channels from other tokens into the given query token. This fundamental operator actively predicts where to capture useful contexts and learns how to fuse th"},"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":"2203.06108","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-03-11T17:29:54Z","cross_cats_sorted":[],"title_canon_sha256":"4e07d9ab3a3b28fd8520e72279c5f12324031635791211ef22f867d2a840806e","abstract_canon_sha256":"34eed38429dd7d10cb20d2fd0715ffe2a1154d098e30281394eda3113a4635be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:27:47.360564Z","signature_b64":"mag24y1dY1AlIerNtAWK1OFOhXvqO/5Lnnfk1QUb95zatVp36UtkGjok4y+vzR7pILvfvBn6e9RODrC0uxN+CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"104956a5ced8be7d255aad6cc3d547d4b5ce9923b8316e9fa76fdf9fe4e3d4d2","last_reissued_at":"2026-07-05T05:27:47.360064Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:27:47.360064Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Active Token Mixer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cuiling Lan, Guoqiang Wei, Yan Lu, Zhibo Chen, Zhizheng Zhang","submitted_at":"2022-03-11T17:29:54Z","abstract_excerpt":"The three existing dominant network families, i.e., CNNs, Transformers, and MLPs, differ from each other mainly in the ways of fusing spatial contextual information, leaving designing more effective token-mixing mechanisms at the core of backbone architecture development. In this work, we propose an innovative token-mixer, dubbed Active Token Mixer (ATM), to actively incorporate flexible contextual information distributed across different channels from other tokens into the given query token. This fundamental operator actively predicts where to capture useful contexts and learns how to fuse th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.06108","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/2203.06108/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":"2203.06108","created_at":"2026-07-05T05:27:47.360124+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.06108v2","created_at":"2026-07-05T05:27:47.360124+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.06108","created_at":"2026-07-05T05:27:47.360124+00:00"},{"alias_kind":"pith_short_12","alias_value":"CBEVNJOO3C7H","created_at":"2026-07-05T05:27:47.360124+00:00"},{"alias_kind":"pith_short_16","alias_value":"CBEVNJOO3C7H2JK2","created_at":"2026-07-05T05:27:47.360124+00:00"},{"alias_kind":"pith_short_8","alias_value":"CBEVNJOO","created_at":"2026-07-05T05:27:47.360124+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.18995","citing_title":"MVFormer: Diversifying Feature Normalization and Token Mixing for Efficient Vision Transformers","ref_index":51,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S","json":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S.json","graph_json":"https://pith.science/api/pith-number/CBEVNJOO3C7H2JK2VVWMHVKH2S/graph.json","events_json":"https://pith.science/api/pith-number/CBEVNJOO3C7H2JK2VVWMHVKH2S/events.json","paper":"https://pith.science/paper/CBEVNJOO"},"agent_actions":{"view_html":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S","download_json":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S.json","view_paper":"https://pith.science/paper/CBEVNJOO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.06108&json=true","fetch_graph":"https://pith.science/api/pith-number/CBEVNJOO3C7H2JK2VVWMHVKH2S/graph.json","fetch_events":"https://pith.science/api/pith-number/CBEVNJOO3C7H2JK2VVWMHVKH2S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S/action/storage_attestation","attest_author":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S/action/author_attestation","sign_citation":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S/action/citation_signature","submit_replication":"https://pith.science/pith/CBEVNJOO3C7H2JK2VVWMHVKH2S/action/replication_record"}},"created_at":"2026-07-05T05:27:47.360124+00:00","updated_at":"2026-07-05T05:27:47.360124+00:00"}