{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:AEXEVRSAHEZE62RCO24N56YVBV","short_pith_number":"pith:AEXEVRSA","schema_version":"1.0","canonical_sha256":"012e4ac64039324f6a2276b8defb150d5bcce1648860c827244d8a09d16b2f9a","source":{"kind":"arxiv","id":"2305.14952","version":2},"attestation_state":"computed","paper":{"title":"Focus Your Attention (with Adaptive IIR Filters)","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.SP"],"primary_cat":"cs.LG","authors_text":"Itamar Zimerman, Lior Wolf, Shahar Lutati","submitted_at":"2023-05-24T09:42:30Z","abstract_excerpt":"We present a new layer in which dynamic (i.e.,input-dependent) Infinite Impulse Response (IIR) filters of order two are used to process the input sequence prior to applying conventional attention. The input is split into chunks, and the coefficients of these filters are determined based on previous chunks to maintain causality. Despite their relatively low order, the causal adaptive filters are shown to focus attention on the relevant sequence elements. The new layer is grounded in control theory, and is shown to generalize diagonal state-space layers. The layer performs on-par with state-of-t"},"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":"2305.14952","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-05-24T09:42:30Z","cross_cats_sorted":["eess.SP"],"title_canon_sha256":"6e85ff18dd7b1455882133062a620cee62cb32d641e755c67e7005039b847068","abstract_canon_sha256":"da5d61a4756d4b30ef5e42ee797f90ac70aad66b49fe3b462defc93fe2145b36"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:02:02.238936Z","signature_b64":"q5EhMUi0tqPQTI2PG2s8vz2H7WBF/JVx9dUBckT3PO/4lFxyHV69QeYMtEcY3pABWpNVv0w6WJlT7r4ElOBvBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"012e4ac64039324f6a2276b8defb150d5bcce1648860c827244d8a09d16b2f9a","last_reissued_at":"2026-07-05T07:02:02.238420Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:02:02.238420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Focus Your Attention (with Adaptive IIR Filters)","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.SP"],"primary_cat":"cs.LG","authors_text":"Itamar Zimerman, Lior Wolf, Shahar Lutati","submitted_at":"2023-05-24T09:42:30Z","abstract_excerpt":"We present a new layer in which dynamic (i.e.,input-dependent) Infinite Impulse Response (IIR) filters of order two are used to process the input sequence prior to applying conventional attention. The input is split into chunks, and the coefficients of these filters are determined based on previous chunks to maintain causality. Despite their relatively low order, the causal adaptive filters are shown to focus attention on the relevant sequence elements. The new layer is grounded in control theory, and is shown to generalize diagonal state-space layers. The layer performs on-par with state-of-t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14952","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/2305.14952/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":"2305.14952","created_at":"2026-07-05T07:02:02.238482+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14952v2","created_at":"2026-07-05T07:02:02.238482+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14952","created_at":"2026-07-05T07:02:02.238482+00:00"},{"alias_kind":"pith_short_12","alias_value":"AEXEVRSAHEZE","created_at":"2026-07-05T07:02:02.238482+00:00"},{"alias_kind":"pith_short_16","alias_value":"AEXEVRSAHEZE62RC","created_at":"2026-07-05T07:02:02.238482+00:00"},{"alias_kind":"pith_short_8","alias_value":"AEXEVRSA","created_at":"2026-07-05T07:02:02.238482+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2312.00752","citing_title":"Mamba: Linear-Time Sequence Modeling with Selective State Spaces","ref_index":69,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV","json":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV.json","graph_json":"https://pith.science/api/pith-number/AEXEVRSAHEZE62RCO24N56YVBV/graph.json","events_json":"https://pith.science/api/pith-number/AEXEVRSAHEZE62RCO24N56YVBV/events.json","paper":"https://pith.science/paper/AEXEVRSA"},"agent_actions":{"view_html":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV","download_json":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV.json","view_paper":"https://pith.science/paper/AEXEVRSA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14952&json=true","fetch_graph":"https://pith.science/api/pith-number/AEXEVRSAHEZE62RCO24N56YVBV/graph.json","fetch_events":"https://pith.science/api/pith-number/AEXEVRSAHEZE62RCO24N56YVBV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV/action/storage_attestation","attest_author":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV/action/author_attestation","sign_citation":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV/action/citation_signature","submit_replication":"https://pith.science/pith/AEXEVRSAHEZE62RCO24N56YVBV/action/replication_record"}},"created_at":"2026-07-05T07:02:02.238482+00:00","updated_at":"2026-07-05T07:02:02.238482+00:00"}