{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6I2AQVF3AFMBPS7F53A7NJ3ZK3","short_pith_number":"pith:6I2AQVF3","schema_version":"1.0","canonical_sha256":"f2340854bb015817cbe5eec1f6a77956eaf71b0ebc0d917b4901e423d41306bc","source":{"kind":"arxiv","id":"2411.04967","version":1},"attestation_state":"computed","paper":{"title":"AsCAN: Asymmetric Convolution-Attention Networks for Efficient Recognition and Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aliaksandr Siarohin, Anil Kag, Huseyin Coskun, Jian Ren, Jierun Chen, Junli Cao, Sergey Tulyakov, Willi Menapace","submitted_at":"2024-11-07T18:43:17Z","abstract_excerpt":"Neural network architecture design requires making many crucial decisions. The common desiderata is that similar decisions, with little modifications, can be reused in a variety of tasks and applications. To satisfy that, architectures must provide promising latency and performance trade-offs, support a variety of tasks, scale efficiently with respect to the amounts of data and compute, leverage available data from other tasks, and efficiently support various hardware. To this end, we introduce AsCAN -- a hybrid architecture, combining both convolutional and transformer blocks. We revisit the "},"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":"2411.04967","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-07T18:43:17Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c0a0ded747532f1073ee24a74c265f54821428102c1ed5ebb7b71e7f8be46b44","abstract_canon_sha256":"651eca7b1a3e8ab6c06b7a4edbbba51b325ca31ef580c122fbcf0c527692b7f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:32:36.969366Z","signature_b64":"a6b844LwAFGaHJReFqENAOc2f7MzAFV2ibEzpl0ZzpT9mLi1Eoy0bElN2D6lG1fpDe5byCbTrvWcFjRmz1f6DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f2340854bb015817cbe5eec1f6a77956eaf71b0ebc0d917b4901e423d41306bc","last_reissued_at":"2026-07-05T09:32:36.968745Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:32:36.968745Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AsCAN: Asymmetric Convolution-Attention Networks for Efficient Recognition and Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aliaksandr Siarohin, Anil Kag, Huseyin Coskun, Jian Ren, Jierun Chen, Junli Cao, Sergey Tulyakov, Willi Menapace","submitted_at":"2024-11-07T18:43:17Z","abstract_excerpt":"Neural network architecture design requires making many crucial decisions. The common desiderata is that similar decisions, with little modifications, can be reused in a variety of tasks and applications. To satisfy that, architectures must provide promising latency and performance trade-offs, support a variety of tasks, scale efficiently with respect to the amounts of data and compute, leverage available data from other tasks, and efficiently support various hardware. To this end, we introduce AsCAN -- a hybrid architecture, combining both convolutional and transformer blocks. We revisit the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04967","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/2411.04967/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":"2411.04967","created_at":"2026-07-05T09:32:36.968820+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.04967v1","created_at":"2026-07-05T09:32:36.968820+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04967","created_at":"2026-07-05T09:32:36.968820+00:00"},{"alias_kind":"pith_short_12","alias_value":"6I2AQVF3AFMB","created_at":"2026-07-05T09:32:36.968820+00:00"},{"alias_kind":"pith_short_16","alias_value":"6I2AQVF3AFMBPS7F","created_at":"2026-07-05T09:32:36.968820+00:00"},{"alias_kind":"pith_short_8","alias_value":"6I2AQVF3","created_at":"2026-07-05T09:32:36.968820+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.09619","citing_title":"SnapGen: Taming High-Resolution Text-to-Image Models for Mobile Devices with Efficient Architectures and Training","ref_index":32,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3","json":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3.json","graph_json":"https://pith.science/api/pith-number/6I2AQVF3AFMBPS7F53A7NJ3ZK3/graph.json","events_json":"https://pith.science/api/pith-number/6I2AQVF3AFMBPS7F53A7NJ3ZK3/events.json","paper":"https://pith.science/paper/6I2AQVF3"},"agent_actions":{"view_html":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3","download_json":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3.json","view_paper":"https://pith.science/paper/6I2AQVF3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.04967&json=true","fetch_graph":"https://pith.science/api/pith-number/6I2AQVF3AFMBPS7F53A7NJ3ZK3/graph.json","fetch_events":"https://pith.science/api/pith-number/6I2AQVF3AFMBPS7F53A7NJ3ZK3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3/action/storage_attestation","attest_author":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3/action/author_attestation","sign_citation":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3/action/citation_signature","submit_replication":"https://pith.science/pith/6I2AQVF3AFMBPS7F53A7NJ3ZK3/action/replication_record"}},"created_at":"2026-07-05T09:32:36.968820+00:00","updated_at":"2026-07-05T09:32:36.968820+00:00"}