{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:GMDROFD7P6ZEX4AZ6DVGJPMPB4","short_pith_number":"pith:GMDROFD7","schema_version":"1.0","canonical_sha256":"330717147f7fb24bf019f0ea64bd8f0f14d125cb3030e94cbc3ab40f6d414c61","source":{"kind":"arxiv","id":"2303.14189","version":2},"attestation_state":"computed","paper":{"title":"FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anurag Ranjan, James Gabriel, Jeff Zhu, Oncel Tuzel, Pavan Kumar Anasosalu Vasu","submitted_at":"2023-03-24T17:58:32Z","abstract_excerpt":"The recent amalgamation of transformer and convolutional designs has led to steady improvements in accuracy and efficiency of the models. In this work, we introduce FastViT, a hybrid vision transformer architecture that obtains the state-of-the-art latency-accuracy trade-off. To this end, we introduce a novel token mixing operator, RepMixer, a building block of FastViT, that uses structural reparameterization to lower the memory access cost by removing skip-connections in the network. We further apply train-time overparametrization and large kernel convolutions to boost accuracy and empiricall"},"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":"2303.14189","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-03-24T17:58:32Z","cross_cats_sorted":[],"title_canon_sha256":"bc1b336d4422cabe1db047a78d716f8e01e3b1ec7d20ba4c6d1b6f51977d29f4","abstract_canon_sha256":"9d2fae527e948d16ef533fb78cab97681d3687ba5203cd01c01ab88067703337"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:42:20.332724Z","signature_b64":"13QLTkfa3XjUvnThDfjuDEGhL75I4g/TxI/Meoew/Kij01ou8Q/pFuBoqNwWi0Ma/PuUV6G7WJOyodvZOEPKBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"330717147f7fb24bf019f0ea64bd8f0f14d125cb3030e94cbc3ab40f6d414c61","last_reissued_at":"2026-07-05T06:42:20.332232Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:42:20.332232Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anurag Ranjan, James Gabriel, Jeff Zhu, Oncel Tuzel, Pavan Kumar Anasosalu Vasu","submitted_at":"2023-03-24T17:58:32Z","abstract_excerpt":"The recent amalgamation of transformer and convolutional designs has led to steady improvements in accuracy and efficiency of the models. In this work, we introduce FastViT, a hybrid vision transformer architecture that obtains the state-of-the-art latency-accuracy trade-off. To this end, we introduce a novel token mixing operator, RepMixer, a building block of FastViT, that uses structural reparameterization to lower the memory access cost by removing skip-connections in the network. We further apply train-time overparametrization and large kernel convolutions to boost accuracy and empiricall"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.14189","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/2303.14189/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":"2303.14189","created_at":"2026-07-05T06:42:20.332291+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.14189v2","created_at":"2026-07-05T06:42:20.332291+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.14189","created_at":"2026-07-05T06:42:20.332291+00:00"},{"alias_kind":"pith_short_12","alias_value":"GMDROFD7P6ZE","created_at":"2026-07-05T06:42:20.332291+00:00"},{"alias_kind":"pith_short_16","alias_value":"GMDROFD7P6ZEX4AZ","created_at":"2026-07-05T06:42:20.332291+00:00"},{"alias_kind":"pith_short_8","alias_value":"GMDROFD7","created_at":"2026-07-05T06:42:20.332291+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/GMDROFD7P6ZEX4AZ6DVGJPMPB4","json":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4.json","graph_json":"https://pith.science/api/pith-number/GMDROFD7P6ZEX4AZ6DVGJPMPB4/graph.json","events_json":"https://pith.science/api/pith-number/GMDROFD7P6ZEX4AZ6DVGJPMPB4/events.json","paper":"https://pith.science/paper/GMDROFD7"},"agent_actions":{"view_html":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4","download_json":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4.json","view_paper":"https://pith.science/paper/GMDROFD7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.14189&json=true","fetch_graph":"https://pith.science/api/pith-number/GMDROFD7P6ZEX4AZ6DVGJPMPB4/graph.json","fetch_events":"https://pith.science/api/pith-number/GMDROFD7P6ZEX4AZ6DVGJPMPB4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4/action/storage_attestation","attest_author":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4/action/author_attestation","sign_citation":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4/action/citation_signature","submit_replication":"https://pith.science/pith/GMDROFD7P6ZEX4AZ6DVGJPMPB4/action/replication_record"}},"created_at":"2026-07-05T06:42:20.332291+00:00","updated_at":"2026-07-05T06:42:20.332291+00:00"}