{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:PSMB3TJ3VWFDPA6UFTIGL26RTC","short_pith_number":"pith:PSMB3TJ3","schema_version":"1.0","canonical_sha256":"7c981dcd3bad8a3783d42cd065ebd1989c64e7be28e40ed148269b6f197097a7","source":{"kind":"arxiv","id":"2206.10555","version":2},"attestation_state":"computed","paper":{"title":"LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jianhui Liu, Jiaya Jia, Xiangyu Zhang, Xiaojuan Qi, Yukang Chen","submitted_at":"2022-06-21T17:35:57Z","abstract_excerpt":"Recent advance in 2D CNNs has revealed that large kernels are important. However, when directly applying large convolutional kernels in 3D CNNs, severe difficulties are met, where those successful module designs in 2D become surprisingly ineffective on 3D networks, including the popular depth-wise convolution. To address this vital challenge, we instead propose the spatial-wise partition convolution and its large-kernel module. As a result, it avoids the optimization and efficiency issues of naive 3D large kernels. Our large-kernel 3D CNN network, LargeKernel3D, yields notable improvement in 3"},"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":"2206.10555","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-21T17:35:57Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2cd87eb4b0b1b10020b59dec5271d37e281e2f04f3816afddc24abe3ba22ad21","abstract_canon_sha256":"eca262c5f8722dd9cdb34ee54a738c07844d8f6bb99349bc98d279f338b179f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:53:32.389188Z","signature_b64":"19CvGcn9y0RW4D++c+v+aAFE6ne0yAXWFYDPo4/ttEYCuZ8VJhPk22xngcZSJ6nDRu37vFugpp0EOm7wvqUXBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7c981dcd3bad8a3783d42cd065ebd1989c64e7be28e40ed148269b6f197097a7","last_reissued_at":"2026-07-05T05:53:32.388755Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:53:32.388755Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jianhui Liu, Jiaya Jia, Xiangyu Zhang, Xiaojuan Qi, Yukang Chen","submitted_at":"2022-06-21T17:35:57Z","abstract_excerpt":"Recent advance in 2D CNNs has revealed that large kernels are important. However, when directly applying large convolutional kernels in 3D CNNs, severe difficulties are met, where those successful module designs in 2D become surprisingly ineffective on 3D networks, including the popular depth-wise convolution. To address this vital challenge, we instead propose the spatial-wise partition convolution and its large-kernel module. As a result, it avoids the optimization and efficiency issues of naive 3D large kernels. Our large-kernel 3D CNN network, LargeKernel3D, yields notable improvement in 3"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.10555","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/2206.10555/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":"2206.10555","created_at":"2026-07-05T05:53:32.388812+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.10555v2","created_at":"2026-07-05T05:53:32.388812+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.10555","created_at":"2026-07-05T05:53:32.388812+00:00"},{"alias_kind":"pith_short_12","alias_value":"PSMB3TJ3VWFD","created_at":"2026-07-05T05:53:32.388812+00:00"},{"alias_kind":"pith_short_16","alias_value":"PSMB3TJ3VWFDPA6U","created_at":"2026-07-05T05:53:32.388812+00:00"},{"alias_kind":"pith_short_8","alias_value":"PSMB3TJ3","created_at":"2026-07-05T05:53:32.388812+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.10706","citing_title":"RelFlexformer: Efficient Attention 3D-Transformers for Integrable Relative Positional Encodings","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC","json":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC.json","graph_json":"https://pith.science/api/pith-number/PSMB3TJ3VWFDPA6UFTIGL26RTC/graph.json","events_json":"https://pith.science/api/pith-number/PSMB3TJ3VWFDPA6UFTIGL26RTC/events.json","paper":"https://pith.science/paper/PSMB3TJ3"},"agent_actions":{"view_html":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC","download_json":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC.json","view_paper":"https://pith.science/paper/PSMB3TJ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.10555&json=true","fetch_graph":"https://pith.science/api/pith-number/PSMB3TJ3VWFDPA6UFTIGL26RTC/graph.json","fetch_events":"https://pith.science/api/pith-number/PSMB3TJ3VWFDPA6UFTIGL26RTC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC/action/storage_attestation","attest_author":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC/action/author_attestation","sign_citation":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC/action/citation_signature","submit_replication":"https://pith.science/pith/PSMB3TJ3VWFDPA6UFTIGL26RTC/action/replication_record"}},"created_at":"2026-07-05T05:53:32.388812+00:00","updated_at":"2026-07-05T05:53:32.388812+00:00"}