{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:N5Z2NKHD3YN7QPATCGFQIDHZQI","short_pith_number":"pith:N5Z2NKHD","schema_version":"1.0","canonical_sha256":"6f73a6a8e3de1bf83c13118b040cf982031b13d10d87e25e113973d2bc96e47c","source":{"kind":"arxiv","id":"2409.02097","version":3},"attestation_state":"computed","paper":{"title":"LinFusion: 1 GPU, 1 Minute, 16K Image","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Songhua Liu, Weihao Yu, Xinchao Wang, Zhenxiong Tan","submitted_at":"2024-09-03T17:54:39Z","abstract_excerpt":"Modern diffusion models, particularly those utilizing a Transformer-based UNet for denoising, rely heavily on self-attention operations to manage complex spatial relationships, thus achieving impressive generation performance. However, this existing paradigm faces significant challenges in generating high-resolution visual content due to its quadratic time and memory complexity with respect to the number of spatial tokens. To address this limitation, we aim at a novel linear attention mechanism as an alternative in this paper. Specifically, we begin our exploration from recently introduced mod"},"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":"2409.02097","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-09-03T17:54:39Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"40486cb2dfce6801d981559a6a26d2689ec4feef68db5b7488401194051ba5a5","abstract_canon_sha256":"c6a7724027ea8ca17905540245039aaed7026ba805aba71f6cc7c196f4dcf7f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:21:48.501099Z","signature_b64":"KjNG0apezldW/NXed493KSm4LbaBgU9FaP1jRSSz7Ry5Udj+lzi+RgkVp9TAUiB6axe1tTEx8WKgaty7uQ/GCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f73a6a8e3de1bf83c13118b040cf982031b13d10d87e25e113973d2bc96e47c","last_reissued_at":"2026-07-05T09:21:48.500625Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:21:48.500625Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LinFusion: 1 GPU, 1 Minute, 16K Image","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Songhua Liu, Weihao Yu, Xinchao Wang, Zhenxiong Tan","submitted_at":"2024-09-03T17:54:39Z","abstract_excerpt":"Modern diffusion models, particularly those utilizing a Transformer-based UNet for denoising, rely heavily on self-attention operations to manage complex spatial relationships, thus achieving impressive generation performance. However, this existing paradigm faces significant challenges in generating high-resolution visual content due to its quadratic time and memory complexity with respect to the number of spatial tokens. To address this limitation, we aim at a novel linear attention mechanism as an alternative in this paper. Specifically, we begin our exploration from recently introduced mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.02097","kind":"arxiv","version":3},"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/2409.02097/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":"2409.02097","created_at":"2026-07-05T09:21:48.500682+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.02097v3","created_at":"2026-07-05T09:21:48.500682+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.02097","created_at":"2026-07-05T09:21:48.500682+00:00"},{"alias_kind":"pith_short_12","alias_value":"N5Z2NKHD3YN7","created_at":"2026-07-05T09:21:48.500682+00:00"},{"alias_kind":"pith_short_16","alias_value":"N5Z2NKHD3YN7QPAT","created_at":"2026-07-05T09:21:48.500682+00:00"},{"alias_kind":"pith_short_8","alias_value":"N5Z2NKHD","created_at":"2026-07-05T09:21:48.500682+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.26636","citing_title":"JetViT: Efficient High-Resolution Vision Transformer with Post-Training Attention Search","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00746","citing_title":"Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders","ref_index":123,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17294","citing_title":"HierEdit: Region-Aware Hierarchical Diffusion for Efficient High-Resolution Editing","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20147","citing_title":"PixVerve: Advancing Native UHR Image Generation to 100MP with a Large-Scale High-Quality Dataset","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI","json":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI.json","graph_json":"https://pith.science/api/pith-number/N5Z2NKHD3YN7QPATCGFQIDHZQI/graph.json","events_json":"https://pith.science/api/pith-number/N5Z2NKHD3YN7QPATCGFQIDHZQI/events.json","paper":"https://pith.science/paper/N5Z2NKHD"},"agent_actions":{"view_html":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI","download_json":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI.json","view_paper":"https://pith.science/paper/N5Z2NKHD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.02097&json=true","fetch_graph":"https://pith.science/api/pith-number/N5Z2NKHD3YN7QPATCGFQIDHZQI/graph.json","fetch_events":"https://pith.science/api/pith-number/N5Z2NKHD3YN7QPATCGFQIDHZQI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI/action/storage_attestation","attest_author":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI/action/author_attestation","sign_citation":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI/action/citation_signature","submit_replication":"https://pith.science/pith/N5Z2NKHD3YN7QPATCGFQIDHZQI/action/replication_record"}},"created_at":"2026-07-05T09:21:48.500682+00:00","updated_at":"2026-07-05T09:21:48.500682+00:00"}