{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:57RDJMXZ5UKC6FG4QQPLI66HD4","short_pith_number":"pith:57RDJMXZ","schema_version":"1.0","canonical_sha256":"efe234b2f9ed142f14dc841eb47bc71f1cffc858ab05b1ab8a8e90e8892f85fc","source":{"kind":"arxiv","id":"2405.18428","version":2},"attestation_state":"computed","paper":{"title":"DiG: Scalable and Efficient Diffusion Models with Gated Linear Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bencheng Liao, Hanshu Yan, Jiashi Feng, Jun Hao Liew, Lianghui Zhu, Xinggang Wang, Zilong Huang","submitted_at":"2024-05-28T17:59:33Z","abstract_excerpt":"Diffusion models with large-scale pre-training have achieved significant success in the field of visual content generation, particularly exemplified by Diffusion Transformers (DiT). However, DiT models have faced challenges with quadratic complexity efficiency, especially when handling long sequences. In this paper, we aim to incorporate the sub-quadratic modeling capability of Gated Linear Attention (GLA) into the 2D diffusion backbone. Specifically, we introduce Diffusion Gated Linear Attention Transformers (DiG), a simple, adoptable solution with minimal parameter overhead. We offer two var"},"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":"2405.18428","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-05-28T17:59:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"25b03eaf37be2b599d170a3d135558ba8a70ae97329e331f599b285a42a2a39f","abstract_canon_sha256":"d24e8f387b819c8589c37cb3b68b3db14eefcd5000bddb52b58c4f4447f48025"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:40:57.912452Z","signature_b64":"boncwSNvEoBiWwne5xcAXrqA/5YLl1x15TN6f+fE5cZO+HgDLe3kIRV2DFejYeH9+eXQJXEJynkin+cBFbhLAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"efe234b2f9ed142f14dc841eb47bc71f1cffc858ab05b1ab8a8e90e8892f85fc","last_reissued_at":"2026-07-05T09:40:57.911944Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:40:57.911944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DiG: Scalable and Efficient Diffusion Models with Gated Linear Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bencheng Liao, Hanshu Yan, Jiashi Feng, Jun Hao Liew, Lianghui Zhu, Xinggang Wang, Zilong Huang","submitted_at":"2024-05-28T17:59:33Z","abstract_excerpt":"Diffusion models with large-scale pre-training have achieved significant success in the field of visual content generation, particularly exemplified by Diffusion Transformers (DiT). However, DiT models have faced challenges with quadratic complexity efficiency, especially when handling long sequences. In this paper, we aim to incorporate the sub-quadratic modeling capability of Gated Linear Attention (GLA) into the 2D diffusion backbone. Specifically, we introduce Diffusion Gated Linear Attention Transformers (DiG), a simple, adoptable solution with minimal parameter overhead. We offer two var"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.18428","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/2405.18428/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":"2405.18428","created_at":"2026-07-05T09:40:57.912008+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.18428v2","created_at":"2026-07-05T09:40:57.912008+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.18428","created_at":"2026-07-05T09:40:57.912008+00:00"},{"alias_kind":"pith_short_12","alias_value":"57RDJMXZ5UKC","created_at":"2026-07-05T09:40:57.912008+00:00"},{"alias_kind":"pith_short_16","alias_value":"57RDJMXZ5UKC6FG4","created_at":"2026-07-05T09:40:57.912008+00:00"},{"alias_kind":"pith_short_8","alias_value":"57RDJMXZ","created_at":"2026-07-05T09:40:57.912008+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02090","citing_title":"FocusDiT: Masking Queries in Diffusion Transformers for Fine-grained Image Generation","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15684","citing_title":"ElasticDiT: Efficient Diffusion Transformers via Elastic Architecture and Sparse Attention for High-Resolution Image Generation on Mobile Devices","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2410.10629","citing_title":"SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4","json":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4.json","graph_json":"https://pith.science/api/pith-number/57RDJMXZ5UKC6FG4QQPLI66HD4/graph.json","events_json":"https://pith.science/api/pith-number/57RDJMXZ5UKC6FG4QQPLI66HD4/events.json","paper":"https://pith.science/paper/57RDJMXZ"},"agent_actions":{"view_html":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4","download_json":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4.json","view_paper":"https://pith.science/paper/57RDJMXZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.18428&json=true","fetch_graph":"https://pith.science/api/pith-number/57RDJMXZ5UKC6FG4QQPLI66HD4/graph.json","fetch_events":"https://pith.science/api/pith-number/57RDJMXZ5UKC6FG4QQPLI66HD4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4/action/storage_attestation","attest_author":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4/action/author_attestation","sign_citation":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4/action/citation_signature","submit_replication":"https://pith.science/pith/57RDJMXZ5UKC6FG4QQPLI66HD4/action/replication_record"}},"created_at":"2026-07-05T09:40:57.912008+00:00","updated_at":"2026-07-05T09:40:57.912008+00:00"}