{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GHWFWNUAIDXER3URQK3K5AM6KG","short_pith_number":"pith:GHWFWNUA","schema_version":"1.0","canonical_sha256":"31ec5b368040ee48ee9182b6ae819e51a9a05dda98822af1711c12a781ace057","source":{"kind":"arxiv","id":"2412.15032","version":2},"attestation_state":"computed","paper":{"title":"DCTdiff: Intriguing Properties of Image Generative Modeling in the DCT Space","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Albert Ali Salah, Haozhe Jia, Itir Onal Ertugrul, Jianlin Su, Lanmiao Liu, Mang Ning, Martin Bene\\v{s}, Mingxiao Li, Wenshuo Chen","submitted_at":"2024-12-19T16:44:01Z","abstract_excerpt":"This paper explores image modeling from the frequency space and introduces DCTdiff, an end-to-end diffusion generative paradigm that efficiently models images in the discrete cosine transform (DCT) space. We investigate the design space of DCTdiff and reveal the key design factors. Experiments on different frameworks (UViT, DiT), generation tasks, and various diffusion samplers demonstrate that DCTdiff outperforms pixel-based diffusion models regarding generative quality and training efficiency. Remarkably, DCTdiff can seamlessly scale up to 512$\\times$512 resolution without using the latent d"},"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":"2412.15032","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-19T16:44:01Z","cross_cats_sorted":["cs.LG","eess.IV"],"title_canon_sha256":"25148de2cbbf48d5afff708f096c037bd21db1cf59e84d4f7f954ffaf4bd348b","abstract_canon_sha256":"50aac7586ba4aeb86c5d2c7ad0d5b6295038e1c55ca2291f114d14bbe10aeb1d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:23.614788Z","signature_b64":"3ML2NziNA6x/qtqUQWV4DsjOIUlhcELHtX7pc0fnY142D40nnSKvQlYqV4RtBVRVfzTn1cC43iJWuWXrO70DDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31ec5b368040ee48ee9182b6ae819e51a9a05dda98822af1711c12a781ace057","last_reissued_at":"2026-07-05T11:12:23.614226Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:23.614226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DCTdiff: Intriguing Properties of Image Generative Modeling in the DCT Space","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Albert Ali Salah, Haozhe Jia, Itir Onal Ertugrul, Jianlin Su, Lanmiao Liu, Mang Ning, Martin Bene\\v{s}, Mingxiao Li, Wenshuo Chen","submitted_at":"2024-12-19T16:44:01Z","abstract_excerpt":"This paper explores image modeling from the frequency space and introduces DCTdiff, an end-to-end diffusion generative paradigm that efficiently models images in the discrete cosine transform (DCT) space. We investigate the design space of DCTdiff and reveal the key design factors. Experiments on different frameworks (UViT, DiT), generation tasks, and various diffusion samplers demonstrate that DCTdiff outperforms pixel-based diffusion models regarding generative quality and training efficiency. Remarkably, DCTdiff can seamlessly scale up to 512$\\times$512 resolution without using the latent d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15032","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/2412.15032/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":"2412.15032","created_at":"2026-07-05T11:12:23.614287+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.15032v2","created_at":"2026-07-05T11:12:23.614287+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15032","created_at":"2026-07-05T11:12:23.614287+00:00"},{"alias_kind":"pith_short_12","alias_value":"GHWFWNUAIDXE","created_at":"2026-07-05T11:12:23.614287+00:00"},{"alias_kind":"pith_short_16","alias_value":"GHWFWNUAIDXER3UR","created_at":"2026-07-05T11:12:23.614287+00:00"},{"alias_kind":"pith_short_8","alias_value":"GHWFWNUA","created_at":"2026-07-05T11:12:23.614287+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02661","citing_title":"Learning to Refine: Spectral-Decoupled Iterative Refinement Framework for Precipitation Nowcasting","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29570","citing_title":"Hierarchical Policy Learning via Spectral Decomposition","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26032","citing_title":"Everything at Every Scale: Scale-Invariant Diffusion with Continuous Super-Resolution","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14417","citing_title":"Before the Body Moves: Learning Anticipatory Joint Intent for Language-Conditioned Humanoid Control","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14417","citing_title":"Before the Body Moves: Learning Anticipatory Joint Intent for Language-Conditioned Humanoid Control","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09275","citing_title":"DiffATS: Diffusion in Aligned Tensor Space","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23536","citing_title":"$Z^2$-Sampling: Zero-Cost Zigzag Trajectories for Semantic Alignment in Diffusion Models","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23540","citing_title":"Oracle Noise: Faster Semantic Spherical Alignment for Interpretable Latent Optimization","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG","json":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG.json","graph_json":"https://pith.science/api/pith-number/GHWFWNUAIDXER3URQK3K5AM6KG/graph.json","events_json":"https://pith.science/api/pith-number/GHWFWNUAIDXER3URQK3K5AM6KG/events.json","paper":"https://pith.science/paper/GHWFWNUA"},"agent_actions":{"view_html":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG","download_json":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG.json","view_paper":"https://pith.science/paper/GHWFWNUA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.15032&json=true","fetch_graph":"https://pith.science/api/pith-number/GHWFWNUAIDXER3URQK3K5AM6KG/graph.json","fetch_events":"https://pith.science/api/pith-number/GHWFWNUAIDXER3URQK3K5AM6KG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG/action/storage_attestation","attest_author":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG/action/author_attestation","sign_citation":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG/action/citation_signature","submit_replication":"https://pith.science/pith/GHWFWNUAIDXER3URQK3K5AM6KG/action/replication_record"}},"created_at":"2026-07-05T11:12:23.614287+00:00","updated_at":"2026-07-05T11:12:23.614287+00:00"}