{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ISB4YSOK3A3DXP3LBUZ52IGSUV","short_pith_number":"pith:ISB4YSOK","schema_version":"1.0","canonical_sha256":"4483cc49cad8363bbf6b0d33dd20d2a5582149ad4766540ad6e3a523f8bdba1b","source":{"kind":"arxiv","id":"2303.09769","version":2},"attestation_state":"computed","paper":{"title":"Denoising Diffusion Autoencoders are Unified Self-supervised Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Di Huang, Hongyu Yang, Weilai Xiang, Yunhong Wang","submitted_at":"2023-03-17T04:20:47Z","abstract_excerpt":"Inspired by recent advances in diffusion models, which are reminiscent of denoising autoencoders, we investigate whether they can acquire discriminative representations for classification via generative pre-training. This paper shows that the networks in diffusion models, namely denoising diffusion autoencoders (DDAE), are unified self-supervised learners: by pre-training on unconditional image generation, DDAE has already learned strongly linear-separable representations within its intermediate layers without auxiliary encoders, thus making diffusion pre-training emerge as a general approach "},"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":"2303.09769","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-03-17T04:20:47Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d451ba639ab82e7025f3f80968c021f6f6a1fd167e60347e1bd6d517ce221fc2","abstract_canon_sha256":"aea3b6c39317aaa27dac6a723c664fbac15070a11a25c9818259ad549b0d3cc4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:42:42.757992Z","signature_b64":"HS2izjvNOnnRHZ6HwYqqpzcbiVCil4Pw3muNBkI9UplzwNC+KLX1Un2QjMOqTxRIocVeKUBdo5aclQ4+8fojAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4483cc49cad8363bbf6b0d33dd20d2a5582149ad4766540ad6e3a523f8bdba1b","last_reissued_at":"2026-07-05T06:42:42.757496Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:42:42.757496Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Denoising Diffusion Autoencoders are Unified Self-supervised Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Di Huang, Hongyu Yang, Weilai Xiang, Yunhong Wang","submitted_at":"2023-03-17T04:20:47Z","abstract_excerpt":"Inspired by recent advances in diffusion models, which are reminiscent of denoising autoencoders, we investigate whether they can acquire discriminative representations for classification via generative pre-training. This paper shows that the networks in diffusion models, namely denoising diffusion autoencoders (DDAE), are unified self-supervised learners: by pre-training on unconditional image generation, DDAE has already learned strongly linear-separable representations within its intermediate layers without auxiliary encoders, thus making diffusion pre-training emerge as a general approach "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.09769","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/2303.09769/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":"2303.09769","created_at":"2026-07-05T06:42:42.757564+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.09769v2","created_at":"2026-07-05T06:42:42.757564+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.09769","created_at":"2026-07-05T06:42:42.757564+00:00"},{"alias_kind":"pith_short_12","alias_value":"ISB4YSOK3A3D","created_at":"2026-07-05T06:42:42.757564+00:00"},{"alias_kind":"pith_short_16","alias_value":"ISB4YSOK3A3DXP3L","created_at":"2026-07-05T06:42:42.757564+00:00"},{"alias_kind":"pith_short_8","alias_value":"ISB4YSOK","created_at":"2026-07-05T06:42:42.757564+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25012","citing_title":"Learning from Semantic Dictionaries: Discriminative Codebook Contrastive Learning for Unified Visual Representation and Generation","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV","json":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV.json","graph_json":"https://pith.science/api/pith-number/ISB4YSOK3A3DXP3LBUZ52IGSUV/graph.json","events_json":"https://pith.science/api/pith-number/ISB4YSOK3A3DXP3LBUZ52IGSUV/events.json","paper":"https://pith.science/paper/ISB4YSOK"},"agent_actions":{"view_html":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV","download_json":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV.json","view_paper":"https://pith.science/paper/ISB4YSOK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.09769&json=true","fetch_graph":"https://pith.science/api/pith-number/ISB4YSOK3A3DXP3LBUZ52IGSUV/graph.json","fetch_events":"https://pith.science/api/pith-number/ISB4YSOK3A3DXP3LBUZ52IGSUV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV/action/storage_attestation","attest_author":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV/action/author_attestation","sign_citation":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV/action/citation_signature","submit_replication":"https://pith.science/pith/ISB4YSOK3A3DXP3LBUZ52IGSUV/action/replication_record"}},"created_at":"2026-07-05T06:42:42.757564+00:00","updated_at":"2026-07-05T06:42:42.757564+00:00"}