{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4ESOKHJXUZD3FBC2VFZOSQEDNT","short_pith_number":"pith:4ESOKHJX","schema_version":"1.0","canonical_sha256":"e124e51d37a647b2845aa972e940836ccd9adfe4796a7d666b707c186e5603af","source":{"kind":"arxiv","id":"2307.12868","version":2},"attestation_state":"computed","paper":{"title":"Understanding the Latent Space of Diffusion Models through the Lens of Riemannian Geometry","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jaewoong Choi, Junghyo Jo, Mingi Kwon, Yong-Hyun Park, Youngjung Uh","submitted_at":"2023-07-24T15:06:42Z","abstract_excerpt":"Despite the success of diffusion models (DMs), we still lack a thorough understanding of their latent space. To understand the latent space $\\mathbf{x}_t \\in \\mathcal{X}$, we analyze them from a geometrical perspective. Our approach involves deriving the local latent basis within $\\mathcal{X}$ by leveraging the pullback metric associated with their encoding feature maps. Remarkably, our discovered local latent basis enables image editing capabilities by moving $\\mathbf{x}_t$, the latent space of DMs, along the basis vector at specific timesteps. We further analyze how the geometric structure o"},"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":"2307.12868","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-07-24T15:06:42Z","cross_cats_sorted":[],"title_canon_sha256":"8a51068ef523b9f78399b246589b2d99c070276098848f64d8b8a032761f44e0","abstract_canon_sha256":"baa1d767da55fb21ccd68582c170c3c13fb79d677a78aada479511493f78621a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:05:46.338617Z","signature_b64":"y9108fSCTnYznDhrsXmkZpIjusLYxH1MV329YkU0DBrTgYuHRCqpmahDyH0Go4af73YVJoOJWnlPtDi8y379Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e124e51d37a647b2845aa972e940836ccd9adfe4796a7d666b707c186e5603af","last_reissued_at":"2026-07-05T07:05:46.338175Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:05:46.338175Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding the Latent Space of Diffusion Models through the Lens of Riemannian Geometry","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jaewoong Choi, Junghyo Jo, Mingi Kwon, Yong-Hyun Park, Youngjung Uh","submitted_at":"2023-07-24T15:06:42Z","abstract_excerpt":"Despite the success of diffusion models (DMs), we still lack a thorough understanding of their latent space. To understand the latent space $\\mathbf{x}_t \\in \\mathcal{X}$, we analyze them from a geometrical perspective. Our approach involves deriving the local latent basis within $\\mathcal{X}$ by leveraging the pullback metric associated with their encoding feature maps. Remarkably, our discovered local latent basis enables image editing capabilities by moving $\\mathbf{x}_t$, the latent space of DMs, along the basis vector at specific timesteps. We further analyze how the geometric structure o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.12868","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/2307.12868/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":"2307.12868","created_at":"2026-07-05T07:05:46.338231+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.12868v2","created_at":"2026-07-05T07:05:46.338231+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.12868","created_at":"2026-07-05T07:05:46.338231+00:00"},{"alias_kind":"pith_short_12","alias_value":"4ESOKHJXUZD3","created_at":"2026-07-05T07:05:46.338231+00:00"},{"alias_kind":"pith_short_16","alias_value":"4ESOKHJXUZD3FBC2","created_at":"2026-07-05T07:05:46.338231+00:00"},{"alias_kind":"pith_short_8","alias_value":"4ESOKHJX","created_at":"2026-07-05T07:05:46.338231+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17673","citing_title":"Grokking of Diffusion Models: Case Study on Modular Addition","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT","json":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT.json","graph_json":"https://pith.science/api/pith-number/4ESOKHJXUZD3FBC2VFZOSQEDNT/graph.json","events_json":"https://pith.science/api/pith-number/4ESOKHJXUZD3FBC2VFZOSQEDNT/events.json","paper":"https://pith.science/paper/4ESOKHJX"},"agent_actions":{"view_html":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT","download_json":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT.json","view_paper":"https://pith.science/paper/4ESOKHJX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.12868&json=true","fetch_graph":"https://pith.science/api/pith-number/4ESOKHJXUZD3FBC2VFZOSQEDNT/graph.json","fetch_events":"https://pith.science/api/pith-number/4ESOKHJXUZD3FBC2VFZOSQEDNT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT/action/storage_attestation","attest_author":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT/action/author_attestation","sign_citation":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT/action/citation_signature","submit_replication":"https://pith.science/pith/4ESOKHJXUZD3FBC2VFZOSQEDNT/action/replication_record"}},"created_at":"2026-07-05T07:05:46.338231+00:00","updated_at":"2026-07-05T07:05:46.338231+00:00"}