{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:V67DGM6FIGJIOTC54E3SDL2TVQ","short_pith_number":"pith:V67DGM6F","schema_version":"1.0","canonical_sha256":"afbe3333c54192874c5de13721af53ac06ca112d48855790550809065f9ead2e","source":{"kind":"arxiv","id":"2507.15900","version":1},"attestation_state":"computed","paper":{"title":"Improving the Generation of VAEs with High Dimensional Latent Spaces by the use of Hyperspherical Coordinates","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alejandro Ascarate, Clinton Fookes, Leo Lebrat, Olivier Salvado, Rodrigo Santa Cruz","submitted_at":"2025-07-21T05:10:43Z","abstract_excerpt":"Variational autoencoders (VAE) encode data into lower-dimensional latent vectors before decoding those vectors back to data. Once trained, decoding a random latent vector from the prior usually does not produce meaningful data, at least when the latent space has more than a dozen dimensions. In this paper, we investigate this issue by drawing insight from high dimensional statistics: in these regimes, the latent vectors of a standard VAE are by construction distributed uniformly on a hypersphere. We propose to formulate the latent variables of a VAE using hyperspherical coordinates, which allo"},"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":"2507.15900","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-21T05:10:43Z","cross_cats_sorted":[],"title_canon_sha256":"2c93b6907c9f0ad5756460c2750b6a8b942e0b22f6776ae5f76e5bddcc006449","abstract_canon_sha256":"89127528fd544083d01fe5f37c5bc0d5b39ed3ba39df3a5ec89900ff1ecf773c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:40:58.876552Z","signature_b64":"nOYSlrpKENQtw9q0qxf/p2WllJFvkhOMKC1oXgUMYSb9fhpx9d5qdY5PZ49+sDaeNvyr9rcsCY9TIWROX2DqAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"afbe3333c54192874c5de13721af53ac06ca112d48855790550809065f9ead2e","last_reissued_at":"2026-07-05T11:40:58.876076Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:40:58.876076Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving the Generation of VAEs with High Dimensional Latent Spaces by the use of Hyperspherical Coordinates","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alejandro Ascarate, Clinton Fookes, Leo Lebrat, Olivier Salvado, Rodrigo Santa Cruz","submitted_at":"2025-07-21T05:10:43Z","abstract_excerpt":"Variational autoencoders (VAE) encode data into lower-dimensional latent vectors before decoding those vectors back to data. Once trained, decoding a random latent vector from the prior usually does not produce meaningful data, at least when the latent space has more than a dozen dimensions. In this paper, we investigate this issue by drawing insight from high dimensional statistics: in these regimes, the latent vectors of a standard VAE are by construction distributed uniformly on a hypersphere. We propose to formulate the latent variables of a VAE using hyperspherical coordinates, which allo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15900","kind":"arxiv","version":1},"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/2507.15900/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":"2507.15900","created_at":"2026-07-05T11:40:58.876129+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.15900v1","created_at":"2026-07-05T11:40:58.876129+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15900","created_at":"2026-07-05T11:40:58.876129+00:00"},{"alias_kind":"pith_short_12","alias_value":"V67DGM6FIGJI","created_at":"2026-07-05T11:40:58.876129+00:00"},{"alias_kind":"pith_short_16","alias_value":"V67DGM6FIGJIOTC5","created_at":"2026-07-05T11:40:58.876129+00:00"},{"alias_kind":"pith_short_8","alias_value":"V67DGM6F","created_at":"2026-07-05T11:40:58.876129+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ","json":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ.json","graph_json":"https://pith.science/api/pith-number/V67DGM6FIGJIOTC54E3SDL2TVQ/graph.json","events_json":"https://pith.science/api/pith-number/V67DGM6FIGJIOTC54E3SDL2TVQ/events.json","paper":"https://pith.science/paper/V67DGM6F"},"agent_actions":{"view_html":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ","download_json":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ.json","view_paper":"https://pith.science/paper/V67DGM6F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.15900&json=true","fetch_graph":"https://pith.science/api/pith-number/V67DGM6FIGJIOTC54E3SDL2TVQ/graph.json","fetch_events":"https://pith.science/api/pith-number/V67DGM6FIGJIOTC54E3SDL2TVQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ/action/storage_attestation","attest_author":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ/action/author_attestation","sign_citation":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ/action/citation_signature","submit_replication":"https://pith.science/pith/V67DGM6FIGJIOTC54E3SDL2TVQ/action/replication_record"}},"created_at":"2026-07-05T11:40:58.876129+00:00","updated_at":"2026-07-05T11:40:58.876129+00:00"}