{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JTOXZ7NVXZCY74VDKZLIZ3EBDQ","short_pith_number":"pith:JTOXZ7NV","schema_version":"1.0","canonical_sha256":"4cdd7cfdb5be458ff2a356568cec811c203413098d3548a7867c01d7a74f56a7","source":{"kind":"arxiv","id":"2403.05300","version":5},"attestation_state":"computed","paper":{"title":"Unity by Diversity: Improved Representation Learning in Multimodal VAEs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andrea Agostini, Babak Shahbaba, Daphn\\'e Chopard, Julia E. Vogt, Norbert Fortin, Stephan Mandt, Thomas M. Sutter, Yang Meng","submitted_at":"2024-03-08T13:29:46Z","abstract_excerpt":"Variational Autoencoders for multimodal data hold promise for many tasks in data analysis, such as representation learning, conditional generation, and imputation. Current architectures either share the encoder output, decoder input, or both across modalities to learn a shared representation. Such architectures impose hard constraints on the model. In this work, we show that a better latent representation can be obtained by replacing these hard constraints with a soft constraint. We propose a new mixture-of-experts prior, softly guiding each modality's latent representation towards a shared ag"},"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":"2403.05300","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-08T13:29:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"40ec5be00c8254734360213e768aa85caa69c9222cbcdc3047ddb68c8e037bb4","abstract_canon_sha256":"0bcdaa3c4b5ef19931eb61e6aad546ab93924820d23ecf15f119beabbb62e894"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:57:40.264955Z","signature_b64":"5qJ/M5rZBcfGzMI9BU7NDoAGBiMRwHsszohSYCVqXKfzCqqO2mwhvWB/uyidS/tSFVTwHbZdulEaHqjmTObsCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cdd7cfdb5be458ff2a356568cec811c203413098d3548a7867c01d7a74f56a7","last_reissued_at":"2026-07-05T09:57:40.264462Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:57:40.264462Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unity by Diversity: Improved Representation Learning in Multimodal VAEs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andrea Agostini, Babak Shahbaba, Daphn\\'e Chopard, Julia E. Vogt, Norbert Fortin, Stephan Mandt, Thomas M. Sutter, Yang Meng","submitted_at":"2024-03-08T13:29:46Z","abstract_excerpt":"Variational Autoencoders for multimodal data hold promise for many tasks in data analysis, such as representation learning, conditional generation, and imputation. Current architectures either share the encoder output, decoder input, or both across modalities to learn a shared representation. Such architectures impose hard constraints on the model. In this work, we show that a better latent representation can be obtained by replacing these hard constraints with a soft constraint. We propose a new mixture-of-experts prior, softly guiding each modality's latent representation towards a shared ag"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.05300","kind":"arxiv","version":5},"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/2403.05300/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":"2403.05300","created_at":"2026-07-05T09:57:40.264521+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.05300v5","created_at":"2026-07-05T09:57:40.264521+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.05300","created_at":"2026-07-05T09:57:40.264521+00:00"},{"alias_kind":"pith_short_12","alias_value":"JTOXZ7NVXZCY","created_at":"2026-07-05T09:57:40.264521+00:00"},{"alias_kind":"pith_short_16","alias_value":"JTOXZ7NVXZCY74VD","created_at":"2026-07-05T09:57:40.264521+00:00"},{"alias_kind":"pith_short_8","alias_value":"JTOXZ7NV","created_at":"2026-07-05T09:57:40.264521+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.11561","citing_title":"Predicting Pulmonary Hypertension in Newborns: A Multi-view VAE Approach","ref_index":27,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ","json":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ.json","graph_json":"https://pith.science/api/pith-number/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/graph.json","events_json":"https://pith.science/api/pith-number/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/events.json","paper":"https://pith.science/paper/JTOXZ7NV"},"agent_actions":{"view_html":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ","download_json":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ.json","view_paper":"https://pith.science/paper/JTOXZ7NV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.05300&json=true","fetch_graph":"https://pith.science/api/pith-number/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/graph.json","fetch_events":"https://pith.science/api/pith-number/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/action/storage_attestation","attest_author":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/action/author_attestation","sign_citation":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/action/citation_signature","submit_replication":"https://pith.science/pith/JTOXZ7NVXZCY74VDKZLIZ3EBDQ/action/replication_record"}},"created_at":"2026-07-05T09:57:40.264521+00:00","updated_at":"2026-07-05T09:57:40.264521+00:00"}