{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:Q7HSEGISGPDLXAR45ZAID54SJB","short_pith_number":"pith:Q7HSEGIS","schema_version":"1.0","canonical_sha256":"87cf22191233c6bb823cee4081f7924852da77ac2c89a5e53b9d97219ef6ed53","source":{"kind":"arxiv","id":"2105.02470","version":2},"attestation_state":"computed","paper":{"title":"Generalized Multimodal ELBO","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Imant Daunhawer, Julia E. Vogt, Thomas M. Sutter","submitted_at":"2021-05-06T07:05:00Z","abstract_excerpt":"Multiple data types naturally co-occur when describing real-world phenomena and learning from them is a long-standing goal in machine learning research. However, existing self-supervised generative models approximating an ELBO are not able to fulfill all desired requirements of multimodal models: their posterior approximation functions lead to a trade-off between the semantic coherence and the ability to learn the joint data distribution. We propose a new, generalized ELBO formulation for multimodal data that overcomes these limitations. The new objective encompasses two previous methods as sp"},"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":"2105.02470","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-06T07:05:00Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"d6d4c0311e68c426250af64250ff127e7678719a0fc37c56704cdb44f9afe74b","abstract_canon_sha256":"3db485c6e1fedea10f5924079b282f6ab42638009520c000933aa7b6b8ed13e7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:52:16.284628Z","signature_b64":"sUJlDaNjLzAXQbxBo3fE+VVqta4aae/QK0NWkkUCMHXswxjWK4E0DkVwOz71u+ZnjEEvAGLKl2AuokZBWyLgAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87cf22191233c6bb823cee4081f7924852da77ac2c89a5e53b9d97219ef6ed53","last_reissued_at":"2026-07-05T02:52:16.284279Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:52:16.284279Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generalized Multimodal ELBO","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Imant Daunhawer, Julia E. Vogt, Thomas M. Sutter","submitted_at":"2021-05-06T07:05:00Z","abstract_excerpt":"Multiple data types naturally co-occur when describing real-world phenomena and learning from them is a long-standing goal in machine learning research. However, existing self-supervised generative models approximating an ELBO are not able to fulfill all desired requirements of multimodal models: their posterior approximation functions lead to a trade-off between the semantic coherence and the ability to learn the joint data distribution. We propose a new, generalized ELBO formulation for multimodal data that overcomes these limitations. The new objective encompasses two previous methods as sp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.02470","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/2105.02470/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":"2105.02470","created_at":"2026-07-05T02:52:16.284341+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.02470v2","created_at":"2026-07-05T02:52:16.284341+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.02470","created_at":"2026-07-05T02:52:16.284341+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q7HSEGISGPDL","created_at":"2026-07-05T02:52:16.284341+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q7HSEGISGPDLXAR4","created_at":"2026-07-05T02:52:16.284341+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q7HSEGIS","created_at":"2026-07-05T02:52:16.284341+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.22644","citing_title":"Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2507.12165","citing_title":"Multi-Component VAE with Gaussian Markov Random Field","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21670","citing_title":"Diverse via bounded Agreement: Geometric Regularization for Multimodal Fusion","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09522","citing_title":"Emergent Communication for Co-constructed Emotion Between Embodied Agents via Collective Predictive Coding","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25352","citing_title":"GraphPL: Leveraging GNN for Efficient and Robust Modalities Imputation in Patchwork Learning","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB","json":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB.json","graph_json":"https://pith.science/api/pith-number/Q7HSEGISGPDLXAR45ZAID54SJB/graph.json","events_json":"https://pith.science/api/pith-number/Q7HSEGISGPDLXAR45ZAID54SJB/events.json","paper":"https://pith.science/paper/Q7HSEGIS"},"agent_actions":{"view_html":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB","download_json":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB.json","view_paper":"https://pith.science/paper/Q7HSEGIS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.02470&json=true","fetch_graph":"https://pith.science/api/pith-number/Q7HSEGISGPDLXAR45ZAID54SJB/graph.json","fetch_events":"https://pith.science/api/pith-number/Q7HSEGISGPDLXAR45ZAID54SJB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB/action/storage_attestation","attest_author":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB/action/author_attestation","sign_citation":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB/action/citation_signature","submit_replication":"https://pith.science/pith/Q7HSEGISGPDLXAR45ZAID54SJB/action/replication_record"}},"created_at":"2026-07-05T02:52:16.284341+00:00","updated_at":"2026-07-05T02:52:16.284341+00:00"}