{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:RSGS7RD25ZYA5DB3KSOFL32NJF","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c1b0a142a428a7b4e77a3a176b4ea46d84e3f4608107e4da50175d3b1928b42c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2021-10-08T13:28:28Z","title_canon_sha256":"6074066a81d90840abebf3ec5060fc2a3ef8680a23558138e96d222186ba4ba5"},"schema_version":"1.0","source":{"id":"2110.04121","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.04121","created_at":"2026-07-05T04:12:28Z"},{"alias_kind":"arxiv_version","alias_value":"2110.04121v2","created_at":"2026-07-05T04:12:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.04121","created_at":"2026-07-05T04:12:28Z"},{"alias_kind":"pith_short_12","alias_value":"RSGS7RD25ZYA","created_at":"2026-07-05T04:12:28Z"},{"alias_kind":"pith_short_16","alias_value":"RSGS7RD25ZYA5DB3","created_at":"2026-07-05T04:12:28Z"},{"alias_kind":"pith_short_8","alias_value":"RSGS7RD2","created_at":"2026-07-05T04:12:28Z"}],"graph_snapshots":[{"event_id":"sha256:91b4dd8031277b30ed85aef26881ec348febedce00781f19a00488ce07936760","target":"graph","created_at":"2026-07-05T04:12:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2110.04121/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal variational autoencoders (VAEs) have shown promise as efficient generative models for weakly-supervised data. Yet, despite their advantage of weak supervision, they exhibit a gap in generative quality compared to unimodal VAEs, which are completely unsupervised. In an attempt to explain this gap, we uncover a fundamental limitation that applies to a large family of mixture-based multimodal VAEs. We prove that the sub-sampling of modalities enforces an undesirable upper bound on the multimodal ELBO and thereby limits the generative quality of the respective models. Empirically, we sh","authors_text":"Emanuele Palumbo, Imant Daunhawer, Julia E. Vogt, Kieran Chin-Cheong, Thomas M. Sutter","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2021-10-08T13:28:28Z","title":"On the Limitations of Multimodal VAEs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.04121","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8de314107a50f62b847ee60460812bbd8b239d4b539c34e8b01220012014c8e3","target":"record","created_at":"2026-07-05T04:12:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c1b0a142a428a7b4e77a3a176b4ea46d84e3f4608107e4da50175d3b1928b42c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2021-10-08T13:28:28Z","title_canon_sha256":"6074066a81d90840abebf3ec5060fc2a3ef8680a23558138e96d222186ba4ba5"},"schema_version":"1.0","source":{"id":"2110.04121","kind":"arxiv","version":2}},"canonical_sha256":"8c8d2fc47aee700e8c3b549c55ef4d4960e5a97a7ea10cc3703758261d2383fe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8c8d2fc47aee700e8c3b549c55ef4d4960e5a97a7ea10cc3703758261d2383fe","first_computed_at":"2026-07-05T04:12:28.443015Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:12:28.443015Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GQeNzZqpwSMVySNdTB4n1QiRYzX9aMZ2355PORAJWyCLd8+dLaqaLapzq2WYUjLtqfFzS7KzSw9ruUTaZCQNCg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:12:28.443469Z","signed_message":"canonical_sha256_bytes"},"source_id":"2110.04121","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8de314107a50f62b847ee60460812bbd8b239d4b539c34e8b01220012014c8e3","sha256:91b4dd8031277b30ed85aef26881ec348febedce00781f19a00488ce07936760"],"state_sha256":"2760c0ff76c71f8570eed5d43614808928495fd6d2392f30272081cfa54e3b9b"}