{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:7RMNL76RC6PK6D36E2HQIWIV2K","short_pith_number":"pith:7RMNL76R","schema_version":"1.0","canonical_sha256":"fc58d5ffd1179eaf0f7e268f045915d2bee0391add1e8872e0ba9a069956c4ff","source":{"kind":"arxiv","id":"1901.05534","version":2},"attestation_state":"computed","paper":{"title":"Lagging Inference Networks and Posterior Collapse in Variational Autoencoders","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Spokoyny, Graham Neubig, Junxian He, Taylor Berg-Kirkpatrick","submitted_at":"2019-01-16T21:32:33Z","abstract_excerpt":"The variational autoencoder (VAE) is a popular combination of deep latent variable model and accompanying variational learning technique. By using a neural inference network to approximate the model's posterior on latent variables, VAEs efficiently parameterize a lower bound on marginal data likelihood that can be optimized directly via gradient methods. In practice, however, VAE training often results in a degenerate local optimum known as \"posterior collapse\" where the model learns to ignore the latent variable and the approximate posterior mimics the prior. In this paper, we investigate pos"},"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":"1901.05534","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-16T21:32:33Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a3f546ea3477b65a02a295329df94838cf8401dbf6911cf265b8e0239fd245db","abstract_canon_sha256":"dd16ab481005f0cc7bc352a835c1f8c9e8f448f2788b8deadd156562f071e2df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:16.818841Z","signature_b64":"x+NZZMe78FMplKrRRiHN8MmoPJPWi74rqYneF+KemLYwV2TmLhHT6ARvbWjabWLD35qzB2F6Ee8meIXBLMSIBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fc58d5ffd1179eaf0f7e268f045915d2bee0391add1e8872e0ba9a069956c4ff","last_reissued_at":"2026-05-17T23:55:16.818231Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:16.818231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lagging Inference Networks and Posterior Collapse in Variational Autoencoders","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Spokoyny, Graham Neubig, Junxian He, Taylor Berg-Kirkpatrick","submitted_at":"2019-01-16T21:32:33Z","abstract_excerpt":"The variational autoencoder (VAE) is a popular combination of deep latent variable model and accompanying variational learning technique. By using a neural inference network to approximate the model's posterior on latent variables, VAEs efficiently parameterize a lower bound on marginal data likelihood that can be optimized directly via gradient methods. In practice, however, VAE training often results in a degenerate local optimum known as \"posterior collapse\" where the model learns to ignore the latent variable and the approximate posterior mimics the prior. In this paper, we investigate pos"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.05534","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":""},"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":"1901.05534","created_at":"2026-05-17T23:55:16.818331+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.05534v2","created_at":"2026-05-17T23:55:16.818331+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.05534","created_at":"2026-05-17T23:55:16.818331+00:00"},{"alias_kind":"pith_short_12","alias_value":"7RMNL76RC6PK","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_16","alias_value":"7RMNL76RC6PK6D36","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_8","alias_value":"7RMNL76R","created_at":"2026-05-18T12:33:12.712433+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.06032","citing_title":"Ensemble-Based Dirichlet Modeling for Predictive Uncertainty and Selective Classification","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20293","citing_title":"Synthetic Flight Data Generation Using Generative Models","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04366","citing_title":"Conditional Flow-VAE for Safety-Critical Traffic Scenario Generation","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02589","citing_title":"Representation learning from OCT images","ref_index":103,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K","json":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K.json","graph_json":"https://pith.science/api/pith-number/7RMNL76RC6PK6D36E2HQIWIV2K/graph.json","events_json":"https://pith.science/api/pith-number/7RMNL76RC6PK6D36E2HQIWIV2K/events.json","paper":"https://pith.science/paper/7RMNL76R"},"agent_actions":{"view_html":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K","download_json":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K.json","view_paper":"https://pith.science/paper/7RMNL76R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.05534&json=true","fetch_graph":"https://pith.science/api/pith-number/7RMNL76RC6PK6D36E2HQIWIV2K/graph.json","fetch_events":"https://pith.science/api/pith-number/7RMNL76RC6PK6D36E2HQIWIV2K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K/action/storage_attestation","attest_author":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K/action/author_attestation","sign_citation":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K/action/citation_signature","submit_replication":"https://pith.science/pith/7RMNL76RC6PK6D36E2HQIWIV2K/action/replication_record"}},"created_at":"2026-05-17T23:55:16.818331+00:00","updated_at":"2026-05-17T23:55:16.818331+00:00"}