{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:CNJZBQZRGAUK7KMQLZ4IX6VPN7","short_pith_number":"pith:CNJZBQZR","schema_version":"1.0","canonical_sha256":"135390c3313028afa9905e788bfaaf6fda43db5428c4eb303f953e0aef4cee30","source":{"kind":"arxiv","id":"2209.12590","version":2},"attestation_state":"computed","paper":{"title":"Learning to Drop Out: An Adversarial Approach to Training Sequence VAEs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Carl Allen, {\\DJ}or{\\dj}e Miladinovi\\'c, Joachim M. Buhmann, Kumar Shridhar, Kushal Jain, Max B. Paulus, Mrinmaya Sachan","submitted_at":"2022-09-26T11:21:19Z","abstract_excerpt":"In principle, applying variational autoencoders (VAEs) to sequential data offers a method for controlled sequence generation, manipulation, and structured representation learning. However, training sequence VAEs is challenging: autoregressive decoders can often explain the data without utilizing the latent space, known as posterior collapse. To mitigate this, state-of-the-art models weaken the powerful decoder by applying uniformly random dropout to the decoder input. We show theoretically that this removes pointwise mutual information provided by the decoder input, which is compensated for by"},"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":"2209.12590","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-09-26T11:21:19Z","cross_cats_sorted":[],"title_canon_sha256":"41f5de885104d070124523557b23d332b77a156b5faff6ec5130283653c3c6a9","abstract_canon_sha256":"938f3d21a89a5fcd16cb5cdf346a0b260a3adcc356cb8ad314ac64b215a0b3e5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:25:50.860919Z","signature_b64":"jrbxR6duV6oIaLktvyauLp+47q0S3nIJRng2dQ3CO9KkMXTuvHvBtK/6iPXKnneTaVie5dKAZHRpavO/47zIDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"135390c3313028afa9905e788bfaaf6fda43db5428c4eb303f953e0aef4cee30","last_reissued_at":"2026-07-05T05:25:50.860483Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:25:50.860483Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Drop Out: An Adversarial Approach to Training Sequence VAEs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Carl Allen, {\\DJ}or{\\dj}e Miladinovi\\'c, Joachim M. Buhmann, Kumar Shridhar, Kushal Jain, Max B. Paulus, Mrinmaya Sachan","submitted_at":"2022-09-26T11:21:19Z","abstract_excerpt":"In principle, applying variational autoencoders (VAEs) to sequential data offers a method for controlled sequence generation, manipulation, and structured representation learning. However, training sequence VAEs is challenging: autoregressive decoders can often explain the data without utilizing the latent space, known as posterior collapse. To mitigate this, state-of-the-art models weaken the powerful decoder by applying uniformly random dropout to the decoder input. We show theoretically that this removes pointwise mutual information provided by the decoder input, which is compensated for by"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.12590","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/2209.12590/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":"2209.12590","created_at":"2026-07-05T05:25:50.860541+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.12590v2","created_at":"2026-07-05T05:25:50.860541+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.12590","created_at":"2026-07-05T05:25:50.860541+00:00"},{"alias_kind":"pith_short_12","alias_value":"CNJZBQZRGAUK","created_at":"2026-07-05T05:25:50.860541+00:00"},{"alias_kind":"pith_short_16","alias_value":"CNJZBQZRGAUK7KMQ","created_at":"2026-07-05T05:25:50.860541+00:00"},{"alias_kind":"pith_short_8","alias_value":"CNJZBQZR","created_at":"2026-07-05T05:25:50.860541+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/CNJZBQZRGAUK7KMQLZ4IX6VPN7","json":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7.json","graph_json":"https://pith.science/api/pith-number/CNJZBQZRGAUK7KMQLZ4IX6VPN7/graph.json","events_json":"https://pith.science/api/pith-number/CNJZBQZRGAUK7KMQLZ4IX6VPN7/events.json","paper":"https://pith.science/paper/CNJZBQZR"},"agent_actions":{"view_html":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7","download_json":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7.json","view_paper":"https://pith.science/paper/CNJZBQZR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.12590&json=true","fetch_graph":"https://pith.science/api/pith-number/CNJZBQZRGAUK7KMQLZ4IX6VPN7/graph.json","fetch_events":"https://pith.science/api/pith-number/CNJZBQZRGAUK7KMQLZ4IX6VPN7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7/action/storage_attestation","attest_author":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7/action/author_attestation","sign_citation":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7/action/citation_signature","submit_replication":"https://pith.science/pith/CNJZBQZRGAUK7KMQLZ4IX6VPN7/action/replication_record"}},"created_at":"2026-07-05T05:25:50.860541+00:00","updated_at":"2026-07-05T05:25:50.860541+00:00"}