{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:ZCI4GZKKS2TNZPZITOYDSY3OZ3","short_pith_number":"pith:ZCI4GZKK","schema_version":"1.0","canonical_sha256":"c891c3654a96a6dcbf289bb039636ecee8b7601cc6b9b2658c23e62feda0d7fd","source":{"kind":"arxiv","id":"2004.04092","version":4},"attestation_state":"computed","paper":{"title":"Optimus: Organizing Sentences via Pre-trained Modeling of a Latent Space","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Baolin Peng, Chunyuan Li, Jianfeng Gao, Xiang Gao, Xiujun Li, Yizhe Zhang, Yuan Li","submitted_at":"2020-04-05T06:20:18Z","abstract_excerpt":"When trained effectively, the Variational Autoencoder (VAE) can be both a powerful generative model and an effective representation learning framework for natural language. In this paper, we propose the first large-scale language VAE model, Optimus. A universal latent embedding space for sentences is first pre-trained on large text corpus, and then fine-tuned for various language generation and understanding tasks. Compared with GPT-2, Optimus enables guided language generation from an abstract level using the latent vectors. Compared with BERT, Optimus can generalize better on low-resource la"},"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":"2004.04092","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-05T06:20:18Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"bf6e317eef980ce9db4a4a8b756812b8e7ecf45513db73e633b4b238f0e1ab1d","abstract_canon_sha256":"9424c924ac1c2c311477044708f80fe9a17658e07fc91442d857177762bcc850"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:41:51.878073Z","signature_b64":"+UAKpn3NNmH9FRrUD3Gk7EM7Q9r/x24EFov2Pa2tqvlolYyGnzmkH6yPmPu6I7tC3kHOqyUEeX4zjqumVgTsCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c891c3654a96a6dcbf289bb039636ecee8b7601cc6b9b2658c23e62feda0d7fd","last_reissued_at":"2026-07-05T01:41:51.877620Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:41:51.877620Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimus: Organizing Sentences via Pre-trained Modeling of a Latent Space","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Baolin Peng, Chunyuan Li, Jianfeng Gao, Xiang Gao, Xiujun Li, Yizhe Zhang, Yuan Li","submitted_at":"2020-04-05T06:20:18Z","abstract_excerpt":"When trained effectively, the Variational Autoencoder (VAE) can be both a powerful generative model and an effective representation learning framework for natural language. In this paper, we propose the first large-scale language VAE model, Optimus. A universal latent embedding space for sentences is first pre-trained on large text corpus, and then fine-tuned for various language generation and understanding tasks. Compared with GPT-2, Optimus enables guided language generation from an abstract level using the latent vectors. Compared with BERT, Optimus can generalize better on low-resource la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.04092","kind":"arxiv","version":4},"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/2004.04092/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":"2004.04092","created_at":"2026-07-05T01:41:51.877682+00:00"},{"alias_kind":"arxiv_version","alias_value":"2004.04092v4","created_at":"2026-07-05T01:41:51.877682+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.04092","created_at":"2026-07-05T01:41:51.877682+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZCI4GZKKS2TN","created_at":"2026-07-05T01:41:51.877682+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZCI4GZKKS2TNZPZI","created_at":"2026-07-05T01:41:51.877682+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZCI4GZKK","created_at":"2026-07-05T01:41:51.877682+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.07748","citing_title":"TextLDM: Language Modeling with Continuous Latent Diffusion","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3","json":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3.json","graph_json":"https://pith.science/api/pith-number/ZCI4GZKKS2TNZPZITOYDSY3OZ3/graph.json","events_json":"https://pith.science/api/pith-number/ZCI4GZKKS2TNZPZITOYDSY3OZ3/events.json","paper":"https://pith.science/paper/ZCI4GZKK"},"agent_actions":{"view_html":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3","download_json":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3.json","view_paper":"https://pith.science/paper/ZCI4GZKK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2004.04092&json=true","fetch_graph":"https://pith.science/api/pith-number/ZCI4GZKKS2TNZPZITOYDSY3OZ3/graph.json","fetch_events":"https://pith.science/api/pith-number/ZCI4GZKKS2TNZPZITOYDSY3OZ3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3/action/storage_attestation","attest_author":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3/action/author_attestation","sign_citation":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3/action/citation_signature","submit_replication":"https://pith.science/pith/ZCI4GZKKS2TNZPZITOYDSY3OZ3/action/replication_record"}},"created_at":"2026-07-05T01:41:51.877682+00:00","updated_at":"2026-07-05T01:41:51.877682+00:00"}