{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:W4OM4RCEXI7F2KS3LXLJP7GKNJ","short_pith_number":"pith:W4OM4RCE","schema_version":"1.0","canonical_sha256":"b71cce4444ba3e5d2a5b5dd697fcca6a739a3b47b5e1dd1e14f9b3f406239396","source":{"kind":"arxiv","id":"1703.01898","version":2},"attestation_state":"computed","paper":{"title":"Generative and Discriminative Text Classification with Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"stat.ML","authors_text":"Chris Dyer, Dani Yogatama, Phil Blunsom, Wang Ling","submitted_at":"2017-03-06T14:40:09Z","abstract_excerpt":"We empirically characterize the performance of discriminative and generative LSTM models for text classification. We find that although RNN-based generative models are more powerful than their bag-of-words ancestors (e.g., they account for conditional dependencies across words in a document), they have higher asymptotic error rates than discriminatively trained RNN models. However we also find that generative models approach their asymptotic error rate more rapidly than their discriminative counterparts---the same pattern that Ng & Jordan (2001) proved holds for linear classification models th"},"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":"1703.01898","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-03-06T14:40:09Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"100b2992543bfacf935f0773f1d2cd2ac1f6cd84cf0ed2aad19328478b083db7","abstract_canon_sha256":"3df4a537168e86a893015e44c73f8f1b60d0324e15f5a2f82f5d2b9c341410f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:43:39.298252Z","signature_b64":"AtPeRgIugelrGv689RnW3DV9dhcKGi2X+EcohqTTP3/85tqBol5o6Y2IFB+afu+ppKBSbJPDHPFIym4jEMo0BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b71cce4444ba3e5d2a5b5dd697fcca6a739a3b47b5e1dd1e14f9b3f406239396","last_reissued_at":"2026-05-18T00:43:39.297762Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:43:39.297762Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generative and Discriminative Text Classification with Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"stat.ML","authors_text":"Chris Dyer, Dani Yogatama, Phil Blunsom, Wang Ling","submitted_at":"2017-03-06T14:40:09Z","abstract_excerpt":"We empirically characterize the performance of discriminative and generative LSTM models for text classification. We find that although RNN-based generative models are more powerful than their bag-of-words ancestors (e.g., they account for conditional dependencies across words in a document), they have higher asymptotic error rates than discriminatively trained RNN models. However we also find that generative models approach their asymptotic error rate more rapidly than their discriminative counterparts---the same pattern that Ng & Jordan (2001) proved holds for linear classification models th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1703.01898","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":"1703.01898","created_at":"2026-05-18T00:43:39.297839+00:00"},{"alias_kind":"arxiv_version","alias_value":"1703.01898v2","created_at":"2026-05-18T00:43:39.297839+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1703.01898","created_at":"2026-05-18T00:43:39.297839+00:00"},{"alias_kind":"pith_short_12","alias_value":"W4OM4RCEXI7F","created_at":"2026-05-18T12:31:49.984773+00:00"},{"alias_kind":"pith_short_16","alias_value":"W4OM4RCEXI7F2KS3","created_at":"2026-05-18T12:31:49.984773+00:00"},{"alias_kind":"pith_short_8","alias_value":"W4OM4RCE","created_at":"2026-05-18T12:31:49.984773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.15905","citing_title":"Generative Long-term User Interest Modeling for Click-Through Rate Prediction","ref_index":36,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ","json":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ.json","graph_json":"https://pith.science/api/pith-number/W4OM4RCEXI7F2KS3LXLJP7GKNJ/graph.json","events_json":"https://pith.science/api/pith-number/W4OM4RCEXI7F2KS3LXLJP7GKNJ/events.json","paper":"https://pith.science/paper/W4OM4RCE"},"agent_actions":{"view_html":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ","download_json":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ.json","view_paper":"https://pith.science/paper/W4OM4RCE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1703.01898&json=true","fetch_graph":"https://pith.science/api/pith-number/W4OM4RCEXI7F2KS3LXLJP7GKNJ/graph.json","fetch_events":"https://pith.science/api/pith-number/W4OM4RCEXI7F2KS3LXLJP7GKNJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ/action/storage_attestation","attest_author":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ/action/author_attestation","sign_citation":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ/action/citation_signature","submit_replication":"https://pith.science/pith/W4OM4RCEXI7F2KS3LXLJP7GKNJ/action/replication_record"}},"created_at":"2026-05-18T00:43:39.297839+00:00","updated_at":"2026-05-18T00:43:39.297839+00:00"}