{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:EOS7PJAKNMQ2UIWMVGTUPI7CHF","short_pith_number":"pith:EOS7PJAK","schema_version":"1.0","canonical_sha256":"23a5f7a40a6b21aa22cca9a747a3e23951764c0c2e45be8bb47611d57b91ece2","source":{"kind":"arxiv","id":"1907.12461","version":2},"attestation_state":"computed","paper":{"title":"Leveraging Pre-trained Checkpoints for Sequence Generation Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aliaksei Severyn, Sascha Rothe, Shashi Narayan","submitted_at":"2019-07-29T14:42:30Z","abstract_excerpt":"Unsupervised pre-training of large neural models has recently revolutionized Natural Language Processing. By warm-starting from the publicly released checkpoints, NLP practitioners have pushed the state-of-the-art on multiple benchmarks while saving significant amounts of compute time. So far the focus has been mainly on the Natural Language Understanding tasks. In this paper, we demonstrate the efficacy of pre-trained checkpoints for Sequence Generation. We developed a Transformer-based sequence-to-sequence model that is compatible with publicly available pre-trained BERT, GPT-2 and RoBERTa c"},"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":"1907.12461","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-07-29T14:42:30Z","cross_cats_sorted":[],"title_canon_sha256":"ff6f5dbc34a42a7836f34b04d8ac93cad5feb15925266bbc547c6582100b5393","abstract_canon_sha256":"d059e8f12eddeac035d514ba980ae6e51fbc1c00da190de8a676f1177edc5134"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:46:50.115912Z","signature_b64":"suyYYXpvMrbCax9mOVPA174PYh4+GF4DYMCfzBDZm3sBDPwiwr9L8sM9imPASgHwmOjhPV3uvITdHoL+j0YJBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23a5f7a40a6b21aa22cca9a747a3e23951764c0c2e45be8bb47611d57b91ece2","last_reissued_at":"2026-07-05T04:46:50.115551Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:46:50.115551Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leveraging Pre-trained Checkpoints for Sequence Generation Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aliaksei Severyn, Sascha Rothe, Shashi Narayan","submitted_at":"2019-07-29T14:42:30Z","abstract_excerpt":"Unsupervised pre-training of large neural models has recently revolutionized Natural Language Processing. By warm-starting from the publicly released checkpoints, NLP practitioners have pushed the state-of-the-art on multiple benchmarks while saving significant amounts of compute time. So far the focus has been mainly on the Natural Language Understanding tasks. In this paper, we demonstrate the efficacy of pre-trained checkpoints for Sequence Generation. We developed a Transformer-based sequence-to-sequence model that is compatible with publicly available pre-trained BERT, GPT-2 and RoBERTa c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.12461","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/1907.12461/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":"1907.12461","created_at":"2026-07-05T04:46:50.115615+00:00"},{"alias_kind":"arxiv_version","alias_value":"1907.12461v2","created_at":"2026-07-05T04:46:50.115615+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.12461","created_at":"2026-07-05T04:46:50.115615+00:00"},{"alias_kind":"pith_short_12","alias_value":"EOS7PJAKNMQ2","created_at":"2026-07-05T04:46:50.115615+00:00"},{"alias_kind":"pith_short_16","alias_value":"EOS7PJAKNMQ2UIWM","created_at":"2026-07-05T04:46:50.115615+00:00"},{"alias_kind":"pith_short_8","alias_value":"EOS7PJAK","created_at":"2026-07-05T04:46:50.115615+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.06818","citing_title":"Ad Headline Generation using Self-Critical Masked Language Model","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2607.08071","citing_title":"COBART: Controlled, Optimized, Bidirectional and Auto-Regressive Transformer for Ad Headline Generation","ref_index":23,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF","json":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF.json","graph_json":"https://pith.science/api/pith-number/EOS7PJAKNMQ2UIWMVGTUPI7CHF/graph.json","events_json":"https://pith.science/api/pith-number/EOS7PJAKNMQ2UIWMVGTUPI7CHF/events.json","paper":"https://pith.science/paper/EOS7PJAK"},"agent_actions":{"view_html":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF","download_json":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF.json","view_paper":"https://pith.science/paper/EOS7PJAK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1907.12461&json=true","fetch_graph":"https://pith.science/api/pith-number/EOS7PJAKNMQ2UIWMVGTUPI7CHF/graph.json","fetch_events":"https://pith.science/api/pith-number/EOS7PJAKNMQ2UIWMVGTUPI7CHF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF/action/storage_attestation","attest_author":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF/action/author_attestation","sign_citation":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF/action/citation_signature","submit_replication":"https://pith.science/pith/EOS7PJAKNMQ2UIWMVGTUPI7CHF/action/replication_record"}},"created_at":"2026-07-05T04:46:50.115615+00:00","updated_at":"2026-07-05T04:46:50.115615+00:00"}