{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:FFJOYFPDDDOVJVQ2HR2WBMROBO","short_pith_number":"pith:FFJOYFPD","schema_version":"1.0","canonical_sha256":"2952ec15e318dd54d61a3c7560b22e0bb4f8d6fda03ccebacfe67cda8671e4dc","source":{"kind":"arxiv","id":"1905.02450","version":5},"attestation_state":"computed","paper":{"title":"MASS: Masked Sequence to Sequence Pre-training for Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Jianfeng Lu, Kaitao Song, Tao Qin, Tie-Yan Liu, Xu Tan","submitted_at":"2019-05-07T10:13:04Z","abstract_excerpt":"Pre-training and fine-tuning, e.g., BERT, have achieved great success in language understanding by transferring knowledge from rich-resource pre-training task to the low/zero-resource downstream tasks. Inspired by the success of BERT, we propose MAsked Sequence to Sequence pre-training (MASS) for the encoder-decoder based language generation tasks. MASS adopts the encoder-decoder framework to reconstruct a sentence fragment given the remaining part of the sentence: its encoder takes a sentence with randomly masked fragment (several consecutive tokens) as input, and its decoder tries to predict"},"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":"1905.02450","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-05-07T10:13:04Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"75ec261d5914331f3984e22442eba417a7b9880970c76ce5ab477f42af074b51","abstract_canon_sha256":"74a040f23546279a800cb63ba357d8388932893e65de6c534a2b47820a795871"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:48.119691Z","signature_b64":"OwpPL+A13lRNFlZATmzO0SLHHX8RhqvheWowc6HkxO4uReo3DslRAM8rIkgGTYoahLw0WVuA0PFuI4y69QLyCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2952ec15e318dd54d61a3c7560b22e0bb4f8d6fda03ccebacfe67cda8671e4dc","last_reissued_at":"2026-05-17T23:42:48.118959Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:48.118959Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MASS: Masked Sequence to Sequence Pre-training for Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Jianfeng Lu, Kaitao Song, Tao Qin, Tie-Yan Liu, Xu Tan","submitted_at":"2019-05-07T10:13:04Z","abstract_excerpt":"Pre-training and fine-tuning, e.g., BERT, have achieved great success in language understanding by transferring knowledge from rich-resource pre-training task to the low/zero-resource downstream tasks. Inspired by the success of BERT, we propose MAsked Sequence to Sequence pre-training (MASS) for the encoder-decoder based language generation tasks. MASS adopts the encoder-decoder framework to reconstruct a sentence fragment given the remaining part of the sentence: its encoder takes a sentence with randomly masked fragment (several consecutive tokens) as input, and its decoder tries to predict"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.02450","kind":"arxiv","version":5},"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":"1905.02450","created_at":"2026-05-17T23:42:48.119096+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.02450v5","created_at":"2026-05-17T23:42:48.119096+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.02450","created_at":"2026-05-17T23:42:48.119096+00:00"},{"alias_kind":"pith_short_12","alias_value":"FFJOYFPDDDOV","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_16","alias_value":"FFJOYFPDDDOVJVQ2","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_8","alias_value":"FFJOYFPD","created_at":"2026-05-18T12:33:15.570797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"2109.10616","citing_title":"Enriching and Controlling Global Semantics for Text Summarization","ref_index":32,"is_internal_anchor":true},{"citing_arxiv_id":"2412.18735","citing_title":"Automatic Self-supervised Learning for Social Recommendations","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2009.01325","citing_title":"Learning to summarize from human feedback","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2207.14255","citing_title":"Efficient Training of Language Models to Fill in the Middle","ref_index":137,"is_internal_anchor":true},{"citing_arxiv_id":"2309.16671","citing_title":"Demystifying CLIP Data","ref_index":170,"is_internal_anchor":true},{"citing_arxiv_id":"1910.10683","citing_title":"Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06582","citing_title":"PairAlign: A Framework for Sequence Tokenization via Self-Alignment with Applications to Audio Tokenization","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2402.06196","citing_title":"Large Language Models: A Survey","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2204.02311","citing_title":"PaLM: Scaling Language Modeling with Pathways","ref_index":147,"is_internal_anchor":false},{"citing_arxiv_id":"2005.14165","citing_title":"Language Models are Few-Shot Learners","ref_index":76,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO","json":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO.json","graph_json":"https://pith.science/api/pith-number/FFJOYFPDDDOVJVQ2HR2WBMROBO/graph.json","events_json":"https://pith.science/api/pith-number/FFJOYFPDDDOVJVQ2HR2WBMROBO/events.json","paper":"https://pith.science/paper/FFJOYFPD"},"agent_actions":{"view_html":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO","download_json":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO.json","view_paper":"https://pith.science/paper/FFJOYFPD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.02450&json=true","fetch_graph":"https://pith.science/api/pith-number/FFJOYFPDDDOVJVQ2HR2WBMROBO/graph.json","fetch_events":"https://pith.science/api/pith-number/FFJOYFPDDDOVJVQ2HR2WBMROBO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO/action/storage_attestation","attest_author":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO/action/author_attestation","sign_citation":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO/action/citation_signature","submit_replication":"https://pith.science/pith/FFJOYFPDDDOVJVQ2HR2WBMROBO/action/replication_record"}},"created_at":"2026-05-17T23:42:48.119096+00:00","updated_at":"2026-05-17T23:42:48.119096+00:00"}