{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:JGI72F4U4VCW4HHLRHQWNLDDQ2","short_pith_number":"pith:JGI72F4U","schema_version":"1.0","canonical_sha256":"4991fd1794e5456e1ceb89e166ac6386a1914c21575cfb899515a40acdaca4c5","source":{"kind":"arxiv","id":"1906.07651","version":1},"attestation_state":"computed","paper":{"title":"Scheduled Sampling for Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andr\\'e F. T. Martins, Tsvetomila Mihaylova","submitted_at":"2019-06-18T15:46:08Z","abstract_excerpt":"Scheduled sampling is a technique for avoiding one of the known problems in sequence-to-sequence generation: exposure bias. It consists of feeding the model a mix of the teacher forced embeddings and the model predictions from the previous step in training time. The technique has been used for improving the model performance with recurrent neural networks (RNN). In the Transformer model, unlike the RNN, the generation of a new word attends to the full sentence generated so far, not only to the last word, and it is not straightforward to apply the scheduled sampling technique. We propose some s"},"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":"1906.07651","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-18T15:46:08Z","cross_cats_sorted":[],"title_canon_sha256":"faf735181501a78076122e84ebf01833be525552c4274a4d2456aa896efd30b4","abstract_canon_sha256":"b49552560274b595fea09132ea6d9773f06cd4d433eb36f8b9fe8089973f7b0b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:07.778450Z","signature_b64":"kRBksM49qapughm5G140cIF8T+SyfWkg88VCTXvTKE5uGLGtqq1FBtIlhIqXAeDnf+tioSysQ3PG0mqEl0ylDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4991fd1794e5456e1ceb89e166ac6386a1914c21575cfb899515a40acdaca4c5","last_reissued_at":"2026-05-17T23:42:07.777641Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:07.777641Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scheduled Sampling for Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andr\\'e F. T. Martins, Tsvetomila Mihaylova","submitted_at":"2019-06-18T15:46:08Z","abstract_excerpt":"Scheduled sampling is a technique for avoiding one of the known problems in sequence-to-sequence generation: exposure bias. It consists of feeding the model a mix of the teacher forced embeddings and the model predictions from the previous step in training time. The technique has been used for improving the model performance with recurrent neural networks (RNN). In the Transformer model, unlike the RNN, the generation of a new word attends to the full sentence generated so far, not only to the last word, and it is not straightforward to apply the scheduled sampling technique. We propose some s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.07651","kind":"arxiv","version":1},"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":"1906.07651","created_at":"2026-05-17T23:42:07.777805+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.07651v1","created_at":"2026-05-17T23:42:07.777805+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.07651","created_at":"2026-05-17T23:42:07.777805+00:00"},{"alias_kind":"pith_short_12","alias_value":"JGI72F4U4VCW","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_16","alias_value":"JGI72F4U4VCW4HHL","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_8","alias_value":"JGI72F4U","created_at":"2026-05-18T12:33:21.387695+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/JGI72F4U4VCW4HHLRHQWNLDDQ2","json":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2.json","graph_json":"https://pith.science/api/pith-number/JGI72F4U4VCW4HHLRHQWNLDDQ2/graph.json","events_json":"https://pith.science/api/pith-number/JGI72F4U4VCW4HHLRHQWNLDDQ2/events.json","paper":"https://pith.science/paper/JGI72F4U"},"agent_actions":{"view_html":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2","download_json":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2.json","view_paper":"https://pith.science/paper/JGI72F4U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.07651&json=true","fetch_graph":"https://pith.science/api/pith-number/JGI72F4U4VCW4HHLRHQWNLDDQ2/graph.json","fetch_events":"https://pith.science/api/pith-number/JGI72F4U4VCW4HHLRHQWNLDDQ2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2/action/storage_attestation","attest_author":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2/action/author_attestation","sign_citation":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2/action/citation_signature","submit_replication":"https://pith.science/pith/JGI72F4U4VCW4HHLRHQWNLDDQ2/action/replication_record"}},"created_at":"2026-05-17T23:42:07.777805+00:00","updated_at":"2026-05-17T23:42:07.777805+00:00"}