{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3TJDNKEN7QTMPJXVL4OPBF6VJH","short_pith_number":"pith:3TJDNKEN","schema_version":"1.0","canonical_sha256":"dcd236a88dfc26c7a6f55f1cf097d549de8b4438c9c95a7306a65d38dcc1f36b","source":{"kind":"arxiv","id":"2303.00257","version":1},"attestation_state":"computed","paper":{"title":"Hidden Markov Transformer for Simultaneous Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Shaolei Zhang, Yang Feng","submitted_at":"2023-03-01T06:28:33Z","abstract_excerpt":"Simultaneous machine translation (SiMT) outputs the target sequence while receiving the source sequence, and hence learning when to start translating each target token is the core challenge for SiMT task. However, it is non-trivial to learn the optimal moment among many possible moments of starting translating, as the moments of starting translating always hide inside the model and can only be supervised with the observed target sequence. In this paper, we propose a Hidden Markov Transformer (HMT), which treats the moments of starting translating as hidden events and the target sequence as the"},"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":"2303.00257","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-03-01T06:28:33Z","cross_cats_sorted":[],"title_canon_sha256":"45cbd533894e4aa3048736e0751527f061930db30c51f2492925357ce56b5740","abstract_canon_sha256":"e26ba040940ed0040347fc5ad6d9ee1cbcb2a23370ae51fcf3d2ab6b1a396ac9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:47:07.192666Z","signature_b64":"v1LXO0Bfshnk9hEIKuDzZfytPoaioHJdFAqTcobc+ESTse9jqMpvkG8UjxFjbd7HqW8pEadoaTwAgbX0RkLGDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dcd236a88dfc26c7a6f55f1cf097d549de8b4438c9c95a7306a65d38dcc1f36b","last_reissued_at":"2026-07-05T05:47:07.192291Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:47:07.192291Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hidden Markov Transformer for Simultaneous Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Shaolei Zhang, Yang Feng","submitted_at":"2023-03-01T06:28:33Z","abstract_excerpt":"Simultaneous machine translation (SiMT) outputs the target sequence while receiving the source sequence, and hence learning when to start translating each target token is the core challenge for SiMT task. However, it is non-trivial to learn the optimal moment among many possible moments of starting translating, as the moments of starting translating always hide inside the model and can only be supervised with the observed target sequence. In this paper, we propose a Hidden Markov Transformer (HMT), which treats the moments of starting translating as hidden events and the target sequence as the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.00257","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2303.00257/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":"2303.00257","created_at":"2026-07-05T05:47:07.192360+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.00257v1","created_at":"2026-07-05T05:47:07.192360+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.00257","created_at":"2026-07-05T05:47:07.192360+00:00"},{"alias_kind":"pith_short_12","alias_value":"3TJDNKEN7QTM","created_at":"2026-07-05T05:47:07.192360+00:00"},{"alias_kind":"pith_short_16","alias_value":"3TJDNKEN7QTMPJXV","created_at":"2026-07-05T05:47:07.192360+00:00"},{"alias_kind":"pith_short_8","alias_value":"3TJDNKEN","created_at":"2026-07-05T05:47:07.192360+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.17527","citing_title":"Seed LiveInterpret 2.0: End-to-end Simultaneous Speech-to-speech Translation with Your Voice","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH","json":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH.json","graph_json":"https://pith.science/api/pith-number/3TJDNKEN7QTMPJXVL4OPBF6VJH/graph.json","events_json":"https://pith.science/api/pith-number/3TJDNKEN7QTMPJXVL4OPBF6VJH/events.json","paper":"https://pith.science/paper/3TJDNKEN"},"agent_actions":{"view_html":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH","download_json":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH.json","view_paper":"https://pith.science/paper/3TJDNKEN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.00257&json=true","fetch_graph":"https://pith.science/api/pith-number/3TJDNKEN7QTMPJXVL4OPBF6VJH/graph.json","fetch_events":"https://pith.science/api/pith-number/3TJDNKEN7QTMPJXVL4OPBF6VJH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH/action/storage_attestation","attest_author":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH/action/author_attestation","sign_citation":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH/action/citation_signature","submit_replication":"https://pith.science/pith/3TJDNKEN7QTMPJXVL4OPBF6VJH/action/replication_record"}},"created_at":"2026-07-05T05:47:07.192360+00:00","updated_at":"2026-07-05T05:47:07.192360+00:00"}