{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GM5FCRJFWCHMRCVNRIBTSWJIYD","short_pith_number":"pith:GM5FCRJF","schema_version":"1.0","canonical_sha256":"333a514525b08ec88aad8a03395928c0f632dd19c745734c9470b982e25a615a","source":{"kind":"arxiv","id":"2104.08692","version":2},"attestation_state":"computed","paper":{"title":"MT6: Multilingual Pretrained Text-to-Text Transformer with Translation Pairs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Furu Wei, Heyan Huang, Li Dong, Shaohan Huang Xian-Ling Mao, Shuming Ma, Zewen Chi","submitted_at":"2021-04-18T03:24:07Z","abstract_excerpt":"Multilingual T5 (mT5) pretrains a sequence-to-sequence model on massive monolingual texts, which has shown promising results on many cross-lingual tasks. In this paper, we improve multilingual text-to-text transfer Transformer with translation pairs (mT6). Specifically, we explore three cross-lingual text-to-text pre-training tasks, namely, machine translation, translation pair span corruption, and translation span corruption. In addition, we propose a partially non-autoregressive objective for text-to-text pre-training. We evaluate the methods on eight multilingual benchmark datasets, includi"},"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":"2104.08692","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-18T03:24:07Z","cross_cats_sorted":[],"title_canon_sha256":"e81ee1fb7b3ff04dea682bf4c42cb3ed6b9ee314e1f9369140a1ffe31b037f1e","abstract_canon_sha256":"cc7f6f126ed17237d5f075297aa1314d7671367e6637ec4d191aca30d53c19eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:27.431885Z","signature_b64":"PseORp8tdDpghY+bR6oNYj68A6eWOcQSZcoM4rUApBqnT3wdxxaLlp4HmYiip5hRF0dOHtYxUCdIhtAWk/8/BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"333a514525b08ec88aad8a03395928c0f632dd19c745734c9470b982e25a615a","last_reissued_at":"2026-07-05T03:13:27.431399Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:27.431399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MT6: Multilingual Pretrained Text-to-Text Transformer with Translation Pairs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Furu Wei, Heyan Huang, Li Dong, Shaohan Huang Xian-Ling Mao, Shuming Ma, Zewen Chi","submitted_at":"2021-04-18T03:24:07Z","abstract_excerpt":"Multilingual T5 (mT5) pretrains a sequence-to-sequence model on massive monolingual texts, which has shown promising results on many cross-lingual tasks. In this paper, we improve multilingual text-to-text transfer Transformer with translation pairs (mT6). Specifically, we explore three cross-lingual text-to-text pre-training tasks, namely, machine translation, translation pair span corruption, and translation span corruption. In addition, we propose a partially non-autoregressive objective for text-to-text pre-training. We evaluate the methods on eight multilingual benchmark datasets, includi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.08692","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/2104.08692/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":"2104.08692","created_at":"2026-07-05T03:13:27.431463+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.08692v2","created_at":"2026-07-05T03:13:27.431463+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.08692","created_at":"2026-07-05T03:13:27.431463+00:00"},{"alias_kind":"pith_short_12","alias_value":"GM5FCRJFWCHM","created_at":"2026-07-05T03:13:27.431463+00:00"},{"alias_kind":"pith_short_16","alias_value":"GM5FCRJFWCHMRCVN","created_at":"2026-07-05T03:13:27.431463+00:00"},{"alias_kind":"pith_short_8","alias_value":"GM5FCRJF","created_at":"2026-07-05T03:13:27.431463+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.09714","citing_title":"How Small Transformation Expose the Weakness of Semantic Similarity Measures","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD","json":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD.json","graph_json":"https://pith.science/api/pith-number/GM5FCRJFWCHMRCVNRIBTSWJIYD/graph.json","events_json":"https://pith.science/api/pith-number/GM5FCRJFWCHMRCVNRIBTSWJIYD/events.json","paper":"https://pith.science/paper/GM5FCRJF"},"agent_actions":{"view_html":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD","download_json":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD.json","view_paper":"https://pith.science/paper/GM5FCRJF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.08692&json=true","fetch_graph":"https://pith.science/api/pith-number/GM5FCRJFWCHMRCVNRIBTSWJIYD/graph.json","fetch_events":"https://pith.science/api/pith-number/GM5FCRJFWCHMRCVNRIBTSWJIYD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD/action/storage_attestation","attest_author":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD/action/author_attestation","sign_citation":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD/action/citation_signature","submit_replication":"https://pith.science/pith/GM5FCRJFWCHMRCVNRIBTSWJIYD/action/replication_record"}},"created_at":"2026-07-05T03:13:27.431463+00:00","updated_at":"2026-07-05T03:13:27.431463+00:00"}