{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GOE5EMIULAFXBRZWMN6N2ORZCY","short_pith_number":"pith:GOE5EMIU","schema_version":"1.0","canonical_sha256":"3389d23114580b70c736637cdd3a39162281ecf31aa2275097bdd91f74087e61","source":{"kind":"arxiv","id":"2101.10044","version":2},"attestation_state":"computed","paper":{"title":"Cross-lingual Visual Pre-training for Multimodal Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Aykut Erdem, Erkut Erdem, Lucia Specia, Menekse Kuyu, Mustafa Sercan Amac, Ozan Caglayan, Pranava Madhyastha","submitted_at":"2021-01-25T12:46:41Z","abstract_excerpt":"Pre-trained language models have been shown to improve performance in many natural language tasks substantially. Although the early focus of such models was single language pre-training, recent advances have resulted in cross-lingual and visual pre-training methods. In this paper, we combine these two approaches to learn visually-grounded cross-lingual representations. Specifically, we extend the translation language modelling (Lample and Conneau, 2019) with masked region classification and perform pre-training with three-way parallel vision & language corpora. We show that when fine-tuned for"},"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":"2101.10044","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-01-25T12:46:41Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"a54679f89913da2299fafd39bd1583dd1c6a51141dc76213587cd5c87e915b02","abstract_canon_sha256":"aa94d7f4bc1b17ed94108385cbeda756fbbfd13fb040f63854900eeff04324a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:33:57.190474Z","signature_b64":"CWlVKBVVVY2wTNnZ9TRci1PFyU6BV+CkP1hznijxDNIW82v1WIKvq7YtOtUIx23Fc5rP3icKxFTh8vy7lMCoDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3389d23114580b70c736637cdd3a39162281ecf31aa2275097bdd91f74087e61","last_reissued_at":"2026-07-05T02:33:57.190022Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:33:57.190022Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cross-lingual Visual Pre-training for Multimodal Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Aykut Erdem, Erkut Erdem, Lucia Specia, Menekse Kuyu, Mustafa Sercan Amac, Ozan Caglayan, Pranava Madhyastha","submitted_at":"2021-01-25T12:46:41Z","abstract_excerpt":"Pre-trained language models have been shown to improve performance in many natural language tasks substantially. Although the early focus of such models was single language pre-training, recent advances have resulted in cross-lingual and visual pre-training methods. In this paper, we combine these two approaches to learn visually-grounded cross-lingual representations. Specifically, we extend the translation language modelling (Lample and Conneau, 2019) with masked region classification and perform pre-training with three-way parallel vision & language corpora. We show that when fine-tuned for"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.10044","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/2101.10044/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":"2101.10044","created_at":"2026-07-05T02:33:57.190080+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.10044v2","created_at":"2026-07-05T02:33:57.190080+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.10044","created_at":"2026-07-05T02:33:57.190080+00:00"},{"alias_kind":"pith_short_12","alias_value":"GOE5EMIULAFX","created_at":"2026-07-05T02:33:57.190080+00:00"},{"alias_kind":"pith_short_16","alias_value":"GOE5EMIULAFXBRZW","created_at":"2026-07-05T02:33:57.190080+00:00"},{"alias_kind":"pith_short_8","alias_value":"GOE5EMIU","created_at":"2026-07-05T02:33:57.190080+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/GOE5EMIULAFXBRZWMN6N2ORZCY","json":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY.json","graph_json":"https://pith.science/api/pith-number/GOE5EMIULAFXBRZWMN6N2ORZCY/graph.json","events_json":"https://pith.science/api/pith-number/GOE5EMIULAFXBRZWMN6N2ORZCY/events.json","paper":"https://pith.science/paper/GOE5EMIU"},"agent_actions":{"view_html":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY","download_json":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY.json","view_paper":"https://pith.science/paper/GOE5EMIU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.10044&json=true","fetch_graph":"https://pith.science/api/pith-number/GOE5EMIULAFXBRZWMN6N2ORZCY/graph.json","fetch_events":"https://pith.science/api/pith-number/GOE5EMIULAFXBRZWMN6N2ORZCY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY/action/storage_attestation","attest_author":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY/action/author_attestation","sign_citation":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY/action/citation_signature","submit_replication":"https://pith.science/pith/GOE5EMIULAFXBRZWMN6N2ORZCY/action/replication_record"}},"created_at":"2026-07-05T02:33:57.190080+00:00","updated_at":"2026-07-05T02:33:57.190080+00:00"}