{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AULNLWRPX7H4URFCEKY3UNQOKH","short_pith_number":"pith:AULNLWRP","schema_version":"1.0","canonical_sha256":"0516d5da2fbfcfca44a222b1ba360e51d37ad88ffab4de8930402e95bc4b5cfa","source":{"kind":"arxiv","id":"2504.21681","version":2},"attestation_state":"computed","paper":{"title":"Investigating the Effect of Parallel Data in the Cross-Lingual Transfer for Vision-Language Encoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrei-Alexandru Manea, Jind\\v{r}ich Libovick\\'y","submitted_at":"2025-04-30T14:19:15Z","abstract_excerpt":"Most pre-trained Vision-Language (VL) models and training data for the downstream tasks are only available in English. Therefore, multilingual VL tasks are solved using cross-lingual transfer: fine-tune a multilingual pre-trained model or transfer the text encoder using parallel data. We study the alternative approach: transferring an already trained encoder using parallel data. We investigate the effect of parallel data: domain and the number of languages, which were out of focus in previous work. Our results show that even machine-translated task data are the best on average, caption-like au"},"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":"2504.21681","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-30T14:19:15Z","cross_cats_sorted":[],"title_canon_sha256":"8e9fce7a043462b71fafe9b8800a807d3ee407fddce9cd134dc7f14a159fac15","abstract_canon_sha256":"eef4f19a10153e10c1984e0cb444b7a6a8d915e7a8ff2ef56bbb6e7110d4a3bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:11.833364Z","signature_b64":"hEYvlUMbw5urZ732pvwrjV7ZnoFT3vl8toXaDkcygPRYXNdSHumaHw+FuF4mu0pY38OOtvVyzMN0H9tbYezEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0516d5da2fbfcfca44a222b1ba360e51d37ad88ffab4de8930402e95bc4b5cfa","last_reissued_at":"2026-07-05T11:54:11.832762Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:11.832762Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Investigating the Effect of Parallel Data in the Cross-Lingual Transfer for Vision-Language Encoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrei-Alexandru Manea, Jind\\v{r}ich Libovick\\'y","submitted_at":"2025-04-30T14:19:15Z","abstract_excerpt":"Most pre-trained Vision-Language (VL) models and training data for the downstream tasks are only available in English. Therefore, multilingual VL tasks are solved using cross-lingual transfer: fine-tune a multilingual pre-trained model or transfer the text encoder using parallel data. We study the alternative approach: transferring an already trained encoder using parallel data. We investigate the effect of parallel data: domain and the number of languages, which were out of focus in previous work. Our results show that even machine-translated task data are the best on average, caption-like au"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.21681","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/2504.21681/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":"2504.21681","created_at":"2026-07-05T11:54:11.832839+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.21681v2","created_at":"2026-07-05T11:54:11.832839+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.21681","created_at":"2026-07-05T11:54:11.832839+00:00"},{"alias_kind":"pith_short_12","alias_value":"AULNLWRPX7H4","created_at":"2026-07-05T11:54:11.832839+00:00"},{"alias_kind":"pith_short_16","alias_value":"AULNLWRPX7H4URFC","created_at":"2026-07-05T11:54:11.832839+00:00"},{"alias_kind":"pith_short_8","alias_value":"AULNLWRP","created_at":"2026-07-05T11:54:11.832839+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.22123","citing_title":"Multilingual Vision-Language Models, A Survey","ref_index":94,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH","json":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH.json","graph_json":"https://pith.science/api/pith-number/AULNLWRPX7H4URFCEKY3UNQOKH/graph.json","events_json":"https://pith.science/api/pith-number/AULNLWRPX7H4URFCEKY3UNQOKH/events.json","paper":"https://pith.science/paper/AULNLWRP"},"agent_actions":{"view_html":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH","download_json":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH.json","view_paper":"https://pith.science/paper/AULNLWRP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.21681&json=true","fetch_graph":"https://pith.science/api/pith-number/AULNLWRPX7H4URFCEKY3UNQOKH/graph.json","fetch_events":"https://pith.science/api/pith-number/AULNLWRPX7H4URFCEKY3UNQOKH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH/action/storage_attestation","attest_author":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH/action/author_attestation","sign_citation":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH/action/citation_signature","submit_replication":"https://pith.science/pith/AULNLWRPX7H4URFCEKY3UNQOKH/action/replication_record"}},"created_at":"2026-07-05T11:54:11.832839+00:00","updated_at":"2026-07-05T11:54:11.832839+00:00"}