{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:A3LGKWJZVRPV7NLAK6TYRUM3CR","short_pith_number":"pith:A3LGKWJZ","schema_version":"1.0","canonical_sha256":"06d6655939ac5f5fb56057a788d19b14743207846363e880ac576fae1d336198","source":{"kind":"arxiv","id":"2202.13083","version":1},"attestation_state":"computed","paper":{"title":"Multi-Level Contrastive Learning for Cross-Lingual Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Beiduo Chen, Bin Gu, Quan Liu, Wu Guo, Yongchao Wang","submitted_at":"2022-02-26T07:14:20Z","abstract_excerpt":"Cross-language pre-trained models such as multilingual BERT (mBERT) have achieved significant performance in various cross-lingual downstream NLP tasks. This paper proposes a multi-level contrastive learning (ML-CTL) framework to further improve the cross-lingual ability of pre-trained models. The proposed method uses translated parallel data to encourage the model to generate similar semantic embeddings for different languages. However, unlike the sentence-level alignment used in most previous studies, in this paper, we explicitly integrate the word-level information of each pair of parallel "},"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":"2202.13083","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-02-26T07:14:20Z","cross_cats_sorted":[],"title_canon_sha256":"77ed8da4b91a4e6a547df968349c2be107b3de6fc7ba35df0311f955ef802de0","abstract_canon_sha256":"ac793cef675f0f1912c5f4632d3bbd640ce41fb80e09266783c80b9c9b678a66"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:00:25.953383Z","signature_b64":"IsSty3GXsb3jWz3BQ6F41Ts5JlysB/jw0G4g7NcaIpWc97THdZ1kAMQ7/3sOrPIZtZGkVfpN+mmf5x2tnVrBDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"06d6655939ac5f5fb56057a788d19b14743207846363e880ac576fae1d336198","last_reissued_at":"2026-07-05T04:00:25.952886Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:00:25.952886Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-Level Contrastive Learning for Cross-Lingual Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Beiduo Chen, Bin Gu, Quan Liu, Wu Guo, Yongchao Wang","submitted_at":"2022-02-26T07:14:20Z","abstract_excerpt":"Cross-language pre-trained models such as multilingual BERT (mBERT) have achieved significant performance in various cross-lingual downstream NLP tasks. This paper proposes a multi-level contrastive learning (ML-CTL) framework to further improve the cross-lingual ability of pre-trained models. The proposed method uses translated parallel data to encourage the model to generate similar semantic embeddings for different languages. However, unlike the sentence-level alignment used in most previous studies, in this paper, we explicitly integrate the word-level information of each pair of parallel "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.13083","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/2202.13083/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":"2202.13083","created_at":"2026-07-05T04:00:25.952944+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.13083v1","created_at":"2026-07-05T04:00:25.952944+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.13083","created_at":"2026-07-05T04:00:25.952944+00:00"},{"alias_kind":"pith_short_12","alias_value":"A3LGKWJZVRPV","created_at":"2026-07-05T04:00:25.952944+00:00"},{"alias_kind":"pith_short_16","alias_value":"A3LGKWJZVRPV7NLA","created_at":"2026-07-05T04:00:25.952944+00:00"},{"alias_kind":"pith_short_8","alias_value":"A3LGKWJZ","created_at":"2026-07-05T04:00:25.952944+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/A3LGKWJZVRPV7NLAK6TYRUM3CR","json":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR.json","graph_json":"https://pith.science/api/pith-number/A3LGKWJZVRPV7NLAK6TYRUM3CR/graph.json","events_json":"https://pith.science/api/pith-number/A3LGKWJZVRPV7NLAK6TYRUM3CR/events.json","paper":"https://pith.science/paper/A3LGKWJZ"},"agent_actions":{"view_html":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR","download_json":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR.json","view_paper":"https://pith.science/paper/A3LGKWJZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.13083&json=true","fetch_graph":"https://pith.science/api/pith-number/A3LGKWJZVRPV7NLAK6TYRUM3CR/graph.json","fetch_events":"https://pith.science/api/pith-number/A3LGKWJZVRPV7NLAK6TYRUM3CR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR/action/storage_attestation","attest_author":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR/action/author_attestation","sign_citation":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR/action/citation_signature","submit_replication":"https://pith.science/pith/A3LGKWJZVRPV7NLAK6TYRUM3CR/action/replication_record"}},"created_at":"2026-07-05T04:00:25.952944+00:00","updated_at":"2026-07-05T04:00:25.952944+00:00"}