{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:72667HI23TTINGIDCZ4EJJB5ZN","short_pith_number":"pith:72667HI2","schema_version":"1.0","canonical_sha256":"febdef9d1adce6869903167844a43dcb6e460d446ff5969a27fa4ec6ab892d83","source":{"kind":"arxiv","id":"2605.29414","version":1},"attestation_state":"computed","paper":{"title":"Beyond Bilingual Transfer: Multilingual Code-Switching in Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jeonghun Baek, Shunta Asano, Toshihiko Yamasaki","submitted_at":"2026-05-28T06:03:52Z","abstract_excerpt":"Recent studies have shown that code-switching data (CSD), in which multiple languages are mixed within the same context, can improve cross-lingual transfer and multilingual alignment in large language models (LLMs). However, existing studies primarily focus on bilingual transfer between English and a target language, leaving multilingual settings involving three or more languages largely unexplored. In this work, we investigate multilingual code-switching instruction tuning across four languages: English, Japanese, Korean, and Chinese. We evaluate multilingual understanding on Belebele. Our ex"},"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":"2605.29414","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-28T06:03:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"83a06dc3388941225fec3cc0058d6e1489b1f3c0c2bf81df5b3b52843cf3c227","abstract_canon_sha256":"7cf2d4772b4e971b75ba989482f4d3a7dc2306ddac6fe443026c0849d3ddc7ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:37.676337Z","signature_b64":"kykYfU3c7g8nyaiBz1MQBn5fb7gvVFlspdQLNqnemnlaYM4W/8QvbbRmtx6To3KpqoJot4ejUZZpM3BkMT/+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"febdef9d1adce6869903167844a43dcb6e460d446ff5969a27fa4ec6ab892d83","last_reissued_at":"2026-05-29T01:05:37.675891Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:37.675891Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Bilingual Transfer: Multilingual Code-Switching in Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jeonghun Baek, Shunta Asano, Toshihiko Yamasaki","submitted_at":"2026-05-28T06:03:52Z","abstract_excerpt":"Recent studies have shown that code-switching data (CSD), in which multiple languages are mixed within the same context, can improve cross-lingual transfer and multilingual alignment in large language models (LLMs). However, existing studies primarily focus on bilingual transfer between English and a target language, leaving multilingual settings involving three or more languages largely unexplored. In this work, we investigate multilingual code-switching instruction tuning across four languages: English, Japanese, Korean, and Chinese. We evaluate multilingual understanding on Belebele. Our ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29414","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/2605.29414/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":"2605.29414","created_at":"2026-05-29T01:05:37.675952+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29414v1","created_at":"2026-05-29T01:05:37.675952+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29414","created_at":"2026-05-29T01:05:37.675952+00:00"},{"alias_kind":"pith_short_12","alias_value":"72667HI23TTI","created_at":"2026-05-29T01:05:37.675952+00:00"},{"alias_kind":"pith_short_16","alias_value":"72667HI23TTINGID","created_at":"2026-05-29T01:05:37.675952+00:00"},{"alias_kind":"pith_short_8","alias_value":"72667HI2","created_at":"2026-05-29T01:05:37.675952+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/72667HI23TTINGIDCZ4EJJB5ZN","json":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN.json","graph_json":"https://pith.science/api/pith-number/72667HI23TTINGIDCZ4EJJB5ZN/graph.json","events_json":"https://pith.science/api/pith-number/72667HI23TTINGIDCZ4EJJB5ZN/events.json","paper":"https://pith.science/paper/72667HI2"},"agent_actions":{"view_html":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN","download_json":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN.json","view_paper":"https://pith.science/paper/72667HI2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29414&json=true","fetch_graph":"https://pith.science/api/pith-number/72667HI23TTINGIDCZ4EJJB5ZN/graph.json","fetch_events":"https://pith.science/api/pith-number/72667HI23TTINGIDCZ4EJJB5ZN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN/action/storage_attestation","attest_author":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN/action/author_attestation","sign_citation":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN/action/citation_signature","submit_replication":"https://pith.science/pith/72667HI23TTINGIDCZ4EJJB5ZN/action/replication_record"}},"created_at":"2026-05-29T01:05:37.675952+00:00","updated_at":"2026-05-29T01:05:37.675952+00:00"}