{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VQMUI7N2VKNAYNKUP2FWEPSKTZ","short_pith_number":"pith:VQMUI7N2","schema_version":"1.0","canonical_sha256":"ac19447dbaaa9a0c35547e8b623e4a9e677516d89977b9e1d09c1a3a9fc2ad93","source":{"kind":"arxiv","id":"2410.01335","version":2},"attestation_state":"computed","paper":{"title":"Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Benjamin Muller, Bing Liu, Hongjiang Lv, Lucas Bandarkar, Nayan Singhal, Pritish Yuvraj, Rui Hou","submitted_at":"2024-10-02T08:53:07Z","abstract_excerpt":"Model merging, such as model souping, is the practice of combining different models with the same architecture together without further training. In this work, we present a model merging methodology that addresses the difficulty of fine-tuning Large Language Models (LLMs) for target tasks in non-English languages, where task-specific data is often unavailable. We focus on mathematical reasoning and without in-language math data, facilitate cross-lingual transfer by composing language and math capabilities. Starting from the same pretrained model, we fine-tune separate \"experts\" on math instruc"},"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":"2410.01335","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-02T08:53:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"db5ce6bcb98debff2eb8e2743c18a79d682569fc81bdacafbf239630a2178d0f","abstract_canon_sha256":"bd9cafb601d74d78892c32e66d0dedfd34d1f81677dbe8a9a73ff39f1fc19c20"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:38.836074Z","signature_b64":"VmRXO7K5Sngbo/Tn5dsO3BD8jDq4eRXjnq/JdIxqMqt4iSMLOudPSAQfi8upAiE3lNeETxRtC58ZOiGO6wpZAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac19447dbaaa9a0c35547e8b623e4a9e677516d89977b9e1d09c1a3a9fc2ad93","last_reissued_at":"2026-07-05T11:08:38.835616Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:38.835616Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Benjamin Muller, Bing Liu, Hongjiang Lv, Lucas Bandarkar, Nayan Singhal, Pritish Yuvraj, Rui Hou","submitted_at":"2024-10-02T08:53:07Z","abstract_excerpt":"Model merging, such as model souping, is the practice of combining different models with the same architecture together without further training. In this work, we present a model merging methodology that addresses the difficulty of fine-tuning Large Language Models (LLMs) for target tasks in non-English languages, where task-specific data is often unavailable. We focus on mathematical reasoning and without in-language math data, facilitate cross-lingual transfer by composing language and math capabilities. Starting from the same pretrained model, we fine-tune separate \"experts\" on math instruc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01335","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/2410.01335/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":"2410.01335","created_at":"2026-07-05T11:08:38.835672+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.01335v2","created_at":"2026-07-05T11:08:38.835672+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01335","created_at":"2026-07-05T11:08:38.835672+00:00"},{"alias_kind":"pith_short_12","alias_value":"VQMUI7N2VKNA","created_at":"2026-07-05T11:08:38.835672+00:00"},{"alias_kind":"pith_short_16","alias_value":"VQMUI7N2VKNAYNKU","created_at":"2026-07-05T11:08:38.835672+00:00"},{"alias_kind":"pith_short_8","alias_value":"VQMUI7N2","created_at":"2026-07-05T11:08:38.835672+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.26735","citing_title":"Rethinking the Multilingual Reasoning Gap with Layer Swap","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ","json":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ.json","graph_json":"https://pith.science/api/pith-number/VQMUI7N2VKNAYNKUP2FWEPSKTZ/graph.json","events_json":"https://pith.science/api/pith-number/VQMUI7N2VKNAYNKUP2FWEPSKTZ/events.json","paper":"https://pith.science/paper/VQMUI7N2"},"agent_actions":{"view_html":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ","download_json":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ.json","view_paper":"https://pith.science/paper/VQMUI7N2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.01335&json=true","fetch_graph":"https://pith.science/api/pith-number/VQMUI7N2VKNAYNKUP2FWEPSKTZ/graph.json","fetch_events":"https://pith.science/api/pith-number/VQMUI7N2VKNAYNKUP2FWEPSKTZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ/action/storage_attestation","attest_author":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ/action/author_attestation","sign_citation":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ/action/citation_signature","submit_replication":"https://pith.science/pith/VQMUI7N2VKNAYNKUP2FWEPSKTZ/action/replication_record"}},"created_at":"2026-07-05T11:08:38.835672+00:00","updated_at":"2026-07-05T11:08:38.835672+00:00"}