{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YEQBCKEWQ76BVF3MT477Z6WEOC","short_pith_number":"pith:YEQBCKEW","schema_version":"1.0","canonical_sha256":"c12011289687fc1a976c9f3ffcfac4709a96535ffa484510a5e2451c69946edb","source":{"kind":"arxiv","id":"2502.08037","version":1},"attestation_state":"computed","paper":{"title":"Franken-Adapter: Cross-Lingual Adaptation of LLMs by Embedding Surgery","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fan Jiang, Grace Chung, Honglin Yu, Trevor Cohn","submitted_at":"2025-02-12T00:38:11Z","abstract_excerpt":"The capabilities of Large Language Models (LLMs) in low-resource languages lag far behind those in English, making their universal accessibility a significant challenge. To alleviate this, we present $\\textit{Franken-Adapter}$, a modular language adaptation approach for decoder-only LLMs with embedding surgery. Our method begins by creating customized vocabularies for target languages and performing language adaptation through embedding tuning on multilingual data. These pre-trained embeddings are subsequently integrated with LLMs that have been instruction-tuned on English alignment data to e"},"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":"2502.08037","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-12T00:38:11Z","cross_cats_sorted":[],"title_canon_sha256":"f9149e8fc41e3ad821b280ba82305633d2386f5c6c47efeafc1e2f218d8cd6ec","abstract_canon_sha256":"fbdf905e586bde9aa30addee460311445fb25409d9dc557a58a8e56d2ad92e07"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:14.215020Z","signature_b64":"PMZmnmGTFyUBpehtw6UmnhBvQiAkTlXjLWNqbEbBjOgQDF1y7LXPnLvv2Fg2UJQDzaejCx7A9+6tKOK+XMXACQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c12011289687fc1a976c9f3ffcfac4709a96535ffa484510a5e2451c69946edb","last_reissued_at":"2026-07-05T10:13:14.214486Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:14.214486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Franken-Adapter: Cross-Lingual Adaptation of LLMs by Embedding Surgery","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fan Jiang, Grace Chung, Honglin Yu, Trevor Cohn","submitted_at":"2025-02-12T00:38:11Z","abstract_excerpt":"The capabilities of Large Language Models (LLMs) in low-resource languages lag far behind those in English, making their universal accessibility a significant challenge. To alleviate this, we present $\\textit{Franken-Adapter}$, a modular language adaptation approach for decoder-only LLMs with embedding surgery. Our method begins by creating customized vocabularies for target languages and performing language adaptation through embedding tuning on multilingual data. These pre-trained embeddings are subsequently integrated with LLMs that have been instruction-tuned on English alignment data to e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.08037","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/2502.08037/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":"2502.08037","created_at":"2026-07-05T10:13:14.214556+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.08037v1","created_at":"2026-07-05T10:13:14.214556+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.08037","created_at":"2026-07-05T10:13:14.214556+00:00"},{"alias_kind":"pith_short_12","alias_value":"YEQBCKEWQ76B","created_at":"2026-07-05T10:13:14.214556+00:00"},{"alias_kind":"pith_short_16","alias_value":"YEQBCKEWQ76BVF3M","created_at":"2026-07-05T10:13:14.214556+00:00"},{"alias_kind":"pith_short_8","alias_value":"YEQBCKEW","created_at":"2026-07-05T10:13:14.214556+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.20720","citing_title":"COMPASS: COntinual Multilingual PEFT with Adaptive Semantic Sampling","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC","json":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC.json","graph_json":"https://pith.science/api/pith-number/YEQBCKEWQ76BVF3MT477Z6WEOC/graph.json","events_json":"https://pith.science/api/pith-number/YEQBCKEWQ76BVF3MT477Z6WEOC/events.json","paper":"https://pith.science/paper/YEQBCKEW"},"agent_actions":{"view_html":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC","download_json":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC.json","view_paper":"https://pith.science/paper/YEQBCKEW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.08037&json=true","fetch_graph":"https://pith.science/api/pith-number/YEQBCKEWQ76BVF3MT477Z6WEOC/graph.json","fetch_events":"https://pith.science/api/pith-number/YEQBCKEWQ76BVF3MT477Z6WEOC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC/action/storage_attestation","attest_author":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC/action/author_attestation","sign_citation":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC/action/citation_signature","submit_replication":"https://pith.science/pith/YEQBCKEWQ76BVF3MT477Z6WEOC/action/replication_record"}},"created_at":"2026-07-05T10:13:14.214556+00:00","updated_at":"2026-07-05T10:13:14.214556+00:00"}