{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/breaking-the-curse-of-multilinguality-with","title":"Breaking the Curse of Multilinguality with Cross-lingual Expert Language Models","arxiv_id":"2401.10440","date":"2024-01-19","proceeding":null,"authors":["Terra Blevins","Tomasz Limisiewicz","Suchin Gururangan","Margaret Li","Hila Gonen","Noah A. Smith","Luke Zettlemoyer"],"abstract":"Despite their popularity in non-English NLP, multilingual language models often underperform monolingual ones due to inter-language competition for model parameters. We propose Cross-lingual Expert Language Models (X-ELM), which mitigate this competition by independently training language models on subsets of the multilingual corpus. This process specializes X-ELMs to different languages while remaining effective as a multilingual ensemble. Our experiments show that when given the same compute budget, X-ELM outperforms jointly trained multilingual models across all considered languages and that these gains transfer to downstream tasks. X-ELM provides additional benefits over performance improvements: new experts can be iteratively added, adapting X-ELM to new languages without catastrophic forgetting. Furthermore, training is asynchronous, reducing the hardware requirements for multilingual training and democratizing multilingual modeling.","url_abs":"https://arxiv.org/abs/2401.10440v2","url_pdf":"https://arxiv.org/pdf/2401.10440v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"breaking-the-curse-of-multilinguality-with","repo_url":"https://github.com/blvns/x-elm","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2401.10440","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2401.10440"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/blvns/x-elm","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":6,"unverified":4},"by_repo_kind":{"official":{"samples":10,"ran":6,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"8060f2ec5c0f9539","entry":"average","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/sequence_scorer_btm.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/sequence_scorer_btm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8060f2ec5c0f9539"}},{"code_sha256_prefix":"840d9070e4fd8ba6","entry":"find_offsets","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/file_chunker_utils.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/file_chunker_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"840d9070e4fd8ba6"}},{"code_sha256_prefix":"c9fd54c09adfb25d","entry":"load_json","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/file_io.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/file_io.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c9fd54c09adfb25d"}},{"code_sha256_prefix":"cec2774d64ea9719","entry":"recursively_cast_dictconfigs","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/file_io.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/file_io.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cec2774d64ea9719"}},{"code_sha256_prefix":"9418b43048e4e4ec","entry":"url_to_filename","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/file_utils.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/file_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9418b43048e4e4ec"}},{"code_sha256_prefix":"1dcaa0202f989ada","entry":"with_incremental_state","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/incremental_decoding_utils.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/incremental_decoding_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1dcaa0202f989ada"}},{"code_sha256_prefix":"aafdc12f01ab388f","entry":"filename_to_url","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/file_utils.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/file_utils.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"aafdc12f01ab388f"}},{"code_sha256_prefix":"0d372aa1e290a613","entry":"load_archive_file","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/file_utils.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/file_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0d372aa1e290a613"}},{"code_sha256_prefix":"e5ec2c0476e20add","entry":"load_model","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/sequence_scorer_btm.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/sequence_scorer_btm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e5ec2c0476e20add"}},{"code_sha256_prefix":"b2e331c14695c39b","entry":"torch_load_cpu","repo":"blvns/x-elm","repo_kind":"official","path":"metaseq/file_io.py","file_url":"https://github.com/blvns/x-elm/blob/HEAD/metaseq/file_io.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b2e331c14695c39b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}