{"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/plug-leveraging-pivot-language-in-cross","title":"PLUG: Leveraging Pivot Language in Cross-Lingual Instruction Tuning","arxiv_id":"2311.08711","date":"2023-11-15","proceeding":null,"authors":["Zhihan Zhang","Dong-Ho Lee","Yuwei Fang","Wenhao Yu","Mengzhao Jia","Meng Jiang","Francesco Barbieri"],"abstract":"Instruction tuning has remarkably advanced large language models (LLMs) in understanding and responding to diverse human instructions. Despite the success in high-resource languages, its application in lower-resource ones faces challenges due to the imbalanced foundational abilities of LLMs across different languages, stemming from the uneven language distribution in their pre-training data. To tackle this issue, we propose pivot language guided generation (PLUG), an approach that utilizes a high-resource language, primarily English, as the pivot to enhance instruction tuning in lower-resource languages. It trains the model to first process instructions in the pivot language, and then produce responses in the target language. To evaluate our approach, we introduce a benchmark, X-AlpacaEval, of instructions in 4 languages (Chinese, Korean, Italian, and Spanish), each annotated by professional translators. Our approach demonstrates a significant improvement in the instruction-following abilities of LLMs by 29% on average, compared to directly responding in the target language alone. Further experiments validate the versatility of our approach by employing alternative pivot languages beyond English to assist languages where LLMs exhibit lower proficiency. Our code and data are available at https://github.com/ytyz1307zzh/PLUG.","url_abs":"https://arxiv.org/abs/2311.08711v2","url_pdf":"https://arxiv.org/pdf/2311.08711v2.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":"plug-leveraging-pivot-language-in-cross","repo_url":"https://github.com/ytyz1307zzh/plug","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"instruction-following","task_name":"Instruction Following"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2311.08711","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2311.08711"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ytyz1307zzh/plug","reach":{"status":"ok"}}],"summary":{"ran":10,"unverified":1},"by_repo_kind":{"official":{"samples":11,"ran":10,"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":11,"samples":[{"code_sha256_prefix":"3e78d709f22fd30c","entry":"extract_answer","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/evaluate/svamp/evaluate_svamp.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/evaluate/svamp/evaluate_svamp.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3e78d709f22fd30c"}},{"code_sha256_prefix":"01ca766b9e5bf4ee","entry":"extract_first_response","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/model/extract_response_from_bilingual_output.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/model/extract_response_from_bilingual_output.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"01ca766b9e5bf4ee"}},{"code_sha256_prefix":"23a42f5111a6b386","entry":"extract_second_response","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/model/extract_response_from_bilingual_output.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/model/extract_response_from_bilingual_output.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"23a42f5111a6b386"}},{"code_sha256_prefix":"36b1fcfb79a1b4a1","entry":"format_answer_list","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/evaluate/truthfulqa/evaluate_truthfulqa_gpt4.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/evaluate/truthfulqa/evaluate_truthfulqa_gpt4.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"36b1fcfb79a1b4a1"}},{"code_sha256_prefix":"7b3b049413a50845","entry":"format_input_prompt","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/model/inference.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/model/inference.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7b3b049413a50845"}},{"code_sha256_prefix":"8615bd4175ccc380","entry":"is_correct","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/evaluate/svamp/evaluate_svamp.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/evaluate/svamp/evaluate_svamp.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8615bd4175ccc380"}},{"code_sha256_prefix":"1010d5fc3c98b8e5","entry":"mean","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/translation/translate_together_chatgpt.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/translation/translate_together_chatgpt.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1010d5fc3c98b8e5"}},{"code_sha256_prefix":"9db179dfefb3aa4c","entry":"read_answer_file","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/evaluate/x-alpacaeval/llm_judge/llm_judge.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/evaluate/x-alpacaeval/llm_judge/llm_judge.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9db179dfefb3aa4c"}},{"code_sha256_prefix":"661ea0d3eb31d657","entry":"read_jsonl_as_list","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/evaluate/truthfulqa/evaluate_truthfulqa_gpt4.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/evaluate/truthfulqa/evaluate_truthfulqa_gpt4.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"661ea0d3eb31d657"}},{"code_sha256_prefix":"43a372b82bfe8052","entry":"read_question_file","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/evaluate/x-alpacaeval/llm_judge/llm_judge.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/evaluate/x-alpacaeval/llm_judge/llm_judge.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"43a372b82bfe8052"}},{"code_sha256_prefix":"4df12d19ed56280d","entry":"chatgpt_single_turn_inference","repo":"ytyz1307zzh/plug","repo_kind":"official","path":"src/utils/gpt_utils.py","file_url":"https://github.com/ytyz1307zzh/plug/blob/HEAD/src/utils/gpt_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4df12d19ed56280d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}