{"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/language-imbalance-driven-rewarding-for","title":"Language Imbalance Driven Rewarding for Multilingual Self-improving","arxiv_id":"2410.08964","date":"2024-10-11","proceeding":null,"authors":["Wen Yang","Junhong Wu","Chen Wang","Chengqing Zong","Jiajun Zhang"],"abstract":"Large Language Models (LLMs) have achieved state-of-the-art performance across numerous tasks. However, these advancements have predominantly benefited \"first-class\" languages such as English and Chinese, leaving many other languages underrepresented. This imbalance, while limiting broader applications, generates a natural preference ranking between languages, offering an opportunity to bootstrap the multilingual capabilities of LLM in a self-improving manner. Thus, we propose $\\textit{Language Imbalance Driven Rewarding}$, where the inherent imbalance between dominant and non-dominant languages within LLMs is leveraged as a reward signal. Iterative DPO training demonstrates that this approach not only enhances LLM performance in non-dominant languages but also improves the dominant language's capacity, thereby yielding an iterative reward signal. Fine-tuning Meta-Llama-3-8B-Instruct over two iterations of this approach results in continuous improvements in multilingual performance across instruction-following and arithmetic reasoning tasks, evidenced by an average improvement of 7.46% win rate on the X-AlpacaEval leaderboard and 13.9% accuracy on the MGSM benchmark. This work serves as an initial exploration, paving the way for multilingual self-improvement of LLMs.","url_abs":"https://arxiv.org/abs/2410.08964v2","url_pdf":"https://arxiv.org/pdf/2410.08964v2.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":"language-imbalance-driven-rewarding-for","repo_url":"https://github.com/znlp/language-imbalance-driven-rewarding","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"arithmetic-reasoning","task_name":"Arithmetic Reasoning"},{"task_slug":"instruction-following","task_name":"Instruction Following"}],"methods":[{"method_slug":"dpo","method_name":"DPO"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.08964","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.08964"}},"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/ZNLP/Language-Imbalance-Driven-Rewarding","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/znlp/language-imbalance-driven-rewarding","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":2,"ran":6,"unverified":5},"by_repo_kind":{"official":{"samples":13,"ran":8,"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":13,"samples":[{"code_sha256_prefix":"c85bc1200cde30f4","entry":"generator","repo":"znlp/language-imbalance-driven-rewarding","repo_kind":"official","path":"utils/batch_inference.py","file_url":"https://github.com/znlp/language-imbalance-driven-rewarding/blob/HEAD/utils/batch_inference.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c85bc1200cde30f4"}},{"code_sha256_prefix":"11e86e115ad397d4","entry":"get_file_names","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"utils/utils.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/utils/utils.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":"11e86e115ad397d4"}},{"code_sha256_prefix":"0c77763ca4cd8bc4","entry":"get_metrics_evaluator","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/analyze.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/analyze.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0c77763ca4cd8bc4"}},{"code_sha256_prefix":"80319fda62fe1a04","entry":"lmsys_parser","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/completion_parsers.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/completion_parsers.py","link_basis":"plan_row","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":"80319fda62fe1a04"}},{"code_sha256_prefix":"10b24db4baa49194","entry":"load_json","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"utils/utils.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/utils/utils.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":"10b24db4baa49194"}},{"code_sha256_prefix":"4ccfe92e7eff858a","entry":"load_jsonl","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"utils/utils.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/utils/utils.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":"4ccfe92e7eff858a"}},{"code_sha256_prefix":"52f4a49b6393e97e","entry":"ranking_parser","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/completion_parsers.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/completion_parsers.py","link_basis":"plan_row","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":"52f4a49b6393e97e"}},{"code_sha256_prefix":"307648f4b5ffd08b","entry":"show_length_gameability","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/plotting.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/plotting.py","link_basis":"plan_row","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":"307648f4b5ffd08b"}},{"code_sha256_prefix":"304d6b64de10aa4f","entry":"get_alpaca_eval_data","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/constants.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/constants.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":"304d6b64de10aa4f"}},{"code_sha256_prefix":"0eb53db38f0faafd","entry":"get_annotations","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/analyze.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/analyze.py","link_basis":"plan_row","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":"0eb53db38f0faafd"}},{"code_sha256_prefix":"de8a267bb7f7d4bd","entry":"get_crossannotations","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/analyze.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/analyze.py","link_basis":"plan_row","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":"de8a267bb7f7d4bd"}},{"code_sha256_prefix":"be8e5acf8fee5a8f","entry":"plot_heatmap_diff_baseline","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/plotting.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/plotting.py","link_basis":"plan_row","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":"be8e5acf8fee5a8f"}},{"code_sha256_prefix":"6f7ea5efc95b3173","entry":"regex_parser","repo":"ZNLP/Language-Imbalance-Driven-Rewarding","repo_kind":"official","path":"alpaca_eval/src/alpaca_eval/completion_parsers.py","file_url":"https://github.com/ZNLP/Language-Imbalance-Driven-Rewarding/blob/HEAD/alpaca_eval/src/alpaca_eval/completion_parsers.py","link_basis":"plan_row","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":"6f7ea5efc95b3173"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}