--- title: "Mind the Gap in Cultural Alignment — paper by Zhang et al. (2026)" description: "The authors propose CultureManager to address both problems. 95% accuracy. 2 references, cited by 2 papers here." authors: ["Binchi Zhang", "Xujiang Zhao", "Jundong Li", "Haifeng Chen", "Zhengzhang Chen"] year: 2026 venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)" source: "https://aclanthology.org/2026.acl-long.766/" doi: "10.18653/v1/2026.acl-long.766" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/zhang-2026-mind-gap-cultural-alignment-task/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/zhang-2026-mind-gap-cultural-alignment-task/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/8/) / Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models # Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models **Authors:** Binchi Zhang, [Xujiang Zhao](https://md.research.elamur.ai/free/en/authors/xujiang-zhao/), [Jundong Li](https://md.research.elamur.ai/free/en/authors/jundong-li/), [Haifeng Chen](https://md.research.elamur.ai/free/en/authors/haifeng-chen/) and [Zhengzhang Chen](https://md.research.elamur.ai/free/en/authors/zhengzhang-chen/) · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2026.acl-long.766/) · [PDF](https://aclanthology.org/2026.acl-long.766.pdf) · [DOI 10.18653/v1/2026.acl-long.766](https://doi.org/10.18653/v1/2026.acl-long.766) · [OpenAlex](https://openalex.org/W7166798393) **Citations:** 0 citations (OpenAlex, 2026-10-14) · [2](https://md.research.elamur.ai/free/en/papers/zhang-2026-mind-gap-cultural-alignment-task/cited-by/) in our library, of the 149,439 papers whose bibliographies we read ## Summary Large language models often carry a Western-leaning cultural bias. Existing cultural alignment methods, such as persona prompting or fine-tuning on value surveys like WVS, have two weaknesses. Broad value data transfers poorly to concrete culture-sensitive tasks such as offensive language detection. Training one model on many cultures causes their conflicting norms to interfere with each other. The authors propose CultureManager to address both problems. It starts from task labels, a few unlabeled task examples and a list of target cultures. GPT-4o writes both task-specific and general cultural search queries. A lightweight SearchGPT agent then collects culturally grounded web pages, and GPT-4o summarizes them and turns them into labeled training samples in the target task's format. Each culture gets its own LoRA adapter trained on its pooled multi-task data. At inference, a zero-shot prompted router picks which adapter to use. In experiments spanning ten tasks where culture matters and the cultures of five countries, CultureManager consistently beats prompt-based and fine-tuning baselines. The results suggest that effective cultural alignment needs task-aware data and modular per-culture handling. ## Key points - **Data synthesis pipeline:** For each query, the search agent fetches the top-10 Google results. GPT-4o condenses them into a culture-grounded passage, which anchors the generated samples in real web content. - **Adapter settings:** - CultureLLM benchmark: LoRA rank 8, α=32, dropout 0.05, applied to q_proj and v_proj. - CulturalBench benchmark: rank 16, applied to all attention projections. - **Culture router:** It reaches over 95% accuracy on most tasks using a roughly 7B-parameter model with zero-shot prompting. Spam detection is the weakest task. - **Llama-3.3-70B-Instruct on Bengali:** CultureManager's average score is 50.73, compared with 41.46 for CultureLLM and 33.01 for the original model. Scale alone does not provide cultural competence. - **Further findings:** - Two-stage training (CultureSFT followed by synthetic task data) beats either stage alone on four of the five cultures. - Zero-shot GPT-5.1 matches or trails a fine-tuned 7B CultureManager on several tasks. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* ## Institutions - [University of Virginia](https://md.research.elamur.ai/free/en/universities/university-of-virginia/) — University, Charlottesville, United States · 36 works · 9,730 citations - [NEC (United States)](https://md.research.elamur.ai/free/en/companies/nec/) — Company, Irving, United States · 6 works · 539 citations 📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works 🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works ## Cite Cite as: Binchi Zhang, Xujiang Zhao, Jundong Li, Haifeng Chen and Zhengzhang Chen. 2026. Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{zhang2026mind, title = {{Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models}}, author = {Binchi Zhang, Xujiang Zhao, Jundong Li, Haifeng Chen and Zhengzhang Chen}, booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)}, year = {2026}, doi = {10.18653/v1/2026.acl-long.766}, url = {https://doi.org/10.18653/v1/2026.acl-long.766} } ``` - [← previous MetaLead: A Comprehensive Human-Curated Leaderboard Dataset for Transparent Reporting of Machine Learning Experiments](https://md.research.elamur.ai/free/en/papers/timmer-2026-metalead-comprehensive-human-curated-leaderboard/) - [next → Mind the Gap: Benchmarking LLM Uncertainty and Calibration with Specialty-Aware Clinical QA and Reasoning-Based Behavioural Features](https://md.research.elamur.ai/free/en/papers/testoni-2026-mind-gap-benchmarking-llm-uncertainty/) - [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/8/) ## References 2 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. This work was supported in part by the National Science Foundation (NSF) under Grants IIS-2144209, IIS-2223769, BCS-2228534, and CMMI-2411248, and by the Office of Naval Research (ONR) under Grant N000142412636. [find](https://scholar.google.com/scholar?q=This+work+was+supported+in+part+by+the+National+Science+Foundation+%28NSF%29+under+Grants+IIS%2D2144209%2C+IIS%2D2223769%2C+BCS%2D2228534%2C+and+CMMI%2D2411248%2C+and+by+the+Office+of+Naval+Research+%28ONR%29+under+Grant+N00) 2. Tuka Alhanai, Adam Kasumovic, Mohammad M Ghassemi, Aven Zitzelberger, Jessica M Lundin, and Guillaume Chabot-Couture. 2025. Bridging the gap: enhancing llm performance for low-resource african languages with new benchmarks, fine-tuning, and cultural adjustments. In Proceedings of the AAAI Conference… [find](https://scholar.google.com/scholar?q=Tuka+Alhanai%2C+Adam+Kasumovic%2C+Mohammad+M+Ghassemi%2C+Aven+Zitzelberger%2C+Jessica+M+Lundin%2C+and+Guillaume+Chabot%2DCouture%2E+2025%2E+Bridging+the+gap%3A+enhancing+llm+performance+for+low%2Dresource+african+languag) *Updated 31 October 2026*