通知：联邦节假日调休：哥伦布日 / 原住民日（Columbus Day / Indigenous Peoples' Day），今日（13日）无更新。通知：联邦节假日调休：哥伦布日 / 原住民日（Columbus Day / Indigenous Peoples' Day），今日（13日）无更新。
arXiv 2603.16827cs.AIcs.CL
提示编程用于大型语言模型的文化偏差与对齐
Prompt Programming for Cultural Bias and Alignment of Large Language Models
Information Systems and Modeling, Los Alamos National Laboratory（信息系统与建模，洛斯阿拉莫斯国家实验室）；Advanced Research in Cyber Systems, Los Alamos National Laboratory（网络安全系统高级研究，洛斯阿拉莫斯国家实验室）；Center for National Security and International Studies（国家安全与国际研究中心）；Analytics Division, Los Alamos National Laboratory（分析部门，洛斯阿拉莫斯国家实验室）
Maksim Eren, Eric Michalak, Brian Cook, Johnny Seales
Culture shapes reasoning, values, prioritization, and strategic decision-making, yet large language models (LLMs) often exhibit cultural biases that misalign with target populations. As LLMs are increasingly used for strategic decision-making, policy support, and document engineering tasks such as summarization, categorization, and compliance-oriented auditing, improving cultural alignment is important for ensuring that downstream analyses and recommendations reflect target-population value profiles rather than default model priors. Previous work introduced a survey-grounded cultural alignment framework and showed that culture-specific prompting can reduce misalignment, but it primarily evaluated proprietary models and relied on manual prompt engineering. In this paper, we validate and extend that framework by reproducing its social sciences survey based projection and distance metrics on open-weight LLMs, testing whether the same cultural skew and benefits of culture conditioning persist outside closed LLM systems. Building on this foundation, we introduce use of prompt programming with DSPy for this problem-treating prompts as modular, optimizable programs-to systematically tune cultural conditioning by optimizing against cultural-distance objectives. In our experiments, we show that prompt optimization often improves upon cultural prompt engineering, suggesting prompt compilation with DSPy can provide a more stable and transferable route to culturally aligned LLM responses.
LLMs and Cultural Values: the Impact of Prompt Language and Explicit Cultural Framing
Brussels Centre for Language Studies, Vrije Universiteit Brussel（布鲁塞尔语言研究中心，布鲁塞尔自由大学）；Université de Montréal & Mila - Quebec AI Institute（蒙特利尔大学及魁北克人工智能研究所）
Whispers of Many Shores: Cultural Alignment through Collaborative Cultural Expertise
Arizona State University（亚利桑那州立大学）；National Taiwan University（台湾国立大学）；Carnegie Mellon University（卡内基梅隆大学）；New York University（纽约大学）；Indian Institute of Technology Hyderabad（印度海得拉巴理工学院）；Minitab（Minitab公司）；Trinity College Dublin（都柏林圣三一学院）
TukaBench: A Culturally Grounded Jailbreak Benchmark for African Languages
Mila - Quebec AI Institute（魁北克人工智能研究院）；McGill University（麦吉尔大学）；Microsoft AI for Good Research Lab（微软人工智能造福人类研究实验室）；Canada CIFAR AI Chair（加拿大CIFAR人工智能主席）
Register Shifts Break LLM Safety: A Bengali Benchmark with Culturally Grounded Harms
Institute of Information Technology, University of Dhaka（达卡大学信息技术学院）；University of Maryland, Baltimore County（马里兰大学巴尔的摩县分校）；Bangladesh Air Force Shaheen College Kurmitola（孟加拉国空军库尔米托拉沙欣学院）；Ciroos Inc.（Ciroos公司）
Department of Linguistics, Tongji University（同济大学语言学系）；Department of Computational Linguistics, Tübingen University（图宾根大学计算语言学系）；The Institute of Natural Language Processing, Stuttgart University（斯图加特大学自然语言处理研究所）
Culture-TRIP: Culturally-Aware Text-to-Image Generation with Iterative Prompt Refinement
Department of Computer Science and Engineering, Dongguk University（计算机科学与工程系，东国大学）；Department of Computer Science and Artificial Intelligence, Dongguk University（计算机科学与人工智能系，东国大学）
Institute for Clarity in Documentation（文档清晰性研究所）；Inria Paris-Rocquencourt（法国国家信息与自动化研究所巴黎-罗克库尔中心）；Rajiv Gandhi University（拉吉夫·甘地大学）；Tsinghua University（清华大学）；Palmer Research Laboratories（帕尔默研究实验室）
Prevailing methods for mapping large generative language models to supervised tasks may fail to sufficiently probe models' novel capabilities. Using GPT-3 as a case study, we show that 0-shot prompts can significantly outperform few-shot prompts. We suggest that the function of few-shot examples in these cases is better described as locating an already learned task rather than meta-learning. This analysis motivates rethinking the role of prompts in controlling and evaluating powerful language models. In this work, we discuss methods of prompt programming, emphasizing the usefulness of consider…
CULEMO: Cultural Lenses on Emotion -- Benchmarking LLMs for Cross-Cultural Emotion Understanding
Instituto Politécnico Nacional（墨西哥国立理工学院）；Haverford College（哈弗福德学院）；Wollo University（沃尔洛大学）；Pennsylvania State University（宾夕法尼亚州立大学）；University of Hamburg（汉堡大学）
Counterfactual Cultural Cues Reduce Medical QA Accuracy in LLMs: Identifier vs Context Effects
Institute of Health Policy, Management, and Evaluation (IHPME)（健康政策、管理与评估研究所）；Dalla Lana School of Public Health（达拉兰公共卫生学院）；University of Toronto（多伦多大学）；Faculty of Information（信息学院）；Schwartz Reisman Institute（施瓦茨-雷曼研究所）
DFKI-MLT at SemEval-2026 TASK 7: Steering Multilingual Models Towards Cultural Knowledge
German Research Center for Artificial Intelligence (DFKI GmbH)（德国人工智能研究中心（DFKI GmbH））；Saarland Informatics Campus（萨尔布吕肯信息学校区）；Barcelona Supercomputing Center (BSC-CNS)（巴塞罗那超级计算中心（BSC-CNS））
Do LLMs Use Cultural Knowledge Without Being Told? A Multilingual Evaluation of Implicit Pragmatic Adaptation
Network Analysis and Social Influence Modelling (NASIM) Lab（网络分析与社会影响力建模实验室）；School of Physics Maths and Computing（物理数学与计算学院）；The University of Western Australia（西澳大学）；School of Computing（计算学院）；Macquarie University（麦考瑞大学）；Computational Social Science & Systems Analysis（计算社会科学与系统分析）；University of Münster（明斯特大学）
