We present a survey of 39 recent papers that aim to study cultural
representation and inclusion in large language models. We observe that none of
the studies define "culture," which is a complex, multifaceted concept;
instead, they probe the models on some specially designed datasets which
represent certain aspects of "culture." We call these aspects the proxies of
cultures, and organize them across three dimensions of demographic, semantic
and linguistic-cultural interaction proxies. We also categorize the probing
methods employed. Our analysis indicates that only certain aspects of
"culture," such as values and objectives, have been studied, leaving several
other interesting and important facets, especially the multitude of semantic
domains (Thompson et al., 2020) and aboutness (Hershcovich et al., 2022),
unexplored. Two other crucial gaps are the lack of robustness and situatedness
of the current methods. Based on these observations, we provide several
recommendations for a holistic and practically useful research agenda for
furthering cultural inclusion in LLMs and LLM-based applications.

对研究大型语言模型中文化表示和包容性的 39 篇最新论文进行概述，发现这些研究中没有明确定义 “文化”，而是通过一些特殊设计的数据集来探索模型，这些数据集代表了文化的某些方面；我们称这些方面为文化的代理，按人口统计学、语义和语言文化交互代理三个维度将其组织起来，并对使用的探测方法进行分类。分析表明，只有文化的某些方面，如价值观和目标，得到了研究，其他一些有趣且重要的方面，尤其是语义领域的多样性和主题向度，尚未被探索。另外，目前的方法缺乏鲁棒性和情境性。基于这些观察，提供了关于进一步促进 LLM 和基于 LLM 的应用中文化包容的全面可行的研究议程的几点建议。

测量和建模 LLMs 中的 “文化”: 一项调查

Towards Measuring and Modeling "Culture" in LLMs: A Survey

The rapid development of Large Language Models (LLMs) and the emergence of
novel abilities with scale have necessitated the construction of holistic,
diverse and challenging benchmarks such as HELM and BIG-bench. However, at the
moment, most of these benchmarks focus only on performance in English and
evaluations that include Southeast Asian (SEA) languages are few in number. We
therefore propose BHASA, a holistic linguistic and cultural evaluation suite
for LLMs in SEA languages. It comprises three components: (1) a NLP benchmark
covering eight tasks across Natural Language Understanding (NLU), Generation
(NLG) and Reasoning (NLR) tasks, (2) LINDSEA, a linguistic diagnostic toolkit
that spans the gamut of linguistic phenomena including syntax, semantics and
pragmatics, and (3) a cultural diagnostics dataset that probes for both
cultural representation and sensitivity. For this preliminary effort, we
implement the NLP benchmark only for Indonesian, Vietnamese, Thai and Tamil,
and we only include Indonesian and Tamil for LINDSEA and the cultural
diagnostics dataset. As GPT-4 is purportedly one of the best-performing
multilingual LLMs at the moment, we use it as a yardstick to gauge the
capabilities of LLMs in the context of SEA languages. Our initial experiments
on GPT-4 with BHASA find it lacking in various aspects of linguistic
capabilities, cultural representation and sensitivity in the targeted SEA
languages. BHASA is a work in progress and will continue to be improved and
expanded in the future.

在东南亚语言方面，BHASA 提供了一个综合的语言和文化评估套件，其中包括自然语言理解、生成和推理等任务的 NLP 基准、语言诊断工具包 LINDSEA 以及文化诊断数据集，初步实验发现 GPT-4 在东南亚语言方面的语言能力、文化表达和敏感性等方面存在不足，BHASA 将在未来继续改进和扩展。

BHASA: 东南亚语言文化综合评估系统

BHASA: A Holistic Southeast Asian Linguistic and Cultural Evaluation  Suite for Large Language Models

Fashion is one of the ways in which we show ourselves to the world. It is a
reflection of our personal decisions and one of the ways in which people
distinguish and represent themselves. In this paper, we focus on the fashion
design process and expand computer vision for fashion beyond its current focus
on western fashion. We discuss the history of Southern African se-Shweshwe
fabric fashion, the collection of a se-Shweshwe dataset, and the application of
sketch-to-design image generation for affordable fashion-design. The
application to fashion raises both technical questions of training with small
amounts of data, and also important questions for computer vision beyond
fairness, in particular ethical considerations on creating and employing
fashion datasets, and how computer vision supports cultural representation and
might avoid algorithmic cultural appropriation.

本文主要关注时尚设计过程中的计算机视觉应用，特别是在南非 se-Shweshwe 面料时尚领域中的应用，包括数据集收集，图像生成，培训技术及计算机视觉在文化代表性和算法文化适当性方面的重要问题。