Abstract
The rapid evolution of Large Language Models (LLMs) has transformed natural language processing but raises critical concerns about biases inherent in their deployment and use across diverse linguistic and sociocultural contexts. This paper presents a framework named ASCenD BDS (Adaptable, Stochastic and Context-aware framework for Detection of Bias, Discrimination and Stereotyping). The framework presents approach to detecting bias, discrimination, stereotyping across various categories such as gender, caste, age, disability, socioeconomic status, linguistic variations, etc., using an approach which is Adaptive, Stochastic and Context-Aware. The existing frameworks rely heavily on usage of datasets to generate scenarios for detection of Bias, Discrimination and Stereotyping. Examples include datasets such as Civil Comments, Wino Gender, WinoBias, BOLD, CrowS Pairs and BBQ. However, such an approach provides point solutions. As a result, these datasets provide a finite number of scenarios for assessment. The current framework overcomes this limitation by having features which enable Adaptability, Stochasticity, Context Awareness. Context awareness can be customized for any nation or culture or sub-culture (for example an organization's unique culture). In this paper, context awareness in the Indian context has been established. Content has been leveraged from Indian Census 2011 to have a commonality of categorization. A framework has been developed using Category, Sub-Category, STEM, X-Factor, Synonym to enable the features for Adaptability, Stochasticity and Context awareness. The framework has been described in detail in Section 3. Overall 800 plus STEMs, 10 Categories, 31 unique SubCategories were developed by a team of consultants at Saint Fox Consultancy Private Ltd. The concept has been tested out in SFCLabs as part of product development.
Abstract (translated)
大型语言模型（LLMs）的快速演变已经改变了自然语言处理领域，但同时也引发了关于其在多样化的语言和社会文化背景中部署和使用时内在偏见的重要担忧。本文提出了一种名为ASCenD BDS（可适应、随机化及上下文感知框架用于检测偏见、歧视与刻板印象）的框架。该框架提供了一种方法来检测跨多个类别（如性别、种姓、年龄、残疾、经济和社会地位、语言变化等）的偏见、歧视和刻板印象，其方法是可适应性、随机化及上下文感知。 现有的框架主要依赖于使用数据集生成检测偏见、歧视与刻板印象的情景。例如，Civil Comments、Wino Gender、WinoBias、BOLD、CrowS Pairs 和 BBQ 数据集等。然而，这种方法只提供点对点解决方案。因此，这些数据集仅提供了有限数量的评估情景。目前提出的框架通过具备可适应性、随机化及上下文感知功能克服了这一限制。 上下文感知可以根据任何国家或文化（例如一个组织的独特文化）进行定制。在本文中，已在印度语境下建立了上下文感知概念。内容利用来自2011年印度人口普查的数据来实现分类的一致性。通过类别、子类别、STEM（情境-主题-情感）、X-Factor 和同义词开发了一种框架，以支持可适应性、随机化及上下文感知的功能。该框架已在第3节中详细描述。 总体而言，圣福咨询私人有限公司的顾问团队开发了800多个STEM、10个类别和31个独特的子类别。这一概念已在SFCLabs的产品研发过程中进行了测试。
URL
https://arxiv.org/abs/2502.02072
