As question answering (QA) systems advance alongside the rapid evolution of foundation models, the need for robust, adaptable, and large-scale evaluation benchmarks becomes increasingly critical. Traditional QA benchmarks are often static and publicly available, making them susceptible to data contamination and memorization by large language models (LLMs). Consequently, static benchmarks may overestimate model generalization and hinder a reliable assessment of real-world performance. In this work, we introduce Dynamic-KGQA, a scalable framework for generating adaptive QA datasets from knowledge graphs (KGs), designed to mitigate memorization risks while maintaining statistical consistency across iterations. Unlike fixed benchmarks, Dynamic-KGQA generates a new dataset variant on every run while preserving the underlying distribution, enabling fair and reproducible evaluations. Furthermore, our framework provides fine-grained control over dataset characteristics, supporting domain-specific and topic-focused QA dataset generation. Additionally, Dynamic-KGQA produces compact, semantically coherent subgraphs that facilitate both training and evaluation of KGQA models, enhancing their ability to leverage structured knowledge effectively. To align with existing evaluation protocols, we also provide static large-scale train/test/validation splits, ensuring comparability with prior methods. By introducing a dynamic, customizable benchmarking paradigm, Dynamic-KGQA enables a more rigorous and adaptable evaluation of QA systems.

Large language models (LLMs) have emerged as an integral part of modern societies, powering user-facing applications such as personal assistants and enterprise applications like recruitment tools. Despite their utility, research indicates that LLMs perpetuate systemic biases. Yet, prior works on LLM harms predominantly focus on Western concepts like race and gender, often overlooking cultural concepts from other parts of the world. Additionally, these studies typically investigate "harm" as a singular dimension, ignoring the various and subtle forms in which harms manifest. To address this gap, we introduce the Covert Harms and Social Threats (CHAST), a set of seven metrics grounded in social science literature. We utilize evaluation models aligned with human assessments to examine the presence of covert harms in LLM-generated conversations, particularly in the context of recruitment. Our experiments reveal that seven out of the eight LLMs included in this study generated conversations riddled with CHAST, characterized by malign views expressed in seemingly neutral language unlikely to be detected by existing methods. Notably, these LLMs manifested more extreme views and opinions when dealing with non-Western concepts like caste, compared to Western ones such as race.

Often machine learning models tend to automatically learn associations present in the training data without questioning their validity or appropriateness. This undesirable property is the root cause of the manifestation of spurious correlations, which render models unreliable and prone to failure in the presence of distribution shifts. Research shows that most methods attempting to remedy spurious correlations are only effective for a model's known spurious associations. Current spurious correlation detection algorithms either rely on extensive human annotations or are too restrictive in their formulation. Moreover, they rely on strict definitions of visual artifacts that may not apply to data produced by generative models, as they are known to hallucinate contents that do not conform to standard specifications. In this work, we introduce a general-purpose method that efficiently detects potential spurious correlations, and requires significantly less human interference in comparison to the prior art. Additionally, the proposed method provides intuitive explanations while eliminating the need for pixel-level annotations. We demonstrate the proposed method's tolerance to the peculiarity of AI-generated images, which is a considerably challenging task, one where most of the existing methods fall short. Consequently, our method is also suitable for detecting spurious correlations that may propagate to downstream applications originating from generative models.

Machine learning (ML) technologies are known to be riddled with ethical and operational problems, however, we are witnessing an increasing thrust by businesses to deploy them in sensitive applications. One major issue among many is that ML models do not perform equally well for underrepresented groups. This puts vulnerable populations in an even disadvantaged and unfavorable position. We propose an approach that leverages the power of web search and generative models to alleviate some of the shortcomings of discriminative models. We demonstrate our method on an image classification problem using ImageNet's People Subtree subset, and show that it is effective in enhancing robustness and mitigating bias in certain classes that represent vulnerable populations (e.g., female doctor of color). Our new method is able to (1) identify weak decision boundaries for such classes; (2) construct search queries for Google as well as text for generating images through DALL-E 2 and Stable Diffusion; and (3) show how these newly captured training samples could alleviate population bias issue. While still improving the model's overall performance considerably, we achieve a significant reduction (77.30\%) in the model's gender accuracy disparity. In addition to these improvements, we observed a notable enhancement in the classifier's decision boundary, as it is characterized by fewer weakspots and an increased separation between classes. Although we showcase our method on vulnerable populations in this study, the proposed technique is extendable to a wide range of problems and domains.