4 accepted papers
Past work seeks to align large language model (LLM)-based assistants with a target set of values, but such assistants are frequently forced to make tradeoffs *between* values when deployed. In response to the scarcity of value conflict in existing alignment datasets, we introduce ConflictScope, an a…
To build fair AI systems we need to understand how social-group biases intrinsic to foundational encoder-based vision-language models (VLMs) manifest in biases in downstream tasks. In this study, we demonstrate that intrinsic biases in VLM representations systematically “carry over” or propagate int…
While recent work has found that vision-language models trained under the Contrastive Language Image Pre-training (CLIP) framework contain intrinsic social biases, the extent to which different upstream pre-training features of the framework relate to these biases, and hence how intrinsic bias and d…
Large Language Models (LLMs), such as the GPT series, have driven significant industrial applications, leading to economic and societal transformations. However, a comprehensive understanding of their real-world applications remains limited.To address this, we introduce **REALM**, a dataset of over…