6 papers
Boundary-targeted Membership Inference Attacks on Safety Classifiers
Anthony Hughes, Alexander Goldberg, Prince Jha +3
Safety classifiers are essential safeguards within generative AI systems, filtering harmful content or identifying at-risk users when interacting with large language models. Despit…
MM-SCALE: Grounded Multimodal Moral Reasoning via Scalar Judgment and Listwise Alignment
Eunkyu Park, Wesley Hanwen Deng, Cheyon Jin +8
Vision-Language Models (VLMs) continue to struggle to make morally salient judgments in multimodal and socially ambiguous contexts. Prior works typically rely on binary or pairwise…
Vipera: Blending Visual and LLM-Driven Guidance for Systematic Auditing of Text-to-Image Generative AI
Yanwei Huang, Wesley Hanwen Deng, Sijia Xiao +4
Despite their increasing capabilities, text-to-image generative AI systems are known to produce biased, offensive, and otherwise problematic outputs. While recent advancements have…
A Human-Centered Approach to Identifying Promises, Risks, & Challenges of Text-to-Image Generative AI in Radiology
Katelyn Morrison, Arpit Mathur, Aidan Bradshaw +7
As text-to-image generative models rapidly improve, AI researchers are making significant advances in developing domain-specific models capable of generating complex medical imager…
Vipera: Towards systematic auditing of generative text-to-image models at scale
Yanwei Huang, Wesley Hanwen Deng, Sijia Xiao +3
Generative text-to-image (T2I) models are known for their risks related such as bias, offense, and misinformation. Current AI auditing methods face challenges in scalability and th…
StructVizor: Interactive Profiling of Semi-Structured Textual Data
Yanwei Huang, Yan Miao, Di Weng +2
Data profiling plays a critical role in understanding the structure of complex datasets and supporting numerous downstream tasks, such as social media analytics and financial fraud…