Vipera: Towards systematic auditing of generative text-to-image models at scale
By: Yanwei Huang , Wesley Hanwen Deng , Sijia Xiao and more
Potential Business Impact:
Helps AI make safer, fairer pictures.
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 thoroughness, and it is even more challenging to enable auditors to explore the auditing space in a structural and effective way. Vipera employs multiple visual cues including a scene graph to facilitate image collection sensemaking and inspire auditors to explore and hierarchically organize the auditing criteria. Additionally, it leverages LLM-powered suggestions to facilitate exploration of unexplored auditing directions. An observational user study demonstrates Vipera's effectiveness in helping auditors organize their analyses while engaging with diverse criteria.
Similar Papers
Vipera: Blending Visual and LLM-Driven Guidance for Systematic Auditing of Text-to-Image Generative AI
Human-Computer Interaction
Helps check AI art for bad or unfair pictures.
Revisiting Data Auditing in Large Vision-Language Models
CV and Pattern Recognition
Finds if AI saw your private pictures.
VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models
Cryptography and Security
Makes AI models with pictures and words unsafe.