This paper was accepted at the Principled Design for Trustworthy AI — Interpretability, Robustness, and Safety across Modalities Workshop at ICLR 2026.
What exactly makes a particular image unsafe? Systematically differentiating between benign and problematic images is a challenging problem, as subtle changes to an image, such as an insulting gesture or symbol, can drastically alter its safety implications. However, existing image safety datasets are coarse and ambiguous, offering only broad safety labels without isolating the specific features that drive these differences. We introduce… Read More
SafetyPairs: Isolating Safety Critical Image Features with Counterfactual Image Generation
This paper was accepted at the Principled Design for Trustworthy AI — Interpretability, Robustness, and Safety across Modalities Workshop at ICLR 2026.
What exactly makes a particular image unsafe? Systematically differentiating between benign and problematic images is a challenging problem, as subtle changes to an image, such as an insulting gesture or symbol, can drastically alter its safety implications. However, existing image safety datasets are coarse and ambiguous, offering only broad safety labels without isolating the specific features that drive these differences. We introduce…
Related Posts
PREDICT: Preference Reasoning by Evaluating Decomposed preferences Inferred from Candidate Trajectories
Accommodating human preferences is essential for creating AI agents that deliver personalized and effective interactions.…
AI in China and the United States
At a private dinner a few months ago, Jensen Huang apparently said what I’ve been…
Learning to Evict from Key-Value Cache
The growing size of Large Language Models (LLMs) makes efficient inference challenging, primarily due to…