SOURCE-LINKED INTELLIGENCE

SafeAtlas-VL: Beyond Binary Multimodal Safety with Large-Scale Data and Guard Models

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

Multimodal safety moderation requires distinguishing risks arising from visual content, user intent, and assistant behavior. Existing safeguards, however, are typically trained for a single judgment target and reduce safety assessment to a binary decision. Consequently, risk becomes difficult to compare across a multimodal interaction, and ambiguous cases are obscured. We introduce SafeAtlas-VL, a dataset of 1.5M training instances that places image-, request-, and response-level judgments on a five-level ordered scale. We curate a broad collection of safety-relevant data from both real-world

Read original source ↗ Open in workspace

recordType

paper

region

Global