A Safe Prototype Is Not a Safety Direction: Reference Dependence and Prompt Confounds in Response-Safety Embeddings
A positive safe-response centroid fails as a safety direction; an explicit safe-minus-unsafe reference scores far better.
1 stories, newest first · one card per story
A positive safe-response centroid fails as a safety direction; an explicit safe-minus-unsafe reference scores far better.
Today 1.1K readers · 1.3K views · 176 reads · all time 202K views