VLM safety steering differs by layer depth, study finds
On Qwen3-VL-4B, early vision-layer steering moved unsafe rejection between 6% and 96% without retraining.
1 stories, newest first · one card per story
On Qwen3-VL-4B, early vision-layer steering moved unsafe rejection between 6% and 96% without retraining.
Today 1.2K readers · 1.8K views · 1.4K reads · all time 160K views