Yang, Z., Ding, W., Feng, S., & Tsvetkov, Y. (2026). Reading, Not Manipulating: Leveraging Router Logits for Multimodal Safety in MoE Vision-Language Models. https://omanscience.com/en/articles/reading-not-manipulating-leveraging-router-logits-for-multimodal-safety-in-moe-vision-language-models