Fu, Z., Zeng, W., Wang, R., & Li, M. (2026). Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse.
Cita Chicago Style (17a ed.)Fu, Zizhuo, Wenxuan Zeng, Runsheng Wang, y Meng Li. Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse. 2026.
Cita MLA (9a ed.)Fu, Zizhuo, et al. Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse. 2026.
Precaución: Estas citas no son 100% exactas.