الباحثون

Zhanpeng Zhou

المنشورات 2

نسخة أولية وصول مفتوح

Looped Transformers as Optimizers

Yulong Huang, Chen Jiang, Zhanpeng Zhou وآخرون · 2026

Looped Transformers provide a parameter-efficient approach to depth scaling by repeatedly applying shared Transformer blocks. Recent reasoning models have likewise highlighted the value of scaling test-time computation through longer computation trajectories. However, the principles for designing effective loop transit …

المؤلفون المشاركون