AI Tools & Models HOTFeaturedAll
arXiv cs.AIScore60

Rectified Looped Transformer

Summary

Looped Transformers deepen a model by reusing one shared block over and over; this study refines that trick to make the parameter-efficient language-modeling path more dependable.

Why this matters

Covers model · Relevant to this topic

Open original ↗

Source: arXiv cs.AI · arxiv.org

Topics

Related items