A Dual-Path Architecture for Scaling Compute and Capacity in LLMs 事件

A Dual-Path Architecture for Scaling Compute and Capacity in LLMs arXiv:2605.30202v1 Announce Type: new Abstract: Looped transformers apply a shared block multiple times and have emerged as a parameter-efficient route to scaling compute in language models. However, at fixed FLOPs a looped model has strictly less capacity than a baseline transformer. We propose a novel dual-path block that can flexibly scale compute, the number of sequential operations applied to a hidden state, and capacity, th