← Search

Ran Guo

2 accepted papers

2026

UltraMemV2: Memory Networks Scaling to 120B Parameters with Superior Long-Context Learning

ICLR 2026poster

While Mixture of Experts (MoE) models achieve remarkable efficiency by activating only subsets of parameters, they suffer from high memory access costs during inference. Memory-layer architectures offer an appealing alternative with very few memory access, but previous attempts like UltraMem have on…

Cited by 0SourcecodeScholar