Author

Zachary Broveak

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Preprint Oct 2026

Spectral-MoE: Heterogeneous Per-Layer Expert Utilization in Dense-to-MoE Conversion

Large language models impose substantial memory demands and arithmetic costs during inference. Dense-to-mixture-of-experts (MoE) conversion is a model optimization technique that aims to reduce these arithmetic costs. Dense-to-MoE conversion partitions pretrained feed-forward network (FFN) neurons into experts and sele...

Zachary Broveak, Christian Kubicka, Thierry Tambe · 0 citations

#artificial intelligence Jun 2026

Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads

This work presents the first systems characterization of agent memory, introducing a system-oriented taxonomy classifying agent memory systems along four axes, and derives 10 system recommendations covering construction scheduling, capability floors, amortization via query volume, freshness-latency tradeoffs, and fleet...

Yasmine Omri, Ziyun Gan, Zachary Broveak et al. · arXiv.org · 7 citations