LLM Digest

Story

arxiv_cs_lg · Nov 1, 2026 · paper

Source brief

MatMoE: Matryoshka Mixture-of-Experts with Dynamic Mixed-Precision Quantization for Efficient Inference

arxiv.orgNov 1, 2026
original source linked

In brief

Mixture-of-Experts (MoE) architecture has been widely adopted in large language models (LLMs) since it reduces the computational cost through sparse expert activation. However, the massive parameter scale of MoE model...

Feed lens

evaluation

Read the original at arxiv.org →Open in live feed Read that day’s brief

MatMoE: Matryoshka Mixture-of-Experts with Dynamic Mixed-Precision Quantization for Efficient Inference

Earlier in this thread 4 items

Studying quantization trade-offs for efficient inference deployment in machine translation

DAIN: Dynamic Agent-Based Interaction Network for Efficient and Collaborative Multimodal Reasoning

RAMP: Robust Adaptive Mixed-Precision Quantization for Edge CPU Vision Models

EmbeddingGemma 2: Unified and Efficient Multimodal Representations