DPA-MoE is a runtime precision control framework that allows Mixture-of-Experts (MoE) models to use any-precision routed experts. It encodes expert feed‑forward weights as prefix‑decodable chunks, profiles reconstruction residuals offline, and allocates chunks during inference based on live router gate scores. This enables a single encoded MoE checkpoint to operate at multiple effective weight budgets without retraining, improving KL divergence and downstream accuracy on Qwen3‑30B‑A3B and DeepSeekMoE‑16B with modest throughput overhead.
Mixture-of-Experts (MoE) models reduce computation by routing each token to a small set of experts, but each selected expert is still used at a fixed precision. We present DPA-MoE, a runtime precision...
Dynamic Expert Quantization (DynaExq) is a runtime-aware mixed-precision serving system designed for single‑GPU Mixture‑of‑Experts (MoE) inference under a hard high‑bandwidth memory (HBM) envelope. It treats the problem as an online, budget‑constrained precision allocation task, keeping the most frequently used experts at higher precision while relegating the rest to low‑precision fallbacks. By estimating expert hotness from router traces and asynchronously promoting or demoting experts, DynaExq maintains a fully materialized expert set during the forward pass, improving accuracy and throughput compared to static post‑training quantization and offloading/prefetch baselines. whyItMatters":"DynaExq enables efficient deployment of large MoE models on memory‑limited GPUs by dynamically allocating precision based on runtime expert usage, thereby reducing memory footprint and latency while boosting accuracy and throughput."
Large Mixture-of-Experts (MoE) language models are attractive for end-device deployment because only a small subset of experts is active per token, but their routed expert weights often exceed accelerator memory. We target latency-critical single-user settings where routed experts are staged on demand from CPU memory to a GPU or from Flash to a mobile NPU.
arXiv:2607. 24434v1 Announce Type: cross Abstract: Large Mixture-of-Experts (MoE) language models are attractive for end-device deployment because only a small subset of experts is active per token, but their routed expert weights often exceed accelerator memory.
ACE introduces a training‑free, calibration‑free framework for adaptive expert skipping in Mixture‑of‑Experts LLMs. It combines a Global Spectral Proxy that estimates global transformation capacity with a Router‑Conditioned Refinement that builds expert‑specific direction prototypes, enabling the model to skip low‑contribution experts while always keeping the top‑1 expert. Offline computation of expert statistics leaves only lightweight table lookups during inference, and experiments on three MoE‑based LLMs show ACE outperforms static and dynamic baselines, especially at high skipping ratios.
arXiv:2607. 24787v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (MoE) models expand foundation model capacity through conditional expert activation, but their full expert pools remain difficult to deploy under limited accelerator memory.
The paper introduces Ban&Pick, a post‑training, plug‑and‑play routing strategy for Sparse Mixture‑of‑Experts large language models. It identifies and reinforces a small group of highly influential experts while dynamically pruning redundant ones, leading to accuracy gains across math, code, and reasoning benchmarks. Experiments on DeepSeek and Qwen3 show notable performance improvements and a 1.25× inference speedup without retraining or architectural changes.
arXiv:2609.36222v1 Announce Type: new Abstract: Large language models are increasingly expensive to serve. In large-scale serving systems, autoregressive decoding is often bottlenecked by transferrin...
arXiv:2510. 19366v2 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) scales model capacity through sparse activation, and is becoming an important architecture for large language models (LLMs).
arXiv:2606. 15453v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) based large language models (LLMs), such as Qwen and DeepSeek, have recently emerged as an effective approach to improving model capacity without proportionally increasing computational cost.
The paper investigates why speculative decoding (SD) performs better than expected on sparse mixture-of-experts (MoE) models. By evaluating 10 large MoE models, the authors find that recent fine‑grained models achieve higher speedups at low batch sizes, contrary to prior predictions of a concave speedup‑batch relationship. They attribute this to lower verification costs driven by temporal overlap of draft tokens and GPU execution dynamics, and develop a maximum‑entropy IPF framework that accurately predicts unique expert counts from routing overlap, closely matching real‑model behavior.
Specter is an inference system for mixture-of-experts (MoE) models that uses quantization to enable speculative decoding and expert offloading. By running a lightweight quantized draft model on the device, Specter predicts future expert usage and pre‑organizes expert movement, reducing PCIe I/O bottlenecks. The approach yields an average 1.50× speedup (up to 1.88×) in time per output token compared to state‑of‑the‑art systems, making single‑GPU MoE inference more practical on commodity hardware.