MoE-Prism: Disentangling Monolithic Experts for Elastic MoE Services via Model-System Co-Designs

• URL: http://arxiv.org/abs/2510.19366v1
• Date: Wed, 22 Oct 2025 08:40:01 GMT
• Title: MoE-Prism: Disentangling Monolithic Experts for Elastic MoE Services via Model-System Co-Designs
• Authors: Xinfeng Xia, Jiacheng Liu, Xiaofeng Hou, Peng Tang, Mingxuan Zhang, Wenfeng Wang, Chao Li,
• Abstract summary: MoE-Prism is a model-system co-design that transforms rigid MoE models into elastic services.<n>Our evaluation shows that MoE-Prismprovides over 4 times more distinct, stable operating points than the baseline.<n>This allows an AI service to dynamically improve throughput by up to 19.9% under a strict budget or reduce latency by up to 10.36% under limited resources.
• Score: 17.827406818899536
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mixture-of-Experts (MoE) models, the state-of-the-art in large-scale AI, achieve high quality by sparsely activating parameters. However, their reliance on routing between a few monolithic experts via a top-k mechanism creates a "quality cliff", offering only a few coarse-grained operating points. This inflexibility forces a difficult trade-off between cost and quality, preventing adaptation to diverse Service Level Objectives (SLOs) and leading to significant resource over-provisioning. This paper introduces MoE-Prism, a model-system co-design that transforms rigid MoE models into elastic services. Our methodology is divided into two phases. First, an \emph{Offline Refactoring Engine} systematically deconstructs monolithic experts into fine-grained "sub-experts." This engine employs a partitioning optimization solver that uses a metaheuristic-based approach to group neurons, preserving functional locality without requiring retraining. Second, an \emph{Online Scheduling Engine} leverages this new elasticity through QoS-aware scheduling. It implements specialized policies to solve complex system problems, including maximizing throughput in cloud deployments and managing latency-optimized offloading for memory-constrained devices. Our evaluation across three different MoE models shows that MoE-Prismprovides over 4 times more distinct, stable operating points than the baseline. This allows an AI service to dynamically improve throughput by up to 19.9\% under a strict latency budget or reduce latency by up to 10.36\% under limited resources. MoE-Prism provides the critical "control knob" to bridge the model-system gap, enabling the next generation of adaptive, efficient, and QoS-aware AI services.

Related papers

• The Curious Case of In-Training Compression of State Space Models [49.819321766705514]
  State Space Models (SSMs) tackle long sequence modeling tasks efficiently, offer both parallelizable training and fast inference.<n>Key design challenge is striking the right balance between maximizing expressivity and limiting this computational burden.<n>Our approach, textscCompreSSM, applies to Linear Time-Invariant SSMs such as Linear Recurrent Units, but is also extendable to selective models.
  arXiv  Detail & Related papers  (2025-10-03T09:02:33Z)
• Breaking the MoE LLM Trilemma: Dynamic Expert Clustering with Structured Compression [14.086434595924716]
  Mixture-of-Experts (MoE) Large Language Models (LLMs) face a trilemma of load imbalance, parameter redundancy, and communication overhead.<n>We introduce a unified framework based on dynamic expert clustering and structured compression to address these issues cohesively.
  arXiv  Detail & Related papers  (2025-09-27T10:45:58Z)
• CoMoE: Collaborative Optimization of Expert Aggregation and Offloading for MoE-based LLMs at Edge [28.961186929739082]
  Large language models (LLMs) have driven the adoption of Mixture-of-Experts (MoE) architectures as a promising solution to scale model capacity while controlling computational costs.<n> deploying MoE models in resource-constrained mobile edge computing environments presents significant challenges due to their large memory footprint and dynamic expert activation patterns.<n>We propose a novel dynamic resource-aware collaborative optimization framework that jointly optimize expert aggregation and offloading strategies.
  arXiv  Detail & Related papers  (2025-08-10T14:05:36Z)
• Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition [95.54406667705999]
  Pangu Embedded is an efficient Large Language Model (LLM) reasoner developed on Ascend Neural Processing Units (NPUs)<n>It addresses the significant computational costs and inference latency challenges prevalent in existing reasoning-optimized LLMs.<n>It delivers rapid responses and state-of-the-art reasoning quality within a single, unified model architecture.
  arXiv  Detail & Related papers  (2025-05-28T14:03:02Z)
• MAS-ZERO: Designing Multi-Agent Systems with Zero Supervision [76.42361936804313]
  We introduce MAS-ZERO, the first self-evolved, inference-time framework for automatic MAS design.<n> MAS-ZERO employs meta-level design to iteratively generate, evaluate, and refine MAS configurations tailored to each problem instance.
  arXiv  Detail & Related papers  (2025-05-21T00:56:09Z)
• D$^{2}$MoE: Dual Routing and Dynamic Scheduling for Efficient On-Device MoE-based LLM Serving [14.607254882119507]
  Combination of experts (MoE) model is a sparse variant of large language models (LLMs)<n>Despite its benefits, MoE is still too expensive to deploy on resource-constrained edge devices.<n>We propose D$2$MoE, an algorithm-system co-design framework that matches diverse task requirements by dynamically allocating the most proper bit-width to each expert.
  arXiv  Detail & Related papers  (2025-04-17T05:37:35Z)
• HOBBIT: A Mixed Precision Expert Offloading System for Fast MoE Inference [54.40808356999408]
  We present HOBBIT, a mixed precision expert offloading system to enable flexible and efficient MoE inference. Our key insight is that dynamically replacing less critical cache-miss experts with low precision versions can substantially reduce expert-loading latency. HOBBIT achieves up to a 9.93x speedup in decoding compared to state-of-the-art MoE offloading systems.
  arXiv  Detail & Related papers  (2024-11-03T04:25:46Z)
• Read-ME: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design [59.00758127310582]
  We propose a novel framework Read-ME that transforms pre-trained dense LLMs into smaller MoE models. Our approach employs activation sparsity to extract experts. Read-ME outperforms other popular open-source dense models of similar scales.
  arXiv  Detail & Related papers  (2024-10-24T19:48:51Z)
• AdapMoE: Adaptive Sensitivity-based Expert Gating and Management for Efficient MoE Inference [13.263938935671646]
  AdapMoE is an algorithm-system co-design framework for efficient MoE inference. AdapMoE features adaptive expert gating and management to reduce the on-demand loading overheads. We show AdapMoE consistently outperforms existing techniques, reducing the average number of activated experts by 25% and achieving a 1.35x speedup without degradation accuracy.
  arXiv  Detail & Related papers  (2024-08-19T03:27:15Z)
• Temporal Feature Matters: A Framework for Diffusion Model Quantization [105.3033493564844]
  Diffusion models rely on the time-step for the multi-round denoising.<n>We introduce a novel quantization framework that includes three strategies.<n>This framework preserves most of the temporal information and ensures high-quality end-to-end generation.
  arXiv  Detail & Related papers  (2024-07-28T17:46:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.