# SlimCaching: Edge Caching of Mixture-of-Experts for Distributed Inference

arXiv: https://arxiv.org/abs/2507.06567 · HTML: https://mlsanity.com/paper/2507.06567/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Efficient MoE training and inference systems, one of 12 directions in Mixture-of-experts.
- Builds on 34 of its 59 references that are in the map.
- First posted on arXiv 9 Jul 2025.

## Placed in

- Mixture-of-experts / [Efficient MoE training and inference systems](https://mlsanity.com/moe/efficient-moe-training-and-inference-systems/)

## Builds on (most cited first, 30 of 34)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [MoE-LLaVA: Mixture of Experts for Large Vision-Language Models](https://mlsanity.com/paper/2401.15947/) (2024)
- [A Joint Learning and Communications Framework for Federated Learning over Wireless Networks](https://mlsanity.com/paper/1909.07972/) (2019)
- [Hash Layers For Large Sparse Models](https://mlsanity.com/paper/2106.04426/) (2021)
- [A Survey on Mixture of Experts in Large Language Models](https://mlsanity.com/paper/2407.06204/) (2024)
- [Mobile Edge Intelligence for Large Language Models: A Contemporary Survey](https://mlsanity.com/paper/2407.18921/) (2024)
- [LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training](https://mlsanity.com/paper/2406.16554/) (2024)
- [On-Device Language Models: A Comprehensive Review](https://mlsanity.com/paper/2409.00088/) (2024)
- [A Review on Edge Large Language Models: Design, Execution, and Applications](https://mlsanity.com/paper/2410.11845/) (2024)
- [Pushing Large Language Models to the 6G Edge: Vision, Challenges, and Opportunities](https://mlsanity.com/paper/2309.16739/) (2023)
- [EdgeMoE: Empowering Sparse Large Language Models on Mobile Devices](https://mlsanity.com/paper/2308.14352/) (2023)
- [SplitLoRA: A Split Parameter-Efficient Fine-Tuning Framework for Large Language Models](https://mlsanity.com/paper/2407.00952/) (2024)
- [Edge Learning for B5G Networks with Distributed Signal Processing: Semantic Communication, Edge Computing, and Wireless Sensing](https://mlsanity.com/paper/2206.00422/) (2022)
- [Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models](https://mlsanity.com/paper/2501.12370/) (2025)
- [AdapMoE: Adaptive Sensitivity-based Expert Gating and Management for Efficient MoE Inference](https://mlsanity.com/paper/2408.10284/) (2024)
- [HSplitLoRA: A Heterogeneous Split Parameter-Efficient Fine-Tuning Framework for Large Language Models](https://mlsanity.com/paper/2505.02795/) (2025)
- [Exploiting Inter-Layer Expert Affinity for Accelerating Mixture-of-Experts Model Inference](https://mlsanity.com/paper/2401.08383/) (2024)
- [AgentsCoMerge: Large Language Model Empowered Collaborative Decision Making for Ramp Merging](https://mlsanity.com/paper/2408.03624/) (2024)
- [Read-ME: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design](https://mlsanity.com/paper/2410.19123/) (2024)
- [Prioritized Information Bottleneck Theoretic Framework with Distributed Online Learning for Edge Video Analytics](https://mlsanity.com/paper/2409.00146/) (2024)
- [TrimCaching: Parameter-sharing AI Model Caching in Wireless Edge Networks](https://mlsanity.com/paper/2405.03990/) (2024)
- [TrimCaching: Parameter-sharing Edge Caching for AI Model Downloading](https://mlsanity.com/paper/2404.14204/) (2024)
- [Semantic Sleuth: Identifying Ponzi Contracts via Large Language Models](https://mlsanity.com/paper/2411.07498/) (2024)

## Built on by (most cited first, 7 of 7)

- [SpaceMoE: Realizing Distributed Mixture-of-Experts Inference over Space Networks](https://mlsanity.com/paper/2605.00515/) (2026)
- [OrderMoE: An expert similarity driven distributed edge MoE inference](https://mlsanity.com/paper/2607.17154/) (2026)
- [OD-MoE: On-Demand Expert Loading for Cacheless Edge-Distributed MoE Inference](https://mlsanity.com/paper/2512.03927/) (2025)
- [CR^2: Cost-Aware Risk-Controlled Routing for Wireless Device-Edge LLM Inference](https://mlsanity.com/paper/2605.12001/) (2026)
- [TrimMoE A communication aware and adaptive depth framework for distributed edge inference](https://mlsanity.com/paper/2608.00573/) (2026)
- [BALANCE: Hybrid Autoregressive-Speculative LLM Inference at the Network Edge](https://mlsanity.com/paper/2608.05926/) (2026)
- [From Semantic to Token Communication: The Next Paradigm for Large-Model-Driven 6G Intelligent Connectivity](https://mlsanity.com/paper/2609.10714/) (2026)

## Abstract (arXiv)

Mixture-of-Experts (MoE) models improve the scalability of large language models (LLMs) by activating only a small subset of relevant experts per input. However, the sheer number of expert networks in an MoE model introduces a significant storage/memory burden for an edge device. To address this challenge, we consider a scenario where experts are dispersed across an edge network for distributed inference. Based on the popular Top-K expert selection strategy, we formulate a latency minimization problem by optimizing expert caching on edge servers under storage constraints. When K=1, the problem reduces to a monotone submodular maximization problem with knapsack constraints, for which we design a greedy-based algorithm with a (1 - 1/e)-approximation guarantee. For the general case where K≥1, expert co-activation within the same MoE layer introduces non-submodularity, which renders greedy methods ineffective. To tackle this issue, we propose a successive greedy decomposition method to decompose the original problem into a series of subproblems, with each being solved by a dynamic programming approach. Furthermore, we design an accelerated algorithm based on the max-convolution technique to obtain the approximate solution with a provable guarantee in polynomial time. Simulation results on various MoE models demonstrate that our method significantly reduces inference latency compared to existing baselines.