DiEP: Adaptive Mixture-of-Experts Compression through Differentiable Expert Pruning

• URL: http://arxiv.org/abs/2509.16105v1
• Date: Fri, 19 Sep 2025 15:47:42 GMT
• Title: DiEP: Adaptive Mixture-of-Experts Compression through Differentiable Expert Pruning
• Abstract summary: Existing MoE pruning methods often lead to suboptimal outcomes and performance degradation due to varying expert redundancy in different MoE layers.<n>We propose a non-uniform pruning strategy, dubbed textbfDiEP, which adaptively adjusts pruning rates at the layer level while jointly learning inter-layer importance.<n>Our method handles exponentially growing non-uniform expert combinations, enabling adaptive gradient-based pruning.
• Score: 24.59026258291765
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Despite the significant breakthrough of Mixture-of-Experts (MoE), the increasing scale of these MoE models presents huge memory and storage challenges. Existing MoE pruning methods, which involve reducing parameter size with a uniform sparsity across all layers, often lead to suboptimal outcomes and performance degradation due to varying expert redundancy in different MoE layers. To address this, we propose a non-uniform pruning strategy, dubbed \textbf{Di}fferentiable \textbf{E}xpert \textbf{P}runing (\textbf{DiEP}), which adaptively adjusts pruning rates at the layer level while jointly learning inter-layer importance, effectively capturing the varying redundancy across different MoE layers. By transforming the global discrete search space into a continuous one, our method handles exponentially growing non-uniform expert combinations, enabling adaptive gradient-based pruning. Extensive experiments on five advanced MoE models demonstrate the efficacy of our method across various NLP tasks. Notably, \textbf{DiEP} retains around 92\% of original performance on Mixtral 8$\times$7B with only half the experts, outperforming other pruning methods by up to 7.1\% on the challenging MMLU dataset.

Related papers

• Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data [119.14301046837822]
  We show that Mixture-of-Experts (MoEs) degrade more rapidly under data repetition.<n>MoEs can outperform dense models even when data is repeated more than 64 times.<n>We analyze internal mechanisms correlated with MoE overfitting in high repetition regimes.
  arXiv  Detail & Related papers  (2026-09-10T17:57:33Z)
• Tying the Loop -- Tied Expert Layers in Mixture-of-Experts Language Models [50.34585122347149]
  We introduce Expert Tying, an architectural modification that shares expert parameters across consecutive transformer layers.<n>We evaluate this approach across common, state-of-the-art architectures, including OLMoE, Qwen3, and DeepSeek-style MoEs.
  arXiv  Detail & Related papers  (2026-06-15T15:08:09Z)
• MoDES: Accelerating Mixture-of-Experts Multimodal Large Language Models via Dynamic Expert Skipping [52.02659589971978]
  We propose MoDES, the first training-free framework that adaptively skips experts to enable efficient and accurate MoE MLLM inference.<n>MoDES significantly enhances inference speed, improving the prefilling time by 2.16$times$ and the decoding time by 1.26$times$.
  arXiv  Detail & Related papers  (2025-11-19T18:48:27Z)
• MoETTA: Test-Time Adaptation Under Mixed Distribution Shifts with MoE-LayerNorm [13.63833424954647]
  Test-Time adaptation (TTA) has proven effective in mitigating performance drops under single-domain distribution shifts.<n>We propose MoETTA, a novel entropy-based TTA framework that integrates the Mixture-of-Experts architecture.
  arXiv  Detail & Related papers  (2025-11-14T10:24:06Z)
• MEPT: Mixture of Expert Prompt Tuning as a Manifold Mapper [75.6582687942241]
  We propose Mixture of Expert Prompt Tuning (MEPT) as an effective and efficient manifold-mapping framework.<n>MEPT integrates multiple prompt experts to adaptively learn diverse and non-stationary data distributions.<n> Empirical evaluations demonstrate that MEPT outperforms several state-of-the-art parameter efficient baselines on SuperGLUE.
  arXiv  Detail & Related papers  (2025-08-31T21:19:25Z)
• MoA: Heterogeneous Mixture of Adapters for Parameter-Efficient Fine-Tuning of Large Language Models [61.89384981175277]
  We propose a emphheterogeneous textbfMixture-of-Adapters (MoA) approach to integrate Low-Rank Adaptation (LoRA) and Mixture-of-Experts (MoE)<n> Experimental results demonstrate that heterogeneous MoA outperforms homogeneous MoE-LoRA methods in both performance and parameter efficiency.
  arXiv  Detail & Related papers  (2025-06-06T09:54:19Z)
• Ensemble-MIX: Enhancing Sample Efficiency in Multi-Agent RL Using Ensemble Methods [0.0]
  Multi-agent reinforcement learning (MARL) methods have achieved state-of-the-art results on a range of multi-agent tasks.<n>Yet, MARL algorithms require significantly more environment interactions than their single-agent counterparts to converge.<n>We propose a novel algorithm that combines a decomposed centralized critic with decentralized ensemble learning.
  arXiv  Detail & Related papers  (2025-06-03T13:13:15Z)
• Cluster-Driven Expert Pruning for Mixture-of-Experts Large Language Models [24.64757529640278]
  Cluster-driven Expert Pruning (C-Prune) is a novel two-stage framework for adaptive task-specific compression of large language models.<n>C-Prune operates through layer-wise expert clustering, which groups functionally similar experts within each MoE layer.<n>We validate C-Prune through extensive experiments on multiple MoE models and benchmarks.
  arXiv  Detail & Related papers  (2025-04-10T14:46:26Z)
• DiffMoE: Dynamic Token Selection for Scalable Diffusion Transformers [86.5541501589166]
  DiffMoE is a batch-level global token pool that enables experts to access global token distributions during training.<n>It achieves state-of-the-art performance among diffusion models on ImageNet benchmark.<n>The effectiveness of our approach extends beyond class-conditional generation to more challenging tasks such as text-to-image generation.
  arXiv  Detail & Related papers  (2025-03-18T17:57:07Z)
• A Sliding Layer Merging Method for Efficient Depth-Wise Pruning in LLMs [13.000188564679998]
  This paper reveals the Patch-like'' feature relationship between layers in large language models by analyzing the correlation of the outputs of different layers in the reproducing kernel Hilbert space.<n>We propose a sliding layer merging method that dynamically selects and fuses consecutive layers from top to bottom according to a pre-defined similarity threshold.<n>Our method outperforms existing pruning techniques in both zero-shot inference performance and retraining recovery quality after pruning.
  arXiv  Detail & Related papers  (2025-02-26T14:15:24Z)
• Determining Layer-wise Sparsity for Large Language Models Through a Theoretical Perspective [55.90119819642064]
  We address the challenge of determining the layer-wise sparsity rates of large language models (LLMs) through a theoretical perspective.<n>This refers to the cumulative effect of reconstruction errors throughout the sparsification process.<n>We derive a simple yet effective approach to layer-wise sparsity allocation that mitigates this issue.
  arXiv  Detail & Related papers  (2025-02-20T17:51:10Z)
• MoE-I$^2$: Compressing Mixture of Experts Models through Inter-Expert Pruning and Intra-Expert Low-Rank Decomposition [32.97035551579975]
  We introduce a two-stage compression method tailored for MoE to reduce the model size and decrease the computational cost. Experiments on Qwen1.5-MoE-A2.7B, DeepSeek-V2-Lite, and Mixtral-8$times$7B demonstrate that our proposed methods can both reduce the model size and enhance inference efficiency.
  arXiv  Detail & Related papers  (2024-11-01T20:37:58Z)
• Diversifying the Expert Knowledge for Task-Agnostic Pruning in Sparse Mixture-of-Experts [75.85448576746373]
  We propose a method of grouping and pruning similar experts to improve the model's parameter efficiency.<n>We validate the effectiveness of our method by pruning three state-of-the-art MoE architectures.<n>The evaluation shows that our method outperforms other model pruning methods on a range of natural language tasks.
  arXiv  Detail & Related papers  (2024-07-12T17:25:02Z)
• Diffusion for Natural Image Matting [88.58577778234036]
  We present DiffMatte, a solution designed to overcome the challenges of image matting.<n>First, DiffMatte decouples the decoder from the intricately coupled matting network design, involving only one lightweight decoder in the iterations of the diffusion process.<n>Second, we employ a self-aligned training strategy with uniform time intervals, ensuring a consistent noise sampling between training and inference across the entire time domain.
  arXiv  Detail & Related papers  (2023-12-10T15:28:56Z)
• ECoFLaP: Efficient Coarse-to-Fine Layer-Wise Pruning for Vision-Language Models [70.45441031021291]
  Large Vision-Language Models (LVLMs) can understand the world comprehensively by integrating rich information from different modalities. LVLMs are often problematic due to their massive computational/energy costs and carbon consumption. We propose Efficient Coarse-to-Fine LayerWise Pruning (ECoFLaP), a two-stage coarse-to-fine weight pruning approach for LVLMs.
  arXiv  Detail & Related papers  (2023-10-04T17:34:00Z)
• Harnessing Hard Mixed Samples with Decoupled Regularizer [69.98746081734441]
  Mixup is an efficient data augmentation approach that improves the generalization of neural networks by smoothing the decision boundary with mixed data. In this paper, we propose an efficient mixup objective function with a decoupled regularizer named Decoupled Mixup (DM) DM can adaptively utilize hard mixed samples to mine discriminative features without losing the original smoothness of mixup.
  arXiv  Detail & Related papers  (2022-03-21T07:12:18Z)

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.