Expert Divergence Learning for MoE-based Language Models

• URL: http://arxiv.org/abs/2603.00054v1
• Date: Tue, 10 Feb 2026 06:58:38 GMT
• Title: Expert Divergence Learning for MoE-based Language Models
• Abstract summary: We introduce Expert Divergence Learning, a novel pre-training strategy that explicitly encourages functional specialization among experts.<n>Our method incorporates a label-driven auxiliary loss that leverages domain labels inherent in pre-training corpora.<n>We validate our approach by pre-training MoE models of up to 15 billion parameters from scratch.
• Score: 21.857507958048696
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The Mixture-of-Experts (MoE) architecture is a powerful technique for scaling language models, yet it often suffers from expert homogenization, where experts learn redundant functionalities, thereby limiting MoE's full potential. To address this, we introduce Expert Divergence Learning, a novel pre-training strategy that explicitly encourages functional specialization among experts. Our method incorporates a label-driven auxiliary loss that leverages domain labels inherent in pre-training corpora to maximize the Jensen-Shannon Divergence between the expert routing distributions of different data domains. This optimization objective guides the model to develop diverged routing policies for varied domains and closer routing policies for the same domain, which leads to emergent and organized expert specialization. We validate our approach by pre-training MoE models of up to 15 billion parameters from scratch. Experimental results demonstrate that models trained with Expert Divergence Learning not only achieve a lower language modeling loss but also exhibit significant performance improvements across a diverse range of downstream benchmarks. Further analysis confirms that our method effectively mitigates expert homogenization and brings greater functional specialization, all with negligible computational overhead during training.

Related papers

• 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)
• Eigenvectors of Experts are Training-free Non-collapsing Routers [14.774596844618396]
  Sparse Mixture of Experts (SMoE) architectures improve the training efficiency of Large Language Models (LLMs) by routing input tokens to a subset of specialized experts.<n>Despite their remarkable success, both training and inference in SMoE models suffer from the expert collapse issue.<n>We propose Singular Value Decomposition SMoE (SSMoE), a novel and training-free framework that leverages spectral properties of the expert weights to address the collapse issue.
  arXiv  Detail & Related papers  (2026-05-29T08:27:10Z)
• Multi-Domain Learning with Global Expert Mapping [102.62297074508147]
  Mixture-of-Experts (MoE) models provide a scalable solution by routing inputs to specializedworks (experts)<n>We propose GEM (Global Expert Mapping), a planner-compiler framework that replaces the learned router with a global scheduler.<n>Our planner, based on linear programming relaxation, computes a fractional assignment of datasets to experts, while the compiler applies hierarchical rounding to convert this soft plan into a deterministic, capacity-aware mapping.<n>Experiments show that GEM-DINO achieves state-of-the-art performance on the UODB benchmark, with notable gains on underrepresented datasets and solves task interference in few
  arXiv  Detail & Related papers  (2026-04-20T21:09:34Z)
• MoE-GRPO: Optimizing Mixture-of-Experts via Reinforcement Learning in Vision-Language Models [36.67777831459829]
  Mixture-of-Experts (MoE) has emerged as an effective approach to reduce the computational overhead of Transformer architectures.<n>We propose MoE-GRPO, a reinforcement learning framework for optimizing expert routing in MoE-based Vision-Language Models.<n>We show that MoE-GRPO consistently outperforms standard top-K routing and its variants by promoting more diverse expert selection.
  arXiv  Detail & Related papers  (2026-03-26T03:23:45Z)
• ERNIE 5.0 Technical Report [244.36480708815316]
  ERNIE 5.0 is a unified autoregressive foundation model for unified multimodal understanding and generation across text, image, video, and audio.<n>To address practical challenges in large-scale deployment under diverse resource constraints, ERNIE 5.0 adopts a novel elastic training paradigm.<n>We show that ERNIE 5.0 achieves strong and balanced performance across multiple modalities.
  arXiv  Detail & Related papers  (2026-02-04T16:18:15Z)
• Training Matryoshka Mixture-of-Experts for Elastic Inference-Time Expert Utilization [60.309915093470416]
  Matryoshka MoE (M-MoE) is a training framework that instills a coarse-to-fine structure directly into the expert ensemble.<n>Our work paves the way for more practical and adaptable deployments of large-scale MoE models.
  arXiv  Detail & Related papers  (2025-09-30T16:56:44Z)
• SpectR: Dynamically Composing LM Experts with Spectral Routing [37.969478059005574]
  This paper introduces SPECTR, an approach for dynamically composing expert models at each time step during inference.<n>We show that SPECTR improves routing accuracy over alternative training-free methods, increasing task performance across expert domains.
  arXiv  Detail & Related papers  (2025-04-04T13:58:44Z)
• LFME: A Simple Framework for Learning from Multiple Experts in Domain Generalization [61.16890890570814]
  Domain generalization (DG) methods aim to maintain good performance in an unseen target domain by using training data from multiple source domains. This work introduces a simple yet effective framework, dubbed learning from multiple experts (LFME) that aims to make the target model an expert in all source domains to improve DG.
  arXiv  Detail & Related papers  (2024-10-22T13:44:10Z)
• Scalable Multi-Domain Adaptation of Language Models using Modular Experts [10.393155077703653]
  MoDE is a mixture-of-experts architecture that augments a general PLM with modular, domain-specialized experts. MoDE achieves comparable target performances to full parameter fine-tuning while achieving 1.65% better retention performance.
  arXiv  Detail & Related papers  (2024-10-14T06:02:56Z)
• 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)
• Learning to Route Among Specialized Experts for Zero-Shot Generalization [39.56470042680907]
  We propose Post-Hoc Adaptive Tokenwise Gating Over an Ocean of Specialized Experts (PHATGOOSE) It learns to route among specialized modules that were produced through parameter-efficient fine-tuning. It does not require simultaneous access to the datasets used to create the specialized models and only requires a modest amount of additional compute after each expert model is trained.
  arXiv  Detail & Related papers  (2024-02-08T17:43:22Z)

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.