NeuronMoE: Neuron-Guided Mixture-of-Experts for Efficient Multilingual LLM Extension

• URL: http://arxiv.org/abs/2603.05046v1
• Date: Thu, 05 Mar 2026 10:55:26 GMT
• Title: NeuronMoE: Neuron-Guided Mixture-of-Experts for Efficient Multilingual LLM Extension
• Authors: Rongzhi Li, Hitomi Yanaka,
• Abstract summary: Mixture-of-Experts (MoE) architectures address this by adding sparse language-specific parameters.<n>Current approaches allocate experts based on layer-level similarity, yet language processing exhibits fine-grained specialization at individual neurons.<n>We propose $textbfNeuronMoE$, a method that analyzes language-specific neurons across all transformer components to guide expert allocation per layer.
• Score: 16.52781433504377
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Extending large language models to low-resource languages is essential for global accessibility, but training separate models per language is prohibitively expensive. Mixture-of-Experts (MoE) architectures address this by adding sparse language-specific parameters, but determining how many experts each layer needs remains an open question. Current approaches allocate experts based on layer-level similarity, yet language processing exhibits fine-grained specialization at individual neurons. We propose $\textbf{NeuronMoE}$, a method that analyzes language-specific neurons across all transformer components to guide expert allocation per layer based on empirically measured cross-lingual neuron diversity. Applied to Llama-3.2-3B for low-resource languages (Greek, Turkish, and Hungarian), this approach achieves approximately 40% average parameter reduction while matching the performance of the LayerMoE baseline. We find that low-resource language experts independently develop neuron specialization patterns mirroring the high-resource language, which are concentrated in early and late layers. This reveals potential universal architectural principles in how multilingual models organize linguistic knowledge.

Related papers

• Understanding Multilingualism in Mixture-of-Experts LLMs: Routing Mechanism, Expert Specialization, and Layerwise Steering [61.0787902713059]
  We propose a routing-guided steering method that adaptively guides routing behavior in middle layers toward shared experts associated with dominant languages at inference time.<n>Our code is available at http://conctsai.com/multilingualism-in-Mixture-of-Experts-LLMs.
  arXiv  Detail & Related papers  (2026-01-20T15:04:25Z)
• Sparse Subnetwork Enhancement for Underrepresented Languages in Large Language Models [11.719190735841407]
  Large language models exhibit uneven performance across languages.<n>We present a framework for enhancing monolingual capabilities of LLMs in underrepresented languages.<n>Our approach identifies language-specific neurons using Language Activation Probability Entropy and fine-tunes only the weights associated with these neurons.
  arXiv  Detail & Related papers  (2025-10-15T14:14:49Z)
• The Emergence of Abstract Thought in Large Language Models Beyond Any Language [95.50197866832772]
  Large language models (LLMs) function effectively across a diverse range of languages.<n>Preliminary studies observe that the hidden activations of LLMs often resemble English, even when responding to non-English prompts.<n>Recent results show strong multilingual performance, even surpassing English performance on specific tasks in other languages.
  arXiv  Detail & Related papers  (2025-06-11T16:00:54Z)
• How does Alignment Enhance LLMs' Multilingual Capabilities? A Language Neurons Perspective [64.79894853375478]
  We propose a new finer-grained neuron identification algorithm, which detects language neurons(including language-specific neurons and language-related neurons) and language-agnostic neurons.<n>Based on the distributional characteristics of different types of neurons, we divide the LLMs' internal process for multilingual inference into four parts.<n>We systematically analyze the models before and after alignment with a focus on different types of neurons.
  arXiv  Detail & Related papers  (2025-05-27T17:59:52Z)
• RetrieveAll: A Multilingual Named Entity Recognition Framework with Large Language Models [7.867158538366131]
  Existing multilingual NER methods face severe language interference during the multi-language adaptation process.<n>We propose RetrieveAll, a universal multilingual NER framework based on dynamic LoRA.<n>We introduce a cross-granularity knowledge augmented method that fully exploits the intrinsic potential of the data.
  arXiv  Detail & Related papers  (2025-05-25T12:52:18Z)
• Language-specific Neurons Do Not Facilitate Cross-Lingual Transfer [21.205821852762362]
  Existing techniques to identify language-specific neurons can be leveraged to enhance cross-lingual task performance of lowresource languages.<n>We find that such neuron-specific interventions are insufficient to yield cross-lingual improvements on downstream tasks.
  arXiv  Detail & Related papers  (2025-03-21T18:08:11Z)
• Efficiently Democratizing Medical LLMs for 50 Languages via a Mixture of Language Family Experts [29.091853631327304]
  Adapting medical Large Language Models to local languages can reduce barriers to accessing healthcare services.<n>We first construct a high-quality medical dataset and conduct analysis to ensure its quality.<n>We propose a novel MoE routing method that employs language-specific experts and cross-lingual routing.
  arXiv  Detail & Related papers  (2024-10-14T15:31:54Z)
• On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons [37.32174349956148]
  We analyze the neuron-level internal behavior of multilingual decoder-based language models (PLMs) We show that language-specific neurons are unique, with a slight overlap ( 5%) between languages. We tamper with less than 1% of the total neurons in each model during inference and demonstrate that tampering with a few language-specific neurons drastically changes the probability of target language occurrence in text generation.
  arXiv  Detail & Related papers  (2024-04-03T03:37:22Z)
• How do Large Language Models Handle Multilingualism? [81.15060972112563]
  This study explores how large language models (LLMs) handle multilingualism. LLMs initially understand the query, converting multilingual inputs into English for task-solving. In the intermediate layers, they employ English for thinking and incorporate multilingual knowledge with self-attention and feed-forward structures.
  arXiv  Detail & Related papers  (2024-02-29T02:55:26Z)
• Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models [117.20416338476856]
  Large language models (LLMs) demonstrate remarkable multilingual capabilities without being pre-trained on specially curated multilingual parallel corpora. We propose a novel detection method, language activation probability entropy (LAPE), to identify language-specific neurons within LLMs. Our findings indicate that LLMs' proficiency in processing a particular language is predominantly due to a small subset of neurons.
  arXiv  Detail & Related papers  (2024-02-26T09:36:05Z)
• High-resource Language-specific Training for Multilingual Neural Machine Translation [109.31892935605192]
  We propose the multilingual translation model with the high-resource language-specific training (HLT-MT) to alleviate the negative interference. Specifically, we first train the multilingual model only with the high-resource pairs and select the language-specific modules at the top of the decoder. HLT-MT is further trained on all available corpora to transfer knowledge from high-resource languages to low-resource languages.
  arXiv  Detail & Related papers  (2022-07-11T14:33:13Z)
• Cross-lingual, Character-Level Neural Morphological Tagging [57.0020906265213]
  We train character-level recurrent neural taggers to predict morphological taggings for high-resource languages and low-resource languages together.<n>Learning joint character representations among multiple related languages successfully enables knowledge transfer from the high-resource languages to the low-resource ones, improving accuracy by up to 30% over a monolingual model.
  arXiv  Detail & Related papers  (2017-08-30T08:14:34Z)

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.