Papers with mixture-of-experts

18 papers
Table-based Fact Verification with Self-adaptive Mixture of Experts (2022.findings-acl)

Copied to clipboard

Challenge: Existing research focuses on table-based fact verification, but a new trend is extending the scope to structured evidence.
Approach: They propose a mixture-of-experts neural network to recognize and execute different types of reasoning . they use a management module to decide the contribution of each expert network to the verification result .
Outcome: The proposed method achieves 85.1% accuracy on the TabFact dataset, comparable with the previous state-of-the-art models.
Parameter-Efficient Sparsity Crafting from Dense to Mixture-of-Experts for Instruction Tuning on General Tasks (2024.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) have demonstrated considerable proficiency in general natural language processing tasks.
Approach: They propose a parameter-efficient sparsity crafting method which crafts dense models into sparse models using the mixture-of-experts architecture.
Outcome: The proposed method significantly reduces computational costs and GPU memory requirements, while maintaining the quality of approximation in function space.
ConstitutionalExperts: Training a Mixture of Principle-based Prompts (2024.acl-short)

Copied to clipboard

Challenge: Large language models (LLMs) are capable at a variety of tasks given the right prompt, but writing one remains a difficult and tedious process.
Approach: They propose a method for learning a prompt consisting of constitutional principles, given a training dataset.
Outcome: The proposed method outperforms other prompt optimization techniques by 10.9% and improves all techniques, suggesting its broad applicability.
Divide, Conquer, and Combine: Mixture of Semantic-Independent Experts for Zero-Shot Dialogue State Tracking (2023.acl-long)

Copied to clipboard

Challenge: Existing methods to enhance the zeroshot generalization of DST fail to effectively decouple semantics of samples, limiting the zero-shot performance of the system.
Approach: They propose a new learning schema that explicitly disentangles the semantics of seen data and leverages the performance and robustness with the mixture-of-experts mechanism.
Outcome: The proposed model achieves state-of-the-art on multiWOZ2.1 with 10M trainable parameters and is robust to the mixture-of experts mechanism.
A Lightweight Mixture-of-Experts Neural Machine Translation Model with Stage-wise Training Strategy (2024.findings-naacl)

Copied to clipboard

Challenge: Using mixture-of-experts (MoE) to deal with language heterogeneity is a challenge in neural machine translation (NMT).
Approach: They propose a lightweight MoE-based NMT model that is trained via an elaborate stage-wise training strategy.
Outcome: The proposed model achieves stable improvements in translation tasks by introducing fewer extra parameters compared to baseline models.
Adaptive Gating in Mixture-of-Experts based Language Models (2023.emnlp-main)

Copied to clipboard

Challenge: Existing models employ a fixed gating network where each token is computed by the same number of experts.
Approach: They propose a flexible training strategy that allows tokens to be processed by a variable number of experts based on expert probability distribution.
Outcome: The proposed model reduces training time and inference quality while maintaining sparsity while maintaining inference accuracy.
H3Fusion: Helpful, Harmless, Honest Fusion of Aligned LLMs (2026.eacl-long)

Copied to clipboard

Challenge: Existing approaches to align pre-trained LLMs with instructions for one property are difficult to fine-tune.
Approach: They propose a mixture-of-experts-based fusion mechanism that models alignment as a controllable drift within the subspace, guided by a drift-regularization loss to balance competing alignment dimensions.
Outcome: Extensive evaluations of three benchmark datasets show that H3Fusion outperforms each individually aligned model by 11.37% and provides stronger robustness compared to the state-of-the-art LLM ensemble approaches by 13.77% and model-merging approaches by 6.18 %.
Neural Transduction for Multilingual Lexical Translation (2020.coling-main)

Copied to clipboard

Challenge: a method for completing multilingual translation dictionaries is proposed . a 27% relative improvement in whole-word accuracy is achieved when multilingual data is unavailable .
Approach: They propose a method for completing multilingual translation dictionaries using multilingual inputs and multilingual decoding objective.
Outcome: The proposed method can synthesize new word forms in multilingual translation dictionaries . it can perform in settings where correct translations have not been observed in text .
CoMoE: Contrastive Representation for Mixture-of-Experts in Parameter-Efficient Fine-tuning (2025.findings-emnlp)

Copied to clipboard

Challenge: Currently, mixture-of-experts (MoE) is underutilized on heterogeneous datasets, ignoring the fact that experts may learn similar knowledge.
Approach: They propose a method to promote modularization and specialization in MoE by specializing functionalities into different experts and sparsely activating them appropriately.
Outcome: The proposed method improves the capacity and specialization of mixture-of-experts (MoE) by sampling from activated and inactivated experts in top-k routing.
PEMT: Multi-Task Correlation Guided Mixture-of-Experts Enables Parameter-Efficient Transfer Learning (2024.findings-acl)

Copied to clipboard

Challenge: Parameter-efficient fine-tuning (PEFT) is an effective method for adapting pre-trained language models to various tasks efficiently.
Approach: They propose a parameter-efficient fine-tuning framework that captures transferable knowledge as a weighted combination of adapters trained on source tasks.
Outcome: The proposed method yields stable improvements over full fine-tuning and knowledge transferring methods on a broad range of tasks over 17 datasets.
Improving Grammatical Error Correction with Multimodal Feature Integration (2023.findings-acl)

Copied to clipboard

Challenge: Experimental results show that multimodal GEC models improve over strong baselines and achieve a new state-of-the-art result on the Falko-MERLIN test set.
Approach: They propose a framework that integrates both speech and text features to enhance GEC by generating audio from text using advanced text-to-speech models.
Outcome: The proposed framework improves on CoNLL14, BEA19 English, and Falko-MERLIN German datasets.
Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts (2024.findings-acl)

Copied to clipboard

Challenge: Neural architecture search (NAS) uses weight-sharing supernets to generate diverse subnetworks without retraining.
Approach: They propose a weight-sharing supernet that leverages mixture-of-experts to enhance supernet model expressiveness with minimal training overhead.
Outcome: The proposed method achieves state-of-the-art (SoTA) performance in NAS for fast machine translation models, surpassing NAS-BERT and AutoDistil across various model sizes.
Maximum Score Routing For Mixture-of-Experts (2025.findings-acl)

Copied to clipboard

Challenge: Traditional mixture-of-experts (MoE) networks impose an expert capacity constraint to ensure GPU-friendly computation.
Approach: They propose a routing paradigm that dynamically allocates input tokens to top-k experts through differentiable sparse transformations, enabling scalable model capacity while preserving computational efficiency.
Outcome: The proposed model achieves lower training losses and higher evaluation scores at equivalent FLOPs compared to constrained and unconstrained baselines.
Masks Can be Learned as an Alternative to Experts (2025.acl-long)

Copied to clipboard

Challenge: a recent study shows that sparse activation techniques can reduce inference performance without sacrificing performance.
Approach: They propose to sparsify a pre-trained dense large language model into a mixture-of-experts architecture for faster inference.
Outcome: The proposed approach is more efficient than one-shot sparsification techniques . it achieves 97% performance retention on downstream tasks with only 50% of parameters activated .
MiLoRA: Efficient Mixture of Low-Rank Adaptation for Large Language Models Fine-tuning (2024.findings-emnlp)

Copied to clipboard

Challenge: Low-rank adaptation and its mixture-of-experts (MOE) methods are highly effective but introduce significant latency in multi-tenant settings due to the LoRA modules and MOE routers added to multiple linear modules.
Approach: They propose a low-rank adaptation variant that considers each LoRA module as an expert and employs a prompt-aware routing mechanism.
Outcome: Extensive analysis on commonsense reasoning tasks and math reasoning tasks show that MiLoRA outperforms strong PEFT baselines with comparable tunable parameter budgets.
V-RoLoRA: RLVR-Driven MoE Routing for Steerable Pluralistic Alignment (2026.findings-acl)

Copied to clipboard

Challenge: Current methods for steering large language models rely on prompt engineering or reasoning-time guidance.
Approach: They propose a value-controllable pluralistic alignment framework enhanced with conditioned gating that dynamically directs the flow among multiple experts based on an input value or moral vector.
Outcome: The proposed method outperforms prompt-based steering and multi-task PEFT benchmarks on two 8-billion-parameter backbones.
Fundamental Reasoning Paradigms Induce Out-of-Domain Generalization in Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Deduction, induction, and abduction are fundamental reasoning paradigms, core for human logical thinking.
Approach: They propose to use a dataset of symbolic tasks to induce deductive skills into large language models (LLMs) they then use FT to fine-tune models to improve OOD generalization .
Outcome: The proposed approach yields strong generalizability with substantial performance gains (up to 14.60) across realistic out-of-domain tasks.
Knowledge Injection Exists in MoE? Exploring Expert-Aware Contrast Decoding in MoE for Mitigating LLMs’ Hallucinations (2026.acl-long)

Copied to clipboard

Challenge: Existing methods to mitigate hallucinations include prompt engineering and model optimization, but lack domain generalization and potential errors in fine-tuning data may exacerbate the hallucism.
Approach: They propose an expert-aware adaptive contrast decoding that uses expert differences in MoE’s higher layers to mitigate hallucinations on QA tasks.
Outcome: The proposed method outperforms baseline models on four datasets Large language models (LLMs) show strong performance but suffer from hallucinations, limiting their application.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations