paper-with-me

홈 › Papers

PMoE: Progressive Mixture of Experts with Asymmetric Transformer for Continual Learning

2024-07-31 · Min Jae Jung, Joohee Kim

Large Language Models (LLMs) encounter significant challenges in continual learning due to catastrophic forgetting, where new information overwrites previously acquired knowledge. This limitation leads to substantial environmental and economic waste. In this study, we introduce the PMoE, Progressive Mixture of Experts with Asymmetric Transformer, which aims to minimize forgetting by utilizing an asymmetric design with shallow layers dedicated to general knowledge and deep layers for new knowledge. PMoE incorporates progressively added experts in deep layers and a router that allocates new knowledge to the appropriate experts efficiently. The router, positioned adjacent to the deep layers, utilizes deep features aggregating consolidated information. This enables the router to perform efficiently, allocating new knowledge to the appropriate experts, which progressively increase in the deep layers. Extensive experiments on TRACE datasets and general language understanding datasets demonstrate that the proposed PMoE outperforms previous state-of-the-art approaches.

📄 PDF Abstract BibTeX arXiv:2407.21571

Code (0)

등록된 구현이 없습니다.

Tasks

Continual LearningGeneral KnowledgeMixture-of-Experts

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Multi-Head Attention 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
Adam 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

SwapMoE: Serving Off-the-shelf MoE-based Large Language Models with Tunable Memory Budget

2023-08-29 · Rui Kong, Yuanchun Li, Qingtian Feng, Weijun Wang 외

Mixture of experts (MoE) is a popular technique to improve capacity of Large Language Models (LLMs) with conditionally-activated parallel experts. However, serving MoE models on memory-constrained devices is challenging …

Mixture-of-Expertsobject-detectionObject DetectionText Summarization

Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks

2023-06-07 · Mohammed Nowaz Rabbani Chowdhury, Shuai Zhang, Meng Wang, Sijia Liu 외

In deep learning, mixture-of-experts (MoE) activates one or few experts (sub-networks) on a per-sample or per-token basis, resulting in significant computation reduction. The recently proposed \underline{p}atch-level rou…

Mixture-of-Experts

SPMoE: Generate Multiple Pattern-Aware Outputs with Sparse Pattern Mixture of Experts

2021-08-17 · Shaobo Cui, Xintong Bao, Xuming Lin, Zhongzhou Zhao 외

Many generation tasks follow a one-to-many mapping relationship: each input could be associated with multiple outputs. Existing methods like Conditional Variational AutoEncoder(CVAE) employ a latent variable to model thi…

DiversityMixture-of-ExpertsParaphrase Generation

AdapMoE: Adaptive Sensitivity-based Expert Gating and Management for Efficient MoE Inference

2024-08-19 · Shuzhang Zhong, Ling Liang, YuAn Wang, Runsheng Wang 외

Mixture-of-Experts (MoE) models are designed to enhance the efficiency of large language models (LLMs) without proportionally increasing the computational demands. However, their deployment on edge devices still faces si…

ManagementMixture-of-ExpertsSensitivity

MLPMoE: Zero-Shot Architectural Metamorphosis of Dense LLM MLPs into Static Mixture-of-Experts

2025-11-26 · Ivan Novikov arxiv

Large Language Models (LLMs) are predominantly deployed as dense transformers, where every parameter in every feed-forward block is activated for every token. While architecturally simple, this is computationally ineffic…