paper-with-me

Papers

MoIN: Mixture of Introvert Experts to Upcycle an LLM

2024-10-13 · Ajinkya Tejankar, KL Navaneet, Ujjawal Panchal, Kossar Pourahmadi, Hamed Pirsiavash

The goal of this paper is to improve (upcycle) an existing large language model without the prohibitive requirements of continued pre-training of the full-model. The idea is to split the pre-training data into semantically relevant groups and train an expert on each subset. An expert takes the form of a lightweight adapter added on the top of a frozen base model. During inference, an incoming query is first routed to the most relevant expert which is then loaded onto the base model for the forward pass. Unlike typical Mixture of Experts (MoE) models, the experts in our method do not work with other experts for a single query. Hence, we dub them "introvert" experts. Freezing the base model and keeping the experts as lightweight adapters allows extreme parallelism during training and inference. Training of all experts can be done in parallel without any communication channels between them. Similarly, the inference can also be heavily parallelized by distributing experts on different GPUs and routing each request to the GPU containing its relevant expert. We implement a proof-of-concept version of this method and show the validity of our approach.

📄 PDF Abstract BibTeX arXiv:2410.09687

Code (0)

등록된 구현이 없습니다.

Tasks

GPULanguage ModelingLanguage ModellingLarge Language ModelMixture-of-Experts

Methods 이 논문이 사용한 방법론

Adapter 설명 없음
BASE 설명 없음

Similar Papers 제목 키워드 기반

DeRS: Towards Extremely Efficient Upcycled Mixture-of-Experts Models

2025-03-03 · CVPR 2025 1 · Yongqi Huang, Peng Ye, Chenyu Huang, JianJian Cao 외

Upcycled Mixture-of-Experts (MoE) models have shown great potential in various tasks by converting the original Feed-Forward Network (FFN) layers in pre-trained dense models into MoE layers. However, these models still s…

Mixture-of-ExpertsQuantization

XFT: Unlocking the Power of Code Instruction Tuning by Simply Merging Upcycled Mixture-of-Experts

2024-04-23 · Yifeng Ding, Jiawei Liu, Yuxiang Wei, Terry Yue Zhuo 외

We introduce XFT, a simple yet powerful training scheme, by simply merging upcycled Mixture-of-Experts (MoE) to unleash the performance limit of instruction-tuned code Large Language Models (LLMs). While vanilla sparse u…

HumanEvalmbppMixture-of-Experts

CuMo: Scaling Multimodal LLM with Co-Upcycled Mixture-of-Experts

2024-05-09 · Jiachen Li, Xinyao Wang, Sijie Zhu, Chia-Wen Kuo 외

Recent advancements in Multimodal Large Language Models (LLMs) have focused primarily on scaling by increasing text-image pair data and enhancing LLMs to improve performance on multimodal tasks. However, these scaling ap…

Image CaptioningInstruction FollowingMixture-of-Expertsvisual instruction following+2

Upcycling Large Language Models into Mixture of Experts

2024-10-10 · Ethan He, Abhinav Khattar, Ryan Prenger, Vijay Korthikanti 외

Upcycling pre-trained dense language models into sparse mixture-of-experts (MoE) models is an efficient approach to increase the model capacity of already trained models. However, optimal techniques for upcycling at scal…

Mixture-of-ExpertsMMLU

Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints

2022-12-09 · Aran Komatsuzaki, Joan Puigcerver, James Lee-Thorp, Carlos Riquelme Ruiz 외

Training large, deep neural networks to convergence can be prohibitively expensive. As a result, often only a small selection of popular, dense models are reused across different contexts and tasks. Increasingly, sparsel…

Mixture-of-Experts