paper-with-me

홈 › Papers

MoEUT: Mixture-of-Experts Universal Transformers

2024-05-25 · Róbert Csordás, Kazuki Irie, Jürgen Schmidhuber, Christopher Potts, Christopher D. Manning

Previous work on Universal Transformers (UTs) has demonstrated the importance of parameter sharing across layers. By allowing recurrence in depth, UTs have advantages over standard Transformers in learning compositional generalizations, but layer-sharing comes with a practical limitation of parameter-compute ratio: it drastically reduces the parameter count compared to the non-shared model with the same dimensionality. Naively scaling up the layer size to compensate for the loss of parameters makes its computational resource requirements prohibitive. In practice, no previous work has succeeded in proposing a shared-layer Transformer design that is competitive in parameter count-dominated tasks such as language modeling. Here we propose MoEUT (pronounced "moot"), an effective mixture-of-experts (MoE)-based shared-layer Transformer architecture, which combines several recent advances in MoEs for both feedforward and attention layers of standard Transformers together with novel layer-normalization and grouping schemes that are specific and crucial to UTs. The resulting UT model, for the first time, slightly outperforms standard Transformers on language modeling tasks such as BLiMP and PIQA, while using significantly less compute and memory.

📄 PDF Abstract BibTeX arXiv:2405.16039

Code (1)

robertcsordas/moeut 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingMixture-of-Experts

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
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…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Adam 설명 없음
Attention 설명 없음
Residual Connection 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
Multi-Head Attention 설명 없음

Similar Papers 제목 키워드 기반

Mixture of Decoupled Message Passing Experts with Entropy Constraint for General Node Classification

2025-02-12 · Xuanze Chen, Jiajun Zhou, Jinsong Chen, Shanqing Yu 외

The varying degrees of homophily and heterophily in real-world graphs persistently constrain the universality of graph neural networks (GNNs) for node classification. Adopting a data-centric perspective, this work reveal…

Mixture-of-ExpertsNode ClassificationRepresentation Learning

LPT++: Efficient Training on Mixture of Long-tailed Experts

2024-09-17 · Bowen Dong, Pan Zhou, WangMeng Zuo

We introduce LPT++, a comprehensive framework for long-tailed classification that combines parameter-efficient fine-tuning (PEFT) with a learnable model ensemble. LPT++ enhances frozen Vision Transformers (ViTs) through …

Mixture-of-Expertsparameter-efficient fine-tuning

Mixture of Universal Experts: Scaling Virtual Width via Depth-Width Transformation

2026-03-05 · Yilong Chen, Naibin Gu, Junyuan Shang, Zhenyu Zhang 외 arxiv

Mixture-of-Experts (MoE) decouples model capacity from per-token computation, yet their scalability remains limited by the physical dimensions of depth and width. To overcome this, we propose Mixture of Universal Experts…

DAMEX: Dataset-aware Mixture-of-Experts for visual understanding of mixture-of-datasets

2023-11-08 · NeurIPS 2023 11 · Yash Jain, Harkirat Behl, Zsolt Kira, Vibhav Vineet

Construction of a universal detector poses a crucial question: How can we most effectively train a model on a large mixture of datasets? The answer lies in learning dataset-specific features and ensembling their knowledg…

Mixture-of-Expertsobject-detectionObject Detection

Towards 3D Acceleration for low-power Mixture-of-Experts and Multi-Head Attention Spiking Transformers

2024-12-07 · Boxun Xu, Junyoung Hwang, Pruek Vanna-iampikul, Yuxuan Yin 외

Spiking Neural Networks(SNNs) provide a brain-inspired and event-driven mechanism that is believed to be critical to unlock energy-efficient deep learning. The mixture-of-experts approach mirrors the parallel distributed…

Mixture-of-Experts