Related papers: MoH: Multi-Head Attention as Mixture-of-Head Atten…
Mixture-of-Experts (MoE) networks have been proposed as an efficient way to scale up model capacity and implement conditional computing. However, the study of MoE components mostly focused on the feedforward layer in Transformer…
Multi-head attentive neural architectures have achieved state-of-the-art results on a variety of natural language processing tasks. Evidence has shown that they are overparameterized; attention heads can be pruned without significant…
Multi-Head Mixture-of-Experts (MH-MoE) demonstrates superior performance by using the multi-head mechanism to collectively attend to information from various representation spaces within different experts. In this paper, we present a novel…
Computational workloads composing traditional transformer models are starkly bifurcated. Multi-Head Attention (MHA) and Grouped-Query Attention are memory-bound due to low arithmetic intensity, while FeedForward Networks are compute-bound.…
Multi-Head Attention (MHA) is a key component of Transformer. In MHA, attention heads work independently, causing problems such as low-rank bottleneck of attention score matrices and head redundancy. We propose Dynamically Composable…
Scaling depth is a key driver for large language models (LLMs). Yet, as LLMs become deeper, they often suffer from signal degradation: informative features formed in shallow layers are gradually diluted by repeated residual updates, making…
Despite many recent works on Mixture of Experts (MoEs) for resource-efficient Transformer language models, existing methods mostly focus on MoEs for feedforward layers. Previous attempts at extending MoE to the self-attention layer fail to…
Sliding-window attention offers a hardware-efficient solution to the memory and throughput challenges of Large Language Models (LLMs) in long-context scenarios. Existing methods typically employ a single window length across all attention…
We present MoE-MLA-RoPE, a novel architecture combination that combines Mixture of Experts (MoE) with Multi-head Latent Attention (MLA) and Rotary Position Embeddings (RoPE) for efficient language modeling. Our approach addresses the…
Recent advances in large language models highlighted the excessive quadratic cost of self-attention. Despite the significant research efforts, subquadratic attention methods still suffer from inferior performance in practice. We hypothesize…
Mixture-of-Experts (MoE) enhances model performance while maintaining computational efficiency, making it well-suited for large-scale applications. Conventional mixture-of-experts (MoE) architectures suffer from suboptimal coordination…
As the development of Large Language Models (LLMs) shifts from parameter scaling to inference-time collaboration, the Mixture-of-Agents (MoA) framework has emerged as a general paradigm to harness collective intelligence by layering diverse…
We present a novel approach called Mixture of Mixture of Expert (MoMoE) that combines the strengths of Mixture-of-Experts (MoE) architectures with collaborative multi-agent frameworks. By modifying the LLaMA 3.1 8B architecture to…
The vanilla self-attention mechanism in Transformers can be viewed as a two-layer fast-weight MLP, whose weights are dynamically induced by inputs and whose hidden dimension is equal to the sequence length $N$. As the context extends, the…
Scaling pre-trained language models has resulted in large performance gains in various natural language processing tasks but comes with a large cost in memory requirements. Inspired by the position embeddings in transformers, we aim to…
Vision Transformers have made remarkable progress in recent years, achieving state-of-the-art performance in most vision tasks. A key component of this success is due to the introduction of the Multi-Head Self-Attention (MHSA) module, which…
Transformer models based on the Mixture of Experts (MoE) architecture have made significant progress in long-sequence modeling, but existing models still have shortcomings in computational efficiency and the ability to capture long-range…
Mixture-of-Experts (MoE) models improve transformer efficiency but lack a unified theoretical explanation, especially when both feed-forward and attention layers are allowed to specialize. To this end, we study the Mixture-of-Transformers…
The choice of attention mechanism in Transformer models involves a critical trade-off between modeling quality and inference efficiency. Multi-Head Attention (MHA) offers the best quality but suffers from large Key-Value (KV) cache memory…
Multi-Head Attention (MHA) is the core computational primitive underlying modern Large Language Models (LLMs). However, MHA suffers from a fundamental linear scaling limitation: $H$ attention heads produce exactly $H$ independent attention…