Mixture of Attention Heads: Selecting Attention Heads Per Token
This addresses the need for more efficient and scalable attention mechanisms in large language models, representing an incremental advance by applying MoE principles to attention heads.
The paper tackles the problem of scaling Transformer models efficiently by proposing Mixture of Attention Heads (MoA), which dynamically selects attention heads per token using a router, achieving stronger performance than standard multi-head attention in tasks like Machine Translation and Masked Language Modeling.
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 architecture. This paper proposes the Mixture of Attention Heads (MoA), a new architecture that combines multi-head attention with the MoE mechanism. MoA includes a set of attention heads that each has its own set of parameters. Given an input, a router dynamically selects a subset of $k$ attention heads per token. This conditional computation schema allows MoA to achieve stronger performance than the standard multi-head attention layer. Furthermore, the sparsely gated MoA can easily scale up the number of attention heads and the number of parameters while preserving computational efficiency. In addition to the performance improvements, MoA also automatically differentiates heads' utilities, providing a new perspective to discuss the model's interpretability. We conducted experiments on several important tasks, including Machine Translation and Masked Language Modeling. Experiments have shown promising results on several tasks against strong baselines that involve large and very deep models.