Mixture of Weight-shared Heterogeneous Group Attention Experts for Dynamic Token-wise KV Optimization
By: Guanghui Song , Dongping Liao , Yiren Zhao and more
Potential Business Impact:
Makes AI remember more without using more computer power.
Transformer models face scalability challenges in causal language modeling (CLM) due to inefficient memory allocation for growing key-value (KV) caches, which strains compute and storage resources. Existing methods like Grouped Query Attention (GQA) and token-level KV optimization improve efficiency but rely on rigid resource allocation, often discarding "low-priority" tokens or statically grouping them, failing to address the dynamic spectrum of token importance. We propose mixSGA, a novel mixture-of-expert (MoE) approach that dynamically optimizes token-wise computation and memory allocation. Unlike prior approaches, mixSGA retains all tokens while adaptively routing them to specialized experts with varying KV group sizes, balancing granularity and efficiency. Our key novelties include: (1) a token-wise expert-choice routing mechanism guided by learned importance scores, enabling proportional resource allocation without token discard; (2) weight-sharing across grouped attention projections to minimize parameter overhead; and (3) an auxiliary loss to ensure one-hot routing decisions for training-inference consistency in CLMs. Extensive evaluations across Llama3, TinyLlama, OPT, and Gemma2 model families show mixSGA's superiority over static baselines. On instruction-following and continued pretraining tasks, mixSGA achieves higher ROUGE-L and lower perplexity under the same KV budgets.
Similar Papers
Mixture of Group Experts for Learning Invariant Representations
Machine Learning (CS)
Makes AI smarter by teaching experts to work together.
MoGA: Mixture-of-Groups Attention for End-to-End Long Video Generation
CV and Pattern Recognition
Makes computers create long videos much faster.
Mixture of Sparse Attention: Content-Based Learnable Sparse Attention via Expert-Choice Routing
Machine Learning (CS)
Makes AI smarter and faster by focusing on important words.