Kavli Affiliate: Hsiao-Mei (Sherry) Cho| First 5 Authors: [#item_custom_name[1, [#item_custom_name[2, [#item_custom_name[3, [#item_custom_name[4, [#item_custom_name[5| Summary: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. This dichotomy has long motivated research into specialized hardware to mitigate the attention bottleneck. This […]
Continue.. Rethinking LLM Inference Bottlenecks: Insights from Latent Attention and Mixture-of-Experts