Related papers: MaxEVA: Maximizing the Efficiency of Matrix Multip…

The growing demand for on-device AI necessitates energy-efficient execution of DNN based applications on resource-constrained ultra-low power (ULP) platforms. Heterogeneous architectures, combining specialized processing elements (PEs),…

Integrated artificial intelligence (AI) and communication has been recognized as a key pillar of 6G and beyond networks. In line with AI-native 6G vision, explainability and robustness in AI-driven systems are critical for establishing…

Regular expression matching is the core function of various network security applications such as network intrusion detection systems. With the network bandwidth increases, it is a great challenge to implement regular expression matching…

Efficient image tokenization with high compression ratios remains a critical challenge for training generative models. We present SoftVQ-VAE, a continuous image tokenizer that leverages soft categorical posteriors to aggregate multiple…

Mixture of Experts (MoE) architectures have recently advanced the scalability and adaptability of large language models (LLMs) for continual multimodal learning. However, efficiently extending these models to accommodate sequential tasks…

The deployment of deep learning inference in production environments continues to grow, where throughput, latency, and hardware efficiency are critical. Although specialized accelerators are increasingly adopted, many inference workloads…

In recent years, DeepSeek has achieved strong inference performance but remains hard to deploy on energy-constrained edge devices. This paper presents the DeepSeek Processing Element (DSPE), an edge-oriented architecture that alleviates the…

We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent…

Deep neural networks trained through end-to-end learning have achieved remarkable success across various domains in the past decade. However, the end-to-end learning strategy, originally designed to minimize predictive loss in a black-box…

Graphics processing units (GPUs) excel at parallel processing, but remain largely unexplored in ultra-low-power edge devices (TinyAI) due to their power and area limitations, as well as the lack of suitable programming frameworks. To…

Mixture-of-Experts (MoE) activates only a subset of experts during inference, allowing the model to maintain low inference FLOPs and latency even as the parameter count scales up. However, since MoE dynamically selects the experts, all the…

Irregular embedding lookups are a critical bottleneck in recommender models, sparse large language models, and graph learning models. In this paper, we first demonstrate that, by offloading these lookups to specialized access units,…

Apple Neural Engine (ANE) is a dedicated neural processing unit (NPU) present in every Apple Silicon chip. Mixture-of-Experts (MoE) LLMs improve inference efficiency via sparse activation but are challenging for NPUs in three ways: expert…

Iterative solutions of sparse linear systems and sparse eigenvalue problems have a fundamental role in vital fields of scientific research and engineering. The crucial computing kernel for such iterative solutions is the multiplication of a…

Existing multi-criteria decision-making (MCDM) methods often face challenges when evaluating a large number of alternatives, leading to skewed results in selecting the optimal choice. Similarly, conventional efficiency analysis (EA)…

We present MegaScale-MoE, a production system tailored for the efficient training of large-scale mixture-of-experts (MoE) models. MoE emerges as a promising architecture to scale large language models (LLMs) to unprecedented sizes, thereby…

Recent advances in Latent Video Diffusion Models (LVDMs) have revolutionized video generation by leveraging Video Variational Autoencoders (Video VAEs) to compress intricate video data into a compact latent space. However, as LVDM training…

As Deep Learning continues to drive a variety of applications in edge and cloud data centers, there is a growing trend towards building large accelerators with several sub-accelerator cores/chiplets. This work looks at the problem of…

We present a cross-architecture evaluation of production LLM inference on AMD Instinct MI325X GPUs, benchmarking four models spanning 235B to 1 trillion parameters across three architectural families (MoE+MLA, Dense+GQA, MoE+GQA) on an…

With the rise of AI in recent years and the increase in complexity of the models, the growing demand in computational resources is starting to pose a significant challenge. The need for higher compute power is being met with increasingly…