Related papers: A Flexible Instruction Set Architecture for Effici…

Scaling Mixture-of-Experts (MoE) training introduces systems challenges absent in dense models. Because each token activates only a subset of experts, this sparsity allows total parameters to grow much faster than per-token computation,…

Multimodal foundation models are transformative in sequential recommender systems, leveraging powerful representation learning capabilities. While Parameter-efficient Fine-tuning (PEFT) is commonly used to adapt foundation models for…

FPGAs have shown great potential in providing low-latency and energy-efficient solutions for deep neural network (DNN) inference applications. Currently, the majority of FPGA-based DNN accelerators in the cloud run in a time-division…

We here introduce a novel scheme for generating smoothly-varying infill graded microstructural (IGM) configurations from a given menu of generating cells. The scheme was originally proposed for essentially improving the variety of…

Shortcut models represent a promising, non-adversarial paradigm for generative modeling, uniquely supporting one-step, few-step, and multi-step sampling from a single trained network. However, their widespread adoption has been stymied by…

Mixture-of-Expert (MoE) models enable efficient inference by employing smaller experts and activating only a subset of them per token. MoE serving engines distribute experts across multiple GPUs and route tokens to appropriate GPUs at…

Large Language Models (LLMs) are increasingly used as autonomous agents for multi-step tasks. However, most existing frameworks fail to maintain a structured understanding of the task state, often relying on linear prompt concatenation or…

Self-supervised pretraining has transformed computer vision by enabling data-efficient fine-tuning, yet high-resolution training typically requires server-scale infrastructure, limiting in-domain foundation model development for many…

This paper investigates the design of parallel general matrix multiplication (GEMM) for a Versal Adaptive Compute Accelerated Platform (ACAP) equipped with a VC1902 system-on-chip and multiple Artificial Intelligence Engines (AIEs). Our…

Mixture-of-Experts (MoE) models have gained popularity in achieving state-of-the-art performance in a wide range of tasks in computer vision and natural language processing. They effectively expand the model capacity while incurring a…

Optical and optoelectronic approaches of performing matrix-vector multiplication (MVM) operations have shown the great promise of accelerating machine learning (ML) algorithms with unprecedented performance. The incorporation of…

Modern transformer-based Large Language Models (LLMs) are constructed with a series of decoder blocks. Each block comprises three key components: (1) QKV generation, (2) multi-head attention, and (3) feed-forward networks. In batched…

Generic matrix multiplication (GEMM) and one-dimensional convolution/cross-correlation (CONV) kernels often constitute the bulk of the compute- and memory-intensive processing within image/audio recognition and matching systems. We propose…

Large multimodal Mixture-of-Experts (MoEs) effectively scale the model size to boost performance while maintaining fixed active parameters. However, previous works primarily utilized full-precision experts during sparse up-cycling. Despite…

Deep learning recommendation models (DLRMs) are widely used in industry, and their memory capacity requirements reach the terabyte scale. Tiered memory architectures provide a cost-effective solution but introduce challenges in…

Specialized accelerators for tensor-operations, such as blocked-matrix operations and multi-dimensional convolutions, have been emerged as powerful architecture choices for high-performance Deep-Learning computing. The rapid development of…

With unprecedented demand for generative AI (GenAI) inference, acceleration of primitives that dominate GenAI such as general matrix-vector multiplication (GEMV) is receiving considerable attention. A challenge with GEMVs is the high memory…

As the need for edge computing grows, many modern consumer devices now contain edge machine learning (ML) accelerators that can compute a wide range of neural network (NN) models while still fitting within tight resource constraints. We…

Class-incremental learning (CIL) requires deep learning models to continuously acquire new knowledge from streaming data while preserving previously learned information. Recently, CIL based on pre-trained models (PTMs) has achieved…

We investigate the integration of stacked intelligent metasurfaces (SIMs) into cell-free massive multiple input multiple output (CF-mMIMO) system to enhance the simultaneous wireless information and power transfer (SWIPT) performance.…