Related papers: M2L Translation Operators for Kernel Independent F…

As Large Language Models (LLMs) are increasingly adopted in edge intelligence to power domain-specific applications and personalized services, the quality and efficiency of the LLM post-training phase-including fine-tuning and inference,…

Multimodal Machine Translation (MMT) enhances translation quality by incorporating visual context, helping to resolve textual ambiguities. While existing MMT methods perform well in bilingual settings, extending them to multilingual…

Quantization significantly accelerates inference in large language models (LLMs) by replacing original high-precision matrices with low-precision counterparts. Recent advances in weight-activation quantization have primarily focused on…

Modern AI relies on huge matrix multiplications (MatMuls), whose computation poses a scalability problem for inference and training. We propose an alternative, GPU native bilinear operator to MatMuls in neural networks, which offers a…

Traditional monolithic kernels dominated kernel structures for long time along with small sized kernels,few hardware companies and limited kernel functionalities. Monolithic kernel structure was not applicable when the number of hardware…

Fast Multipole Methods (FMM) are a fundamental operation for the simulation of many physical problems. The high performance design of such methods usually requires to carefully tune the algorithm for both the targeted physics and the…

There has been significant progress in open-source text-only translation large language models (LLMs) with better language coverage and quality. However, these models can be only used in cascaded pipelines for speech translation (ST),…

Multi-user multiple-input, multiple-output (MU-MIMO) designs can substantially increase wireless systems' achievable throughput and connectivity capabilities. However, existing MU-MIMO deployments typically utilize linear processing…

Large Foundation Models (LFMs), including multi-modal and generative models, promise to unlock new capabilities for next-generation Edge AI applications. However, performing inference with LFMs in resource-constrained and heterogeneous edge…

Modern large-scale services such as search engines, messaging platforms, and serverless functions, rely on key-value (KV) stores to maintain high performance at scale. When such services are deployed in constrained memory environments, they…

This study explores the use of automatic BLAS offloading and INT8-based emulation for accelerating traditional HPC workloads on modern GPU architectures. Through the use of low-bitwidth integer units and cache-coherent Unified Memory…

Recently, a new framework to compute the photoionization rate in streamer discharges accurately and efficiently using the integral form and the fast multipole method (FMM) was presented. This paper further improves the efficiency of this…

We present hls4ml, a free and open-source platform that translates machine learning (ML) models from modern deep learning frameworks into high-level synthesis (HLS) code that can be integrated into full designs for field-programmable gate…

Boundary element methods for elliptic partial differential equations typically lead to boundary integral operators with translation-invariant kernel functions. Taking advantage of this property is fairly simple for particle methods, e.g.,…

This paper proposes a novel scheme for cell-free massive multiple-input multiple-output (CFmMIMO) networks to support any federated learning (FL) framework. This scheme allows each instead of all the iterations of the FL framework to happen…

Translation is important for cross-language communication, and many efforts have been made to improve its accuracy. However, less investment is conducted in aligning translations with human preferences, such as translation tones or styles.…

Multiple kernel learning (MKL) algorithms combine different base kernels to obtain a more efficient representation in the feature space. Focusing on discriminative tasks, MKL has been used successfully for feature selection and finding the…

In this paper, we propose CUDA-L2, a system that combines large language models (LLMs) and reinforcement learning (RL) to automatically optimize Half-precision General Matrix Multiply (HGEMM) CUDA kernels. Using CUDA execution speed as the…

Large Multimodal Models (LMMs) have demonstrated impressive capabilities in visual-language tasks but face significant deployment challenges due to their high computational demands. While recent token reduction methods show promise for…

Field-aware Factorization Machines (FFMs) have emerged as a powerful model for click-through rate prediction, particularly excelling in capturing complex feature interactions. In this work, we present an in-depth analysis of our in-house,…