相关论文: Data-Oriented Language Implementation of Lattice-B…

Numerical codes using the Lattice Boltzmann Methods (LBM) for simulating one- or two-phase flows are widely compiled and run on graphical process units. However, those computational units necessitate to re-write the program by using a…

Fine-tuning large language models is a popular choice among users trying to adapt them for specific applications. However, fine-tuning these models is a demanding task because the user has to examine several factors, such as resource…

Recently Transformers have provided state-of-the-art performance in sparse matching, crucial to realize high-performance 3D vision applications. Yet, these Transformers lack efficiency due to the quadratic computational complexity of their…

We consider the problem of accurate sparse fine-tuning of large language models (LLMs), that is, fine-tuning pretrained LLMs on specialized tasks, while inducing sparsity in their weights. On the accuracy side, we observe that standard…

A portable implementation of elaborated algorithm is important to use variety of architectures in HPC applications. In this work we implement and benchmark an algebraic multi-grid solver for Lattice QCD on three different architectures,…

Despite the high computational throughput of GPUs, limited memory capacity and bandwidth-limited CPU-GPU communication via PCIe links remain significant bottlenecks for accelerating large-scale data analytics workloads. This paper…

Learned Sparse Retrieval (LSR) has traditionally focused on small-scale encoder-only transformer architectures. With the advent of large-scale pre-trained language models, their capability to generate sparse representations for retrieval…

Large language models (LLMs) have shown exceptional performance and vast potential across diverse tasks. However, the deployment of LLMs with high performance in low-resource environments has garnered significant attention in the industry.…

FlashAttention (Dao, 2023) effectively reduces the quadratic peak memory usage to linear in training transformer-based large language models (LLMs) on a single GPU. In this paper, we introduce DISTFLASHATTN, a distributed memory-efficient…

Lattice Boltzmann methods (LBM) are an important part of current computational fluid dynamics (CFD). They allow easy implementations and boundary handling. However, competitive time to solution not only depends on the choice of a reasonable…

The Lattice Boltzmann Method algorithm is simplified by assuming constant numerical viscosity (the relaxation time is fixed at $\tau=1$). This leads to the removal of the distribution function from the computer memory. To test the solver…

We derive the partial differential equation (PDE) to which the pseudo-potential lattice Boltzmann method (P-LBM) converges under diffusive scaling, providing a rigorous basis for its consistency analysis. By establishing a direct link…

Block-wise diffusion language models (DLMs) generate multiple tokens in any order, offering a promising alternative to the autoregressive decoding pipeline. However, they still remain bottlenecked by memory-bound attention in long-context…

Self-attention has recently been adopted for a wide range of sequence modeling problems. Despite its effectiveness, self-attention suffers from quadratic compute and memory requirements with respect to sequence length. Successful approaches…

Sparse tensors appear in many large-scale applications with multidimensional and sparse data. While multidimensional sparse data often need to be processed on manycore processors, attempts to develop highly-optimized GPU-based…

Vector-based retrieval systems have become a common staple for academic and industrial search applications because they provide a simple and scalable way of extending the search to leverage contextual representations for documents and…

Optimizing GPU kernels with LLM agents is an iterative process over a large design space. Every candidate must be generated, compiled, validated, and profiled, so fewer trials will save both runtime and cost. We make two key observations.…

Dictionary training for sparse representations involves dealing with large chunks of data and complex algorithms that determine time consuming implementations. SBO is an iterative dictionary learning algorithm based on constructing unions…

Vision-Language Models (VLMs) excel across diverse tasks but suffer from high inference costs in time and memory. Token sparsity mitigates inefficiencies in token usage, while neuron sparsity reduces high-dimensional computations, both…

The evolution of large language models (LLMs) towards applications with ultra-long contexts faces challenges posed by the high computational and memory costs of the Transformer architecture. While existing sparse and linear attention…