Related papers: DWDP: Distributed Weight Data Parallelism for High…

How can we effectively handle queries for on-device large language models (LLMs) with varying runtime constraints, such as latency and accuracy? Multi-scale quantization addresses this challenge by enabling memory-efficient runtime model…

Differential Privacy (DP) is a widely adopted technique, valued for its effectiveness in protecting the privacy of task-specific datasets, making it a critical tool for large language models. However, its effectiveness in Multimodal Large…

We consider a three-level parallelisation scheme. The second and third levels define a classical two-level parallelisation scheme and some load balancing algorithm is used to distribute tasks among processes. It is well-known that for many…

Dynamic voltage and frequency scaling (DVFS) and task-to-core allocation are critical for thermal management and balancing energy and performance in embedded systems. Existing approaches either rely on utilization-based heuristics that…

Since the introduction of the GRPO algorithm, reinforcement learning (RL) has attracted increasing attention for LLM post-training, yet training efficiency remains a critical challenge. In mainstream RL frameworks, inference and training…

Massively multi-task learning with large language models has recently made substantial progress on few-shot generalization. However, this is usually performed in a centralized learning fashion, ignoring the privacy sensitivity issue of…

Retrieval-Augmented Generation is a technology that enhances large language models by integrating information retrieval. In the industry, inference services based on LLMs are highly sensitive to cost-performance ratio, prompting the need…

AI accelerator processing capabilities and memory constraints largely dictate the scale in which machine learning workloads (e.g., training and inference) can be executed within a desirable time frame. Training a state of the art,…

While Graph Neural Networks (GNNs) are popular in the deep learning community, they suffer from several challenges including over-smoothing, over-squashing, and gradient vanishing. Recently, a series of models have attempted to relieve…

Distributed synchronous stochastic gradient descent has been widely used to train deep neural networks (DNNs) on computer clusters. With the increase of computational power, network communications generally limit the system scalability.…

Large Language Models (LLMs) have revolutionized AI applications, but deploying them at scale presents significant challenges. We present RTP-LLM, a high-performance inference engine for industrial-scale LLM deployment, successfully…

Transformer models have emerged as the leading approach for achieving state-of-the-art performance across various application domains, serving as the foundation for advanced large-scale deep learning (DL) models. However, efficiently…

LLM-based applications have been widely used in various industries, but with the increasing of models size, an efficient large language model (LLM) inference system is an urgent problem to be solved for service providers. Since the…

As large language models (LLMs) continue to scale, their workloads increasingly rely on distributed execution across multiple GPUs. However, the conventional bulk synchronous parallel~(BSP) model used in such settings introduces significant…

Diffusion-based large language models (dLLMs) have recently gained significant attention for their exceptional performance and inherent potential for parallel decoding. Existing frameworks further enhance its inference efficiency by…

Large-scale models rely heavily on 3D parallelism for distributed training, which utilizes tensor parallelism (TP) as the intra-operator parallelism to partition model states across GPUs. However, TP introduces significant communication…

As LLMs scale to multi-million-token KV histories, real-time autoregressive decoding under tight Token-to-Token Latency (TTL) constraints faces growing pressure. Two core bottlenecks dominate: accessing Feed-Forward Network (FFN) weights…

Motivated by the growing demand for serving large language model inference requests, we study distributed load balancing for global serving systems with network latencies. We consider a fluid model in which continuous flows of requests…

Large language models (LLMs) demand significant memory and computation resources. Wafer-scale chips (WSCs) provide high computation power and die-to-die (D2D) bandwidth but face a unique trade-off between on-chip memory and compute…

Large language models (LLMs) have demonstrated remarkable performance and tremendous potential across a wide range of tasks. However, deploying these models has been challenging due to the astronomical amount of model parameters, which…