Related papers: Progress and status of APEmille
We consider the problem of maximum a posteriori (MAP) inference in discrete graphical models. We present a parallel MAP inference algorithm called Bethe-ADMM based on two ideas: tree-decomposition of the graph and the alternating direction…
Edge deployment of low-batch large language models (LLMs) faces critical memory bandwidth bottlenecks when executing memory-intensive general matrix-vector multiplications (GEMV) operations. While digital processing-in-memory (PIM)…
Regulatory efforts to govern large language model (LLM) development have predominantly focused on restricting access to high-performance computational resources. This study evaluates the efficacy of such measures by examining whether LLM…
This paper considers a mobile edge computing-enabled cell-free massive MIMO wireless network. An optimization problem for the joint allocation of uplink powers and remote computational resources is formulated, aimed at minimizing the total…
The exponential growth of Internet of Things (IoT) applications has intensified the demand for efficient, high-throughput, and energy-efficient data processing at the edge. Conventional CPU-centric encryption methods suffer from performance…
Deploying large language models (LLMs) for online inference is often constrained by limited GPU memory, particularly due to the growing KV cache during auto-regressive decoding. Hybrid GPU-CPU execution has emerged as a promising solution…
LLM ensembles are widely used for LLM judges. However, how to estimate their accuracy, especially in an efficient way, is unknown. In this paper, we present a principled maximum a posteriori (MAP) framework for an economical and precise…
Edge computing's growing prominence, due to its ability to reduce communication latency and enable real-time processing, is promoting the rise of high-performance, heterogeneous System-on-Chip solutions. While current approaches often…
Speculative decoding can substantially accelerate LLM inference, but realizing its benefits in practice is challenging due to evolving workloads and system-level constraints. We present TIDE (Temporal Incremental Draft Engine), a…
With the development of large-scale integrated circuits, electronic design automation~(EDA) tools are increasingly emphasizing efficiency, with parallel algorithms becoming a trend. The optimization of delay reduction is a crucial factor…
We have developed and implemented a new technique of parallel information transmission through a multimode graded-index fiber in real time based on the approach of intensity moments. To do this, it was necessary to measure the spectrum of…
(Abridged) We have developed a numerical software library for collisionless N-body simulations named "Phantom-GRAPE" which highly accelerates force calculations among particles by use of a new SIMD instruction set extension to the x86…
Large language models (LLMs) are rapidly pushing the limits of contemporary computing hardware. For example, training GPT-3 has been estimated to consume around 1300 MWh of electricity, and projections suggest future models may require…
This paper introduces a parallel and asynchronous Transformer framework designed for efficient and accurate multilingual lip synchronization in real-time video conferencing systems. The proposed architecture integrates translation, speech…
Federated inference enhances LLM performance in edge computing through weighted averaging of distributed model predictions. However, autoregressive LLM inference requires frequent full-model forward passes across workers, severely limiting…
Current connectionist simulations require huge computational resources. We describe a neural network simulator for the IBM GF11, an experimental SIMD machine with 566 processors and a peak arithmetic performance of 11 Gigaflops. We present…
The longitudinal tracking engine of the particle accelerator simulation application PyHEADTAIL shows a heavy potential for parallelisation. For basic beam circulation, the tracking functionality with the leap-frog algorithm is extracted and…
Processing large-scale graph datasets is computationally intensive and time-consuming. Processor-centric CPU and GPU architectures, commonly used for graph applications, often face bottlenecks caused by extensive data movement between the…
In a recent funded Key Project of National Science Foundation of China, we have planned to do large scale simulations of lattice Quantum Chromodynamics, using the parallel supercomputing facilities in mainland China. Here I briefly review…
In this paper, we present FASE (Faster Asynchronous Systems Evaluation), a tool for evaluating the worst-case efficiency of asynchronous systems. The tool is based on some well-established results in the setting of a timed process algebra…