Related papers: Event Tensor: A Unified Abstraction for Compiling …

We study how to support elasticity, i.e., the ability to dynamically adjust the parallelism (number of GPUs), for deep neural network (DNN) training. Elasticity can benefit multi-tenant GPU cluster management in many ways, e.g., achieving…

Heterogeneous deep learning systems (DLS) such as GPUs and ASICs have been widely deployed in industrial data centers, which requires to develop multiple low-level tensor programs for different platforms. An attractive solution to relieve…

By adequate employing of complex event processing (CEP), valuable information can be extracted from the underlying complex system and used in controlling and decision situations. An example application area is management of IT systems for…

Transformer models have advanced the state of the art in many Natural Language Processing (NLP) tasks. In this paper, we present a new Transformer architecture, Extended Transformer Construction (ETC), that addresses two key challenges of…

Graph neural networks (GNNs) emerge as a powerful approach to process non-euclidean data structures and have been proved powerful in various application domains such as social networks and e-commerce. While such graph data maintained in…

The efficient deployment of large language models (LLMs) is hindered by memory architecture heterogeneity, where traditional compilers suffer from fragmented workflows and high adaptation costs. We present nncase, an open-source, end-to-end…

According to the increasing complexity of network application and internet traffic, network processor as a subset of embedded processors have to process more computation intensive tasks. By scaling down the feature size and emersion of chip…

Large Language Model (LLM) inference uses an autoregressive manner to generate one token at a time, which exhibits notably lower operational intensity compared to earlier Machine Learning (ML) models such as encoder-only transformers and…

With the High Luminosity LHC coming online in the near future, event generators will need to provide very large event samples to match the experimental precision. Currently, the estimated cost to generate these events exceeds the computing…

Deep learning (DL) is one of the most prominent branches of machine learning. Due to the immense computational cost of DL workloads, industry and academia have developed DL libraries with highly-specialized kernels for each…

Whilst most engineered systems use signals that are continuous in time, there is a domain of systems in which signals consist of events. Events, like Dirac delta functions, have no meaningful time duration. Many important real-world systems…

Event extraction (EE) has considerably benefited from pre-trained language models (PLMs) by fine-tuning. However, existing pre-training methods have not involved modeling event characteristics, resulting in the developed EE models cannot…

The increasing complexity and scale of Deep Neural Networks (DNNs) necessitate specialized tensor accelerators, such as Tensor Processing Units (TPUs), to meet various computational and energy efficiency requirements. Nevertheless,…

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…

Text-to-motion generation has advanced with diffusion models, yet existing systems often collapse complex multi-action prompts into a single embedding, leading to omissions, reordering, or unnatural transitions. In this work, we shift…

Sparse tensors are rapidly becoming critical components of modern deep learning workloads. However, developing high-performance sparse operators can be difficult and tedious, and existing vendor libraries cannot satisfy the escalating…

On the way to Exascale, programmers face the increasing challenge of having to support multiple hardware architectures from the same code base. At the same time, portability of code and performance are increasingly difficult to achieve as…

Event-driven sensors, which produce data only when there is a change in the input signal, are increasingly used in applications that require low-latency and low-power real-time sensing, such as robotics and edge devices. To fully achieve…

The rapid scaling of Large Language Models (LLMs) has pushed training workloads far beyond the limits of single-node analysis, demanding a deeper understanding of how these models behave across large-scale, multi-GPU systems. In this paper,…

Training large language models (LLMs) in the cloud faces growing memory bottlenecks due to the limited capacity and high cost of GPUs. While GPU memory offloading to CPU and NVMe has made large-scale training more feasible, existing…