Related papers: Flint: Compiler Enabled Cluster-Free Design Space …

Training large language models (LLMs) requires substantial compute and energy. At the same time, renewable energy sources regularly produce more electricity than the grid can absorb, leading to curtailment, the deliberate reduction of clean…

Just-in-Time (JIT) compilers are used by many modern programming systems in order to improve performance. Bugs in JIT compilers provide exploitable security vulnerabilities and debugging them is difficult as they are large, complex, and…

Mechanistic interpretability seeks to understand how Large Language Models (LLMs) represent and process information. Recent approaches based on dictionary learning and transcoders enable representing model computation in terms of sparse,…

Aligning future system design with the ever-increasing compute needs of large language models (LLMs) is undoubtedly an important problem in today's world. Here, we propose a general performance modeling methodology and workload analysis of…

In today's production machine learning (ML) systems, models are continuously trained, improved, and deployed. ML design and training are becoming a continuous workflow of various tasks that have dynamic resource demands. Serverless…

The rapidly growing size of deep neural network (DNN) models and datasets has given rise to a variety of distribution strategies such as data, tensor-model, pipeline parallelism, and hybrid combinations thereof. Each of these strategies…

The interpretability of machine learning, particularly for deep neural networks, is crucial for decision making in real-world applications. One approach is replacing the un-interpretable machine learning model with a surrogate model, which…

Many recent machine learning models show dynamic shape characteristics. However, existing AI compiler optimization systems suffer a lot from problems brought by dynamic shape models, including compilation overhead, memory usage,…

Machine learning applications are increasingly deployed not only to serve predictions using static models, but also as tightly-integrated components of feedback loops involving dynamic, real-time decision making. These applications pose a…

We introduce xLLM, an intelligent and efficient Large Language Model (LLM) inference framework designed for high-performance, large-scale enterprise-grade serving, with deep optimizations for diverse AI accelerators. To address these…

Experimental designs reflect assumptions about variable relationships that determine what causal queries researchers can answer through the experiment. Accounting for and communicating these assumptions is essential for drawing valid,…

We study the problem of efficient, scalable set-sharing analysis of logic programs. We use the idea of representing sharing information as a pair of abstract substitutions, one of which is a worst-case sharing representation called a clique…

Multi-level intermediate representations (MLIR) show great promise for reducing the cost of building domain-specific compilers by providing a reusable and extensible compiler infrastructure. This work presents TPU-MLIR, an end-to-end…

Machine learning interatomic potentials (MLIPs) enable atomistic simulations with near ab initio accuracy at significantly reduced computational cost, but their broader adoption is often limited by fragmented tooling, limited scalability,…

With the rapid growth of large language models (LLMs), a wide range of methods have been developed to distribute computation and memory across hardware devices for efficient training and inference. While existing surveys provide descriptive…

We introduce Diffuse, a system that dynamically performs task and kernel fusion in distributed, task-based runtime systems. The key component of Diffuse is an intermediate representation of distributed computation that enables the necessary…

In the big data era, the key feature that each algorithm needs to have is the possibility of efficiently running in parallel in a distributed environment. The popular Silhouette metric to evaluate the quality of a clustering, unfortunately,…

There is increasing demand for specialized hardware for training deep neural networks, both in edge/IoT environments and in high-performance computing systems. The design space of such hardware is very large due to the wide range of…

We propose an end-to-end Multitask Learning Transformer framework, named MulT, to simultaneously learn multiple high-level vision tasks, including depth estimation, semantic segmentation, reshading, surface normal estimation, 2D keypoint…

The amount of available data about complex systems is increasing every year, measurements of larger and larger systems are collected and recorded. A natural representation of such data is given by networks, whose size is following the size…