14,842 papers · page 6 of 743
Compiling neural networks and Transformers for edge devices faces significant challenges due to resource constraints and the reliance on manually optimized operations for performance among others. These limitations hinder the scalability and portability of neural networks on reso…
A computational graph in a tensor compiler represents a program as a set of kernels connected by edges that describe how tensors flow between them. Tensors may be dense or sparse, the latter storing only nonzero values. Recognizing sparsity allows the compiler to avoid redundant …
Compiler optimizations may lead to loss of debug information, hampering developer productivity and techniques that rely on binary-to-source mappings, such as sampling-based feedback-directed optimization. While recent endeavors exposed debug information correctness and completene…
Qubit Mapping is a critical task in Quantum Compilation, as modern Quantum Processing Units (QPUs) are constrained to nearest-neighbor interactions defined by a qubit coupling graph. This compiler pass repairs the connectivity of two-qubit gates whose operands are not adjacent by…
Modern tensor compiler frameworks like JAX and PyTorch accelerate numerical programs by compiling or mapping Domain-Specific Language (DSL) code to efficient executables. However, they rely on a fixed set of transformation rules and heuristics, which means they can miss profitabl…
Modern AI accelerators adopt dataflow architectures to achieve both high peak throughput (TOPS) and energy efficiency (TOPS/W). These designs feature wide datapaths and hierarchical scratchpad memories that supply dense compute arrays with high-bandwidth data access and extensive…
AI Accelerators employ dataflow architectures to achieve impressive peak compute performance (TOPS) and processing efficiencies (TOPS/W). Typically, dataflow architectures use wide data-paths to connect off-chip memory to dense compute arrays (via hierarchy of on-chip memories/ve…
Modern GPUs feature specialized hardware units that enable high-performance, asynchronous dataflow execution. However, the conventional SIMT programming model is fundamentally misaligned with this task-parallel hardware, creating a significant programmability gap. While hardware-…
DataFrame libraries are widely adopted in data science for their flexible, Pythonic interfaces, but their fragmented APIs and unstructured query patterns limit systematic optimization. Existing work has explored parallel execution or SQL-style logical rewrites, yet these approach…
Compiling functional programs into efficient Field Programmable Gate Array (FPGA) designs is difficult. Hardware resources must be explicitly allocated and shared to maximize resource efficiency. This requires careful orchestration of several transformations to expose and exploit…
Instruction selection in compiler backends traditionally depends on huge handwritten rule libraries that map IR patterns to target instruction sequences. Porting to new architectures or extending them for new hardware features is a very error-prone process, results in a significa…
Large Language Models (LLMs) for compiler optimization have recently emerged as a frontier research direction, with many studies demonstrating their potential to automate and improve low-level code transformations. While various techniques have been proposed to enhance LLMs’ abil…
Fault injection attacks are a growing concern for the security of cyber systems. To mitigate these threats, hardware protections are often complemented by a series of software countermeasures, the implementation of which can be complex, time-consuming, and error-prone even for cy…
Java is often considered a superior programming language choice owing to its high portability, strong memory safety, and rapid development cycle. However, this superiority comes with increased complexity within Java software stacks, driven by the extensive use of layered librarie…
Partial evaluation allows specializing programs by substituting some of their inputs with static values and evaluating the resulting static expressions. When applied to a language interpreter, it specializes the interpreter to a fixed source program, resulting in a specialized in…
Sparsity is inherent in many applications such as machine learning and graph analytics. However, achieving high efficiency in sparse computations requires specialized hardware accelerators like FPGAs, as traditional accelerators typically cater to dense data. While high level syn…
With the increasing popularity of artificial intelligence (AI) applications, deep neural networks (DNNs) are in demand for on-device serving in various real-life fields. Running DNN inference on a resource-constrained edge device requires aggressive memory optimization. While sev…
Compiling files individually lends itself well to parallelization, but forces the compiler to operate on incomplete programs. State-of-the-art points-to analyses guarantee sound solutions only for complete programs, requiring summary functions to describe any missing program part…
Object flattening is a non-trivial optimization that inlines the fields of an object inside its containers. Owing to its direct applicability for immutable objects, Java would soon allow programmers to mark compatible classes as "value types", and Java Virtual Machines (JVMs) to …
Large Language Models (LLMs) have recently shown promise in compiler optimizations such as loop vectorization and memory access restructuring. However, due to their generative nature, LLM-optimized code may contain syntax errors or semantic inconsistencies. While state-of-the-art…