Related papers: Target-Aware Implementation of Real Expressions
Modern computing systems increasingly rely on composing heterogeneous devices to improve performance and efficiency. Programming these systems is often unproductive: algorithm implementations must be coupled to system-specific logic,…
User simulation is a promising approach for automatically training and evaluating conversational information access agents, enabling the generation of synthetic dialogues and facilitating reproducible experiments at scale. However, the…
This paper addresses task planning problems for language-instructed robot teams. Tasks are expressed in natural language (NL), requiring the robots to apply their capabilities at various locations and semantic objects. Several recent works…
Natural language understanding (NLU) in the context of goal-oriented dialog systems typically includes intent classification and slot labeling tasks. Existing methods to expand an NLU system to new languages use machine translation with…
Modern scientific discovery increasingly relies on high-performance computing for complex modeling and simulation. A key challenge in improving parallel program performance is efficiently mapping tasks to processors and data to memory, a…
Machine-learning architectures, such as Convolutional Neural Networks (CNNs) are vulnerable to adversarial attacks: inputs crafted carefully to force the system output to a wrong label. Since machine-learning is being deployed in…
Deploying mixed-precision neural networks on edge devices is friendly to hardware resources and power consumption. To support fully mixed-precision neural network inference, it is necessary to design flexible hardware accelerators for…
Modern software systems are built to be used in dynamic environments using configuration capabilities to adapt to changes and external uncertainties. In a self-adaptation context, we are often interested in reasoning about the performance…
Deep learning compilers and vendor libraries deliver strong baseline performance but their performance is bounded by finite, engineer-curated catalogs. When these omit needed optimizations, practitioners substitute hand-written CUDA or…
Transformer-based language models for code have shown remarkable performance in various software analytics tasks, but their adoption is hindered by high computational costs, slow inference speeds, and substantial environmental impact. Model…
Finding bugs is key to the correctness of compilers in wide use today. If the behaviour of a compiled program, as allowed by its architecture memory model, is not a behaviour of the source program under its source model, then there is a…
Despite much success in natural language processing (NLP), pre-trained language models typically lead to a high computational cost during inference. Multi-exit is a mainstream approach to address this issue by making a trade-off between…
We propose a novel Transformer encoder-based architecture with syntactical knowledge encoded for intent detection and slot filling. Specifically, we encode syntactic knowledge into the Transformer encoder by jointly training it to predict…
The aliasing question (can two reference expressions point, during an execution, to the same object?) is both one of the most critical in practice, for applications ranging from compiler optimization to programmer verification, and one of…
Superoptimization is the task of transforming a program into a faster one while preserving its input-output behavior. In this work, we investigate whether large language models (LLMs) can serve as superoptimizers, generating assembly…
The next generation of cosmological surveys is expected to generate unprecedented high-quality data, consequently increasing the already substantial computational costs of Bayesian statistical methods. This will pose a significant challenge…
Optimizing software performance through automated code refinement offers a promising avenue for enhancing execution speed and efficiency. Despite recent advancements in LLMs, a significant gap remains in their ability to perform in-depth…
Currently, agent-based simulation frameworks force the user to choose between simulations involving a large number of agents (at the expense of limited agent reasoning capability) or simulations including agents with increased reasoning…
Computing-in-Memory (CIM) macros have gained popularity for deep learning acceleration due to their highly parallel computation and low power consumption. However, limited macro size and ADC precision introduce throughput and accuracy…
Complex software-driven systems often interleave distributed, concurrent computation processes with physical interactions with the environment. Developing these systems more efficiently and safely can be achieved by employing actionable,…