Related papers: Louvre: Lightweight Ordering Using Versioning for …

Today, network devices share buffer across priority queues to avoid drops during transient congestion. While cost-effective most of the time, this sharing can cause undesired interference among seemingly independent traffic. As a result,…

While modern software development heavily uses versioned packages, programming languages rarely support the concept of versions in their semantics, which makes software updates more bulky and unsafe. This paper proposes a programming…

Reducing computational costs is an important issue for development of embedded systems. Binary-weight Neural Networks (BNNs), in which weights are binarized and activations are quantized, are employed to reduce computational costs of…

Developing concurrent software is challenging, especially if it has to run on modern architectures with Weak Memory Models (WMMs) such as ARMv8, Power, or RISC-V. For the sake of performance, WMMs allow hardware and compilers to…

The rapid adoption of large language models (LLMs) has created significant challenges for efficient inference at scale. Unlike traditional workloads, LLM inference is constrained by both computation and the memory overhead of key-value (KV)…

In this work, we study progress conditions for commutativity-aware, linearizable implementations of shared objects. Motivated by the observation that commuting operations can be executed in parallel, we introduce…

In scientific computing and data science disciplines, it is often necessary to share application workflows and repeat results. Current tools containerize application workflows, and share the resulting container for repeating results. These…

Accurate document parsing requires both robust content recognition and a stable parser interface. In explicit Document Layout Analysis (DLA) pipelines, downstream parsers do not consume the full detector output. Instead, they operate on a…

This paper presents a lightweight formalism (a trace) to model message-passing concurrent executions where some common common problems can be identified, like lost or delayed messages, some forms of deadlock, etc. In particular, we consider…

Application compartmentalization and privilege separation are our primary weapons against ever-increasing security threats and privacy concerns on connected devices. Despite significant progress, it is still challenging to privilege…

With the wide adoption of language models for IR -- and specifically RAG systems -- the latency of the underlying LLM becomes a crucial bottleneck, since the long contexts of retrieved passages lead large prompts and therefore, compute…

When large AI models are deployed as cloud-based services, clients have no guarantee that responses are correct or were produced by the intended model. Rerunning inference locally is infeasible for large models, and existing cryptographic…

Convolutional Neural Networks (CNNs) are crucial in various applications, but their deployment on resource-constrained edge devices poses challenges. This study presents the Sum-of-Products (SOP) units for convolution, which utilize…

Language Models (LMs) struggle with complex, interdependent instructions, particularly in high-stakes domains like finance where precision is critical. We introduce FIFE, a novel, high-difficulty benchmark designed to assess LM…

Optimising deep learning inference across edge devices and optimisation targets such as inference time, memory footprint and power consumption is a key challenge due to the ubiquity of neural networks. Today, production deep learning…

On a system that exposes disjoint memory spaces to the software, a program has to address memory consistency issues and perform data transfers so that it always accesses valid data. Several approaches exist to ensure the consistency of the…

Reading and writing memory are, besides computation, the most common operations a processor performs. The correctness of these operations is therefore essential for the proper execution of any program. However, as soon as fault attacks are…

Oblivious RAM (ORAM) hides the memory access patterns, enhancing data privacy by preventing attackers from discovering sensitive information based on the sequence of memory accesses. The performance of ORAM is often limited by its inherent…

Large language models (LLMs) have demonstrated significant advancements in reasoning capabilities, performing well on various challenging benchmarks. Techniques like Chain-of-Thought prompting have been introduced to further improve…

Contemporary GPUs allow concurrent execution of small computational kernels in order to prevent idling of GPU resources. Despite the potential concurrency between independent kernels, the order in which kernels are issued to the GPU will…