相关论文: Mir: Automated Quantifiable Privilege Reduction Ag…

The secure domination problem, a variation of the domination problem with some important real-world applications, is considered. Very few algorithmic attempts to solve this problem have been presented in literature, and the most successful…

Software intensively depends on external libraries whose relevance may change during its life cycle. As a consequence, software developers must periodically reconsider the libraries they depend on, and must think about \textit{library…

Reinforcement Learning (RL) has shown promise for aligning Large Language Models (LLMs) to follow instructions with various constraints. Despite the encouraging results, RL improvement inevitably relies on sampling successful, high-quality…

Data movement between main memory and the CPU is a major bottleneck in parallel data-intensive applications. In response, researchers have proposed using compilers and intermediate representations (IRs) that apply optimizations such as loop…

This work considers the problem of privately outsourcing the computation of a matrix product over a finite field $\mathbb{F}_q$ to $N$ helper servers. These servers are considered to be honest but curious, i.e., they behave according to the…

Large Language Models have advanced automated software development, however, it remains a challenge to correctly infer dependencies, namely, identifying the internal components and external packages required for a repository to successfully…

Personalized alignments for individual users have been a long-standing goal in large language models (LLMs). We introduce Drift, a novel framework that personalizes LLMs at decoding time with implicit user preferences. Traditional…

The Model Context Protocol (MCP) has emerged as a widely adopted mechanism for connecting large language models to external tools and resources. While MCP promises seamless extensibility and rich integrations, it also introduces a…

We propose using reinforcement learning to address the challenges of discovering microarchitectural vulnerabilities, such as Spectre and Meltdown, which exploit subtle interactions in modern processors. Traditional methods like random…

Personalization in language models aims to tailor model behavior to individual users or user groups. Prompt-based methods incorporate user preferences into queries, while training-based methods encode them into model parameters. Model…

Today's LLMs are susceptible to prompt injections, jailbreaks, and other attacks that allow adversaries to overwrite a model's original instructions with their own malicious prompts. In this work, we argue that one of the primary…

The selection of third-party libraries is an essential element of virtually any software development project. However, deciding which libraries to choose is a challenging practical problem. Selecting the wrong library can severely impact a…

Third-party Python libraries introduce dependency management overhead, supply chain risk, and deployment friction in constrained environments. A natural question is how much of this ecosystem can be replicated using only Python's standard…

Modern datacenter applications are prone to high tail latencies since their requests typically follow highly-dispersive distributions. Delivering fast interrupts is essential to reducing tail latency. Prior work has proposed both OS- and…

In shared-memory concurrent programming, shared resources can be protected using synchronization mechanisms such as monitors or channels. The connection between these mechanisms and the resources they protect is, however, only given…

Small language models (SLMs) are increasingly valued for their efficiency and deployability in resource-constrained environments, making them useful for on-device, privacy-sensitive, and edge computing applications. On the other hand,…

Regular expressions (regexes) are foundational to modern computing for critical tasks like input validation and data parsing, yet their ubiquity exposes systems to regular expression denial of service (ReDoS), a vulnerability requiring…

Large language models (LLMs) are increasingly deployed in security-sensitive applications, where they must follow system- or developer-specified instructions that define the intended task behavior, while completing benign user requests.…

Dense process rewards have proven a more effective alternative to the sparse outcome-level rewards in the inference-time scaling of large language models (LLMs), particularly in tasks requiring complex multi-step reasoning. While dense…

The use of LLMs as automated judges ("LLM-as-a-judge") is now widespread, yet standard judges suffer from a multitude of reliability issues. To address these challenges, we introduce Verdict, an open-source library for scaling judge-time…