Related papers: LTL Model Checking of Self Modifying Code

This paper introduces an innovative approach that combines Large Language Models (LLMs) with Formal Verification strategies for automatic software vulnerability repair. Initially, we employ Bounded Model Checking (BMC) to identify…

Large Language Models (LLMs) have transformed code completion tasks, providing context-based suggestions to boost developer productivity in software engineering. As users often fine-tune these models for specific applications, poisoning and…

We address the model checking problem for shared memory concurrent programs modeled as multi-pushdown systems. We consider here boolean programs with a finite number of threads and recursive procedures. It is well-known that the model…

As Large Language Models (LLMs) become increasingly integrated into software development workflows, they also become prime targets for adversarial attacks. Among these, backdoor attacks are a significant threat, allowing attackers to…

Recent applications of LLMs in Machine Reading Comprehension (MRC) systems have shown impressive results, but the use of shortcuts, mechanisms triggered by features spuriously correlated to the true label, has emerged as a potential threat…

A promising research direction in enabling LLMs to generate consistently correct code involves addressing their inability to properly estimate program execution, particularly for code they generate. In this work, we demonstrate that Code…

Background: Fine-tuning is central to adapting pre-trained Large Language Models (LLMs) to downstream tasks, but its reliance on training data, parameter updates, and reusable components opens entry points for attackers. Threats have…

Despite various approaches being employed to detect vulnerabilities, the number of reported vulnerabilities shows an upward trend over the years. This suggests the problems are not caught before the code is released, which could be caused…

Pushdown systems (PDS) are known as an abstract model of recursive programs, and model checking methods for PDS have been studied. Register PDS (RPDS) are PDS augmented by registers to deal with data values from an infinite domain in a…

Self-modifying code has many intriguing applications in a broad range of fields including software security, artificial general intelligence, and open-ended evolution. Having control over self-modifying code, however, is still an open…

Large language models (LLMs) for code are increasingly used in software development, but they remain static after pretraining while APIs and software libraries continue to evolve. Model editing offers a lightweight alternative to retraining…

Black-box finetuning is an emerging interface for adapting state-of-the-art language models to user needs. However, such access may also let malicious actors undermine model safety. To demonstrate the challenge of defending finetuning…

The reuse of code fragments by copying and pasting is widely practiced in software development and results in code clones. Cloning is considered an anti-pattern as it negatively affects program correctness and increases maintenance efforts.…

With the increasingly rapid development of new malicious computer software by bad faith actors, both commercial and research-oriented antivirus detectors have come to make greater use of machine learning tactics to identify such malware as…

Large language models (LLMs) are now largely involved in software development workflows, and the code they generate routinely includes third-party library (TPL) imports annotated with specific version identifiers. These version choices can…

Model checking of open pushdown systems (OPD) w.r.t. standard branching temporal logics (pushdown module checking or PMC) has been recently investigated in the literature, both in the context of environments with perfect and imperfect…

Mutation testing is a well-established technique for assessing a test suite's quality by injecting artificial faults into production code. In recent years, mutation testing has been extended to machine learning (ML) systems, and deep…

Recent Language Models (LMs) achieve breakthrough performance in code generation when trained on human-authored problems, even solving some competitive-programming problems. Self-play has proven useful in games such as Go, and thus it is…

Large language models (LLMs) have demonstrated an impressive ability to generate codes on competitive programming tasks. However, with limited sample numbers, LLMs still suffer from poor accuracy. Inspired by the process of human…

Large Language Models (LLMs) can acquire deceptive behaviors through backdoor attacks, where the model executes prohibited actions whenever secret triggers appear in the input. Existing safety training methods largely fail to address this…