Software vulnerabilities remain pervasive, even with the rise of AI-powered code assistants, advanced static analysis tools, and comprehensive testing frameworks. It’s clear that we must move beyond merely preventing these bugs; we need to eliminate them swiftly and efficiently. However, manual code intervention is slow, expensive, and can often introduce new security flaws, especially in legacy codebases. The advent of highly advanced Large Language Models (LLMs) presents a significant opportunity for automated software defect patching. We introduce LLM4CVE, an LLM-based iterative pipeline designed for robust and accurate repair of vulnerable functions in real-world code. We evaluate our pipeline using State-of-the-Art LLMs, including GPT-3.5, GPT-4o, Llama 3 8B, and Llama 3 70B. Our results demonstrate a human-verified quality score of 8.51/10 and a 20% increase in ground-truth code similarity with Llama 3 70B. To foster further research in LLM-based vulnerability repair, we release our evaluation framework, fine-tuned model weights, and experimental results on our website: https://sites.google.com/view/llm4cve

Industrial Control Systems (ICS) rely heavily on ProgrammableLogic Controllers (PLCs) to manage critical infrastructure, yet analyzing PLC executables remains challenging due to diverse proprietary compilers and limited access to source code.To bridge this gap, we introduce PLC-BEAD, a comprehensive dataset containing 2431 compiled binaries from 700+ PLC programs across four major industrial compilers (CoDeSys, GEB, OpenPLC-V2, OpenPLC-V3).This novel dataset uniquely pairs each binary with its original Structured Text source code and standardized functionality labels, enabling both binary-level and source-level analysis.We demonstrate the dataset's utility through PLCEmbed, a transformer-based framework for binary code analysis that achieves 93% accuracy in compiler provenance identification and 42% accuracy in finegrained functionality classification across 22 industrial control categories.Through comprehensive ablation studies, we analyze how compiler optimization levels, code patterns, and class distributions influence model performance.We provide detailed documentation of the dataset creation process, labeling taxonomy, and benchmark protocols to ensure reproducibility.Both PLC-BEAD and PLCEmbed are released as open-source resources to foster research in PLC security, reverse engineering, and ICS forensics, establishing new baselines for data-driven approaches to industrial cybersecurity.

RobotCat uses a goal-oriented, declarative approach for robot programming that leverages mathematical representations found in category theory as a way to formally model modularity and behavioral knowledge. We define hierarchical interfaces that require only local modifications when new software or work cell models are introduced, thus making robot programming more interoperable.Note to Practitioners—RoboCat can be put into practice and be beneficial to different stakeholders: 1) a student (or small, individual manufacturer) is in charge of planning the production of a new product. Instead of tedious programming, she sketches out process diagrams using conceptual templates for actions (e.g., “move,” “carry,” “locate”) that can produce vastly more resilient production plans than would normally be expected from someone without many years of experience; 2) a robot platform manufacturer has a new manipulator that they would like to provide to the existing customers. Usually, such an upgrade requires a complex retooling of the fabrication lines to accommodate the new functionality. Further complicating matters, the new manipulator contains third-party hardware and software, requiring a full re-architecting of the integrated system. Rather than a wholesale swap of the old system for the new, the third-party application programming interfaces (APIs) can be systematically translated to fit snuggly within the existing modular system with seamless interoperability; and 3) a company is designing the production process associated with their new product launch and they wish to spread their supply chain across several manufacturing vendors. Rather than writing individual process plans customized for each manufacturer, they compose a high-level plan in a functional language. This plan is compiled into implementations specific to each manufacturer’s systems, without the need for human intervention.

Automation engineering is the task of integrating, via software, various sensors, actuators, and controls to automate a real-world process. Today, automation engineering is supported by a suite of software tools, including integrated development environments (IDEs), hardware configurators, compilers, and runtimes. These tools focus on the automation code itself but leave the automation engineer unassisted in their decision-making. This can lead to longer software development cycles due to the imperfections in the decision-making, which arise when integrating software and hardware. To address this problem, this article addresses multiple challenges often faced in automation engineering and proposes machine learning-based solutions to assist engineers tackle these challenges. We show that machine learning can be leveraged to assist the automation engineer in classifying automation code, finding similar code snippets, and reasoning about the hardware selection of sensors and actuators. We validate our architecture on two real data sets consisting of 2927 Arduino projects and 683 programmable logic controller (PLC) projects. Our results show that paragraph embedding techniques can be utilized to classify automation using code snippets with precision close to human annotation, giving an$F_{1}$-score of 72%. Furthermore, we show that such embedding techniques can help us find similar code snippets with high accuracy. Finally, we use autoencoder models for hardware recommendation and achieve a$p\text{@}3$of 0.79 and$p\text{@}5$of 0.95. We also present the implementation of ArduCode in a proof-of-concept user interface integrated into an existing automation engineering system platform.Note to Practitioners—This article is motivated by the use of artificial intelligence methods to improve the efficiency and quality of the automation engineering software development process. Our goal is to develop and integrate intelligent assistants in existing automation engineering development tools to minimally disrupt existing workflows. Practitioners should be able to adapt our framework to other tools and data. Our contributions address important practical problems: 1) we address the lack of realistic data sets in automation engineering with two publicly available data sources; 2) we make the reference implementation of our algorithms publicly available on GitHub for other practitioners to have a starting point for future research; and 3) we demonstrate the integration of our framework as an add-on to an existing automation engineering toolchain.

This position article presents dispersed automation, a novel approach for reprogramming the omnipresent industrial Internet of Things (IIoT) devices in critical infrastructure for the co-execution of general-purpose computation workloads in an optimal and reliable manner. The key observation is that IIoT devices are available in large quantities in critical infrastructure. However, these devices are often underutilized due to the conservative allocation of resources, static task assignment, and lack of exploitation of parallelism. Dispersed automation gives new life to IIoT devices, some of them as powerful as general-purpose computers, and makes them accessible for the co-execution of various computational workloads, while dynamically adapting to different applications and operational environments. Two key contributions make this possible: the use of a domain-specific language and on-channel computation. We present dispersed automation as an enabling technology for the implementation of future smart and adaptive IIoT systems.

Cloud simulation studies are realized by the combination of different tools and services. Currently, these tools cannot be easily combined and designers do not have full control over the experiments. This paper introduces Cloud Simulation Mashups (CSM), a framework for creating mashups of simulation models, data, and algorithms that give the designer full control over the simulation studies. This research framework is currently being implemented in a domain-specific language (DSL) in Python and we have demonstrated its feasibility using two industrial-strength tools and the deployment in the cloud.