The future of software engineering--SE 3.0--is unfolding with the rise of AI teammates: autonomous, goal-driven systems collaborating with human developers. Among these, autonomous coding agents are especially transformative, now actively initiating, reviewing, and evolving code at scale. This paper introduces AIDev, the first large-scale dataset capturing how such agents operate in the wild. Spanning over 456,000 pull requests by five leading agents--OpenAI Codex, Devin, GitHub Copilot, Cursor, and Claude Code--across 61,000 repositories and 47,000 developers, AIDev provides an unprecedented empirical foundation for studying autonomous teammates in software development. Unlike prior work that has largely theorized the rise of AI-native software engineering, AIDev offers structured, open data to support research in benchmarking, agent readiness, optimization, collaboration modeling, and AI governance. The dataset includes rich metadata on PRs, authorship, review timelines, code changes, and integration outcomes--enabling exploration beyond synthetic benchmarks like SWE-bench. For instance, although agents often outperform humans in speed, their PRs are accepted less frequently, revealing a trust and utility gap. Furthermore, while agents accelerate code submission--one developer submitted as many PRs in three days as they had in three years--these are structurally simpler (via code complexity metrics). We envision AIDev as a living resource: extensible, analyzable, and ready for the SE and AI communities. Grounding SE 3.0 in real-world evidence, AIDev enables a new generation of research into AI-native workflows and supports building the next wave of symbiotic human-AI collaboration. The dataset is publicly available at https://github.com/SAILResearch/AI_Teammates_in_SE3. > AI Agent, Agentic AI, Coding Agent, Agentic Coding, Software Engineering Agent
Armin Ariamajd, Raquel López-Ríos de Castro, Andrea Volkamer
The increasing importance of Computational Science and Engineering has highlighted the need for high-quality scientific software. However, research software development is often hindered by limited funding, time, staffing, and technical resources. To address these challenges, we introduce PyPackIT, a cloud-based automation tool designed to streamline research software engineering in accordance with FAIR (Findable, Accessible, Interoperable, and Reusable) and Open Science principles. PyPackIT is a user-friendly, ready-to-use software that enables scientists to focus on the scientific aspects of their projects while automating repetitive tasks and enforcing best practices throughout the software development life cycle. Using modern Continuous software engineering and DevOps methodologies, PyPackIT offers a robust project infrastructure including a build-ready Python package skeleton, a fully operational documentation and test suite, and a control center for dynamic project management and customization. PyPackIT integrates seamlessly with GitHub's version control system, issue tracker, and pull-based model to establish a fully-automated software development workflow. Exploiting GitHub Actions, PyPackIT provides a cloud-native Agile development environment using containerization, Configuration-as-Code, and Continuous Integration, Deployment, Testing, Refactoring, and Maintenance pipelines. PyPackIT is an open-source software suite that seamlessly integrates with both new and existing projects via a public GitHub repository template at https://github.com/repodynamics/pypackit.
Leonhard Applis, Yuntong Zhang, Shanchao Liang
et al.
The growth of Large Language Model (LLM) technology has raised expectations for automated coding. However, software engineering is more than coding and is concerned with activities including maintenance and evolution of a project. In this context, the concept of LLM agents has gained traction, which utilize LLMs as reasoning engines to invoke external tools autonomously. But is an LLM agent the same as an AI software engineer? In this paper, we seek to understand this question by developing a Unified Software Engineering agent or USEagent. Unlike existing work which builds specialized agents for specific software tasks such as testing, debugging, and repair, our goal is to build a unified agent which can orchestrate and handle multiple capabilities. This gives the agent the promise of handling complex scenarios in software development such as fixing an incomplete patch, adding new features, or taking over code written by others. We envision USEagent as the first draft of a future AI Software Engineer which can be a team member in future software development teams involving both AI and humans. To evaluate the efficacy of USEagent, we build a Unified Software Engineering bench (USEbench) comprising of myriad tasks such as coding, testing, and patching. USEbench is a judicious mixture of tasks from existing benchmarks such as SWE-bench, SWT-bench, and REPOCOD. In an evaluation on USEbench consisting of 1,271 repository-level software engineering tasks, USEagent shows improved efficacy compared to existing general agents such as OpenHands CodeActAgent. There exist gaps in the capabilities of USEagent for certain coding tasks, which provides hints on further developing the AI Software Engineer of the future.
Lingyun Chen, Abdeldjallil Naceri, Abdalla Swikir
et al.
A drawing robot avatar is a robotic system that allows for telepresence-based drawing, enabling users to remotely control a robotic arm and create drawings in real-time from a remote location. The proposed control framework aims to improve bimanual robot telepresence quality by reducing the user workload and required prior knowledge through the automation of secondary or auxiliary tasks. The introduced novel method calculates the near-optimal Cartesian end-effector pose in terms of visual feedback quality for the attached eye-to-hand camera with motion constraints in consideration. The effectiveness is demonstrated by conducting user studies of drawing reference shapes using the implemented robot avatar compared to stationary and teleoperated camera pose conditions. Our results demonstrate that the proposed control framework offers improved visual feedback quality and drawing performance.
In recent years, the field of generative artificial intelligence, particularly in the domain of image generation, has exerted a profound influence on society. Despite the capability of AI to produce images of high quality, the augmentation of users' drawing abilities through the provision of drawing support systems emerges as a challenging issue. In this study, we propose that a cognitive factor, specifically, the size of the canvas, may exert a considerable influence on the outcomes of imitative drawing sketches when utilizing reference images. To investigate this hypothesis, a web based drawing interface was utilized, designed specifically to evaluate the effect of the canvas size's proportionality to the reference image on the fidelity of the drawings produced. The findings from our research lend credence to the hypothesis that a drawing interface, featuring a canvas whose dimensions closely match those of the reference image, markedly improves the precision of user-generated sketches.
There is abundant observational data in the software engineering domain, whereas running large-scale controlled experiments is often practically impossible. Thus, most empirical studies can only report statistical correlations -- instead of potentially more insightful and robust causal relations. To support analyzing purely observational data for causal relations, and to assess any differences between purely predictive and causal models of the same data, this paper discusses some novel techniques based on structural causal models (such as directed acyclic graphs of causal Bayesian networks). Using these techniques, one can rigorously express, and partially validate, causal hypotheses; and then use the causal information to guide the construction of a statistical model that captures genuine causal relations -- such that correlation does imply causation. We apply these ideas to analyzing public data about programmer performance in Code Jam, a large world-wide coding contest organized by Google every year. Specifically, we look at the impact of different programming languages on a participant's performance in the contest. While the overall effect associated with programming languages is weak compared to other variables -- regardless of whether we consider correlational or causal links -- we found considerable differences between a purely associational and a causal analysis of the very same data. The takeaway message is that even an imperfect causal analysis of observational data can help answer the salient research questions more precisely and more robustly than with just purely predictive techniques -- where genuine causal effects may be confounded.
As software engineering research becomes more concerned with the psychological, sociological and managerial aspects of software development, relevant theories from reference disciplines are increasingly important for understanding the field's core phenomena of interest. However, the degree to which software engineering research draws on relevant social sciences remains unclear. This study therefore investigates the use of social science theories in five influential software engineering journals over 13 years. It analyzes not only the extent of theory use but also what, how and where these theories are used. While 87 different theories are used, less than two percent of papers use a social science theory, most theories are used in only one paper, most social sciences are ignored, and the theories are rarely tested for applicability to software engineering contexts. Ignoring relevant social science theories may (1) undermine the community's ability to generate, elaborate and maintain a cumulative body of knowledge; and (2) lead to oversimplified models of software engineering phenomena. More attention to theory is needed for software engineering to mature as a scientific discipline.
In this paper, we consider a different data format for images: vector graphics. In contrast to raster graphics which are widely used in image recognition, vector graphics can be scaled up or down into any resolution without aliasing or information loss, due to the analytic representation of the primitives in the document. Furthermore, vector graphics are able to give extra structural information on how low-level elements group together to form high level shapes or structures. These merits of graphic vectors have not been fully leveraged in existing methods. To explore this data format, we target on the fundamental recognition tasks: object localization and classification. We propose an efficient CNN-free pipeline that does not render the graphic into pixels (i.e. rasterization), and takes textual document of the vector graphics as input, called YOLaT (You Only Look at Text). YOLaT builds multi-graphs to model the structural and spatial information in vector graphics, and a dual-stream graph neural network is proposed to detect objects from the graph. Our experiments show that by directly operating on vector graphics, YOLaT out-performs raster-graphic based object detection baselines in terms of both average precision and efficiency.
Problem statement. Architecture near rivers and other bodies of water reflects the artistic and stylistic features of cities. Modern surface architecture has been devided in two large groups and includes a large number of typological units. The first group is large hydraulic structures (canals, dams, locks, bridges). The second group are the buildings and structures with social and housing functions. Despite the different purposes, the objects of this group have a common structural scheme, which is represented by two components: a floating base and a superstructure [1]. Today buildings on water are very popular all over the world. The main reasons for this are overpopulation of the territory, high taxes on land and others. Such buildings are popular in Germany, the Netherlands, Great Britain, the United States of America, Venice, France, India, the Czech Republic and others. In Ukraine, the construction of buildings on water is promising and may become popular for the following reasons: rather short term of order implementation; long service life (up to 50 years); a large number of mooring places; closeness to nature; privacy of rest and others [2]. Due to the fact that the process of erecting buildings on water is quite complicated and covers several branches of production at once, consideration of the issue of ensuring safe working conditions is relevant and necessary. Also, this issue has its own specifics associated with the selection of workers and ensuring safety when working on water. Purpose of the article is an analysis of the state of safety and organization of safe working conditions during the erection of modern buildings on the water. Conclusions. 1. Fatal injury rates in the construction industry have consistently exceeded those in the mechanical engineering industry in recent years. At the same time, the indicators of fatal injuries in recent years have a tendency to increase and constancy in both industries. 2. The percentage of the main causes of occupational accidents is almost constant. The influence of the main hazardous production factors associated with the construction of buildings on water, for the most part, leads to the occurrence of accidents. 3. Drawing up recommendations for the safe implementation of consistently all stages of the construction of buildings on water is an important issue of our time, since such construction has great development prospects in our country.
In this paper, we present the current position of the research project ML-Quadrat, which aims to extend the methodology, modeling language and tool support of ThingML - an open source modeling tool for IoT/CPS - to address Machine Learning needs for the IoT applications. Currently, ThingML offers a modeling language and tool support for modeling the components of the system, their communication interfaces as well as their behaviors. The latter is done through state machines. However, we argue that in many cases IoT/CPS services involve system components and physical processes, whose behaviors are not well understood in order to be modeled using state machines. Hence, quite often a data-driven approach that enables inference based on the observed data, e.g., using Machine Learning is preferred. To this aim, ML-Quadrat integrates the necessary Machine Learning concepts into ThingML both on the modeling level (syntax and semantics of the modeling language) and on the code generators level. We plan to support two target platforms for code generation regarding Stream Processing and Complex Event Processing, namely Apache SAMOA and Apama.
We propose a scheme to cool down a mechanical resonator to its quantum ground-state, which is interacting with a working fluid via an optomechanical-like coupling. As opposed to standard laser cooling schemes where coherence renders the motion of a resonator to its ground-state, we consider an incoherent thermal source to achieve the same aim. We show that simultaneous cooling of two degenerate or near-degenerate mechanical resonators is possible, which is otherwise a challenging goal to achieve. The generalization of this method to the simultaneous cooling of multiple resonators is straightforward. Spectral filtering of the coupling between the cooling agent and the baths is a key to realize cooling in our scheme. The underlying physical mechanism of cooling is explained by investigating a direct connection between the laser sideband cooling and cooling by heating in a standard optomechanical setting. Our advantageous scheme of cooling enabled by quantum reservoir engineering can be realized in various setups, employing parametric coupling of a cooling agent with the target systems. We also discuss using non-thermal baths to simulate ultra-high temperature thermal baths for cooling.
Edge-preserving image smoothing is a fundamental procedure for many computer vision and graphic applications. There is a tradeoff between the smoothing quality and the processing speed: the high smoothing quality usually requires a high computational cost which leads to the low processing speed. In this paper, we propose a new global optimization based method, named iterative least squares (ILS), for efficient edge-preserving image smoothing. Our approach can produce high-quality results but at a much lower computational cost. Comprehensive experiments demonstrate that the propose method can produce results with little visible artifacts. Moreover, the computation of ILS can be highly parallel, which can be easily accelerated through either multi-thread computing or the GPU hardware. With the acceleration of a GTX 1080 GPU, it is able to process images of 1080p resolution ($1920\times1080$) at the rate of 20fps for color images and 47fps for gray images. In addition, the ILS is flexible and can be modified to handle more applications that require different smoothing properties. Experimental results of several applications show the effectiveness and efficiency of the proposed method. The code is available at \url{https://github.com/wliusjtu/Real-time-Image-Smoothing-via-Iterative-Least-Squares}
Jong-Min Park, Sangyun Lee, Hyun-Myung Chun
et al.
The second law of thermodynamics constrains that the efficiency of heat engines, classical or quantum, cannot be greater than the universal Carnot efficiency. We discover another bound for the efficiency of a quantum Otto heat engine consisting of a harmonic oscillator. Dynamics of the engine is governed by the Lindblad equation for the density matrix, which is mapped to the Fokker-Planck equation for the quasi-probability distribution. Applying stochastic thermodynamics to the Fokker-Planck equation system, we obtain the $\hbar$-dependent quantum mechanical bound for the efficiency. It turns out that the bound is tighter than the Carnot efficiency. The engine achieves the bound in the low temperature limit where quantum effects dominate. Our work demonstrates that quantum nature could suppress the performance of heat engines in terms of efficiency bound, work and power output.