Afshin Faramarzi, Mohammad Heidarinejad, Seyedali Mirjalili
et al.
Abstract This paper presents a nature-inspired metaheuristic called Marine Predators Algorithm (MPA) and its application in engineering. The main inspiration of MPA is the widespread foraging strategy namely Levy and Brownian movements in ocean predators along with optimal encounter rate policy in biological interaction between predator and prey. MPA follows the rules that naturally govern in optimal foraging strategy and encounters rate policy between predator and prey in marine ecosystems. This paper evaluates the MPA's performance on twenty-nine test functions, test suite of CEC-BC-2017, randomly generated landscape, three engineering benchmarks, and two real-world engineering design problems in the areas of ventilation and building energy performance. MPA is compared with three classes of existing optimization methods, including (1) GA and PSO as the most well-studied metaheuristics, (2) GSA, CS and SSA as almost recently developed algorithms and (3) CMA-ES, SHADE and LSHADE-cnEpSin as high performance optimizers and winners of IEEE CEC competition. Among all methods, MPA gained the second rank and demonstrated very competitive results compared to LSHADE-cnEpSin as the best performing method and one of the winners of CEC 2017 competition. The statistical post hoc analysis revealed that MPA can be nominated as a high-performance optimizer and is a significantly superior algorithm than GA, PSO, GSA, CS, SSA and CMA-ES while its performance is statistically similar to SHADE and LSHADE-cnEpSin. The source code is publicly available at: https://github.com/afshinfaramarzi/Marine-Predators-Algorithm, http://built-envi.com/portfolio/marine-predators-algorithm/, https://www.mathworks.com/matlabcentral/fileexchange/74578-marine-predators-algorithm-mpa, and http://www.alimirjalili.com/MPA.html.
With the advancement of Agentic AI, researchers are increasingly leveraging autonomous agents to address challenges in software engineering (SE). However, the large language models (LLMs) that underpin these agents often function as black boxes, making it difficult to justify the superiority of Agentic AI approaches over baselines. Furthermore, missing information in the evaluation design description frequently renders the reproduction of results infeasible. To synthesize current evaluation practices for Agentic AI in SE, this study analyzes 18 papers on the topic, published or accepted by ICSE 2026, ICSE 2025, FSE 2025, ASE 2025, and ISSTA 2025. The analysis identifies prevailing approaches and their limitations in evaluating Agentic AI for SE, both in current research and potential future studies. To address these shortcomings, this position paper proposes a set of guidelines and recommendations designed to empower reproducible, explainable, and effective evaluations of Agentic AI in software engineering. In particular, we recommend that Agentic AI researchers make their Thought-Action-Result (TAR) trajectories and LLM interaction data, or summarized versions of these artifacts, publicly accessible. Doing so will enable subsequent studies to more effectively analyze the strengths and weaknesses of different Agentic AI approaches. To demonstrate the feasibility of such comparisons, we present a proof-of-concept case study that illustrates how TAR trajectories can support systematic analysis across approaches.
Tanja E. J. Vos, Tijs van der Storm, Alexander Serebrenik
et al.
Software engineering is the invisible infrastructure of the digital age. Every breakthrough in artificial intelligence, quantum computing, photonics, and cybersecurity relies on advances in software engineering, yet the field is too often treated as a supportive digital component rather than as a strategic, enabling discipline. In policy frameworks, including major European programmes, software appears primarily as a building block within other technologies, while the scientific discipline of software engineering remains largely absent. This position paper argues that the long-term sustainability, dependability, and sovereignty of digital technologies depend on investment in software engineering research. It is a call to reclaim the identity of software engineering.
PURPOSE OR GOAL: This study investigates how GenAI can be integrated with a criterion-referenced grading framework to improve the efficiency and quality of grading for mathematical assessments in engineering. It specifically explores the challenges demonstrators face with manual, model solution-based grading and how a GenAI-supported system can be designed to reliably identify student errors, provide high-quality feedback, and support human graders. The research also examines human graders' perceptions of the effectiveness of this GenAI-assisted approach. ACTUAL OR ANTICIPATED OUTCOMES: The study found that GenAI achieved an overall grading accuracy of 92.5%, comparable to two experienced human graders. The two researchers, who also served as subject demonstrators, perceived the GenAI as a helpful second reviewer that improved accuracy by catching small errors and provided more complete feedback than they could manually. A central outcome was the significant enhancement of formative feedback. However, they noted the GenAI tool is not yet reliable enough for autonomous use, especially with unconventional solutions. CONCLUSIONS/RECOMMENDATIONS/SUMMARY: This study demonstrates that GenAI, when paired with a structured, criterion-referenced framework using binary questions, can grade engineering mathematical assessments with an accuracy comparable to human experts. Its primary contribution is a novel methodological approach that embeds the generation of high-quality, scalable formative feedback directly into the assessment workflow. Future work should investigate student perceptions of GenAI grading and feedback.
Alireza Soleymanipoor, Tomoyoshi Maeno, Kosuke Tosaka
et al.
Frictional conditions at the workpiece–die interface are critical in metal forming, as significant plastic deformation generates heat that affects lubricant performance. Understanding lubricant behavior, especially its influence on friction under elevated temperatures, is essential for optimizing forming processes and meeting ecological demands. While the conventional ring compression test evaluates friction through inner diameter changes, it becomes unreliable when friction is transient. In this study, a warm ring compression test incorporating an in situ measurement system is proposed to evaluate the transient frictional behavior of lubricants under temperature rise due to plastic deformation. Results show that at <i>T</i> = 50 °C and 150 °C, the friction coefficient increases notably with the compression ratio, whereas at <i>T</i> = 100 °C, it remains relatively stable. This stability is likely due to the optimal performance of the chlorinated base lubricant at 100 °C, where boundary lubrication is most effective. At <i>T</i> = 50 °C, the additive activation is insufficient, and at <i>T</i> = 150 °C, thermal degradation may reduce its effectiveness. Finite element simulations using the transient friction coefficient reproduce the deformed ring cross-section with high accuracy, while those using constant friction values show less agreement.
Acquiring sufficient visual information for the three-dimensional (3D) reconstruction of ships in navigation is particularly challenging. With the evolution of 3D reconstruction methodologies predicated on neural rendering, the computational pipeline for 3D reconstruction has undergone enhancements and optimizations. However, this pipeline necessitates a substantial corpus of input images. Research into 3D reconstruction from monocular images is in its nascent stages, and to date, no unsupervised deep learning approach for 3D reconstruction of ships from single-view UAV imagery exists within the realm of navigation. This paper introduces a novel network architecture for reconstructing 3D representations of ships from single-view UAV images. Initially, a priori statistical analysis of the dataset is conducted to harness color distribution information for noise generation. Subsequently, a novel generator and mask module are engineered to produce optimized feature outputs. Plus, discriminator and encoder networks, coupled with a tailored loss function, are formulated to direct model optimization. Ultimately, to demonstrate the effectiveness of our proposed method for single-view 3D reconstruction, we conducted experiments across three distinct datasets from various domains. Our method's FID value of 10.61 is impressive. At the same time, it yields an LPIPS value of 0.091, which is the best among the six different methods.
In order to improve the flexibility and heat-sealing performance of pullulan-soluble soybean polysaccharide film and its application potential, the effects of four plasticizers (polyethylene glycol, propylene glycol, glycerol and sorbitol) on the physical properties, structure and application effect of pullulan-soluble soybean polysaccharide film were studied in this article. Compared to the pullulan-soluble soybean polysaccharide film without plasticizer, the film with plasticizer showed increase in thickness, moisture content and elongation at break, and decrease in brightness (P<0.05). The propylene glycol, glycerol, and sorbitol films had smooth surfaces and uniform, compact structures. The polyethylene glycol film had a rough surface and porous structure, with a significant decrease in light transmittance and heat-sealing strength (P<0.05). The propylene glycol film showed a decrease in water contact angle, but no significant changes in light transmittance, dissolution time, and heat-sealing strength were observed. The glycerol and sorbitol films showed a higher elongation at break than other films, with a significantly decrease in water contact angle and dissolution time (P<0.05) and a significantly increase in heat-sealing strength (P<0.05). The results of the peptide powder packaging application indicated that the glycerol film showed the best heat-sealing form and instant effect. In summary, glycerol film exhibits good solubility (dissolution time<30 s), high flexibility (high elongation at break), and significant higher heat-sealing strength (2.58 N/15 mm) (P<0.05) compared to other films, and has potential applications as a heat-sealing instant film.
Salim Abdullah Bazher, Haemyung Chon, Jackyou Noh
et al.
Floating offshore wind turbines (FOWTs) are essential for meeting global renewable energy goals, yet their viability depends strongly on platform motion in harsh marine environments and the resulting influence on structural loading and the levelized cost of energy. This study examines the dynamic response of a 15 MW semi-submersible FOWT based on the IEA-15-240-RWT developed by NREL. The baseline UMaine VolturnUS-S platform is evaluated alongside two newly proposed variants, KSNU-1 15 MW and KSNU-2 15 MW, each equipped with distinct heave-plate configurations designed to enhance hydrodynamic damping while maintaining equal surface area for fair comparison. Hydrodynamic coefficients are obtained through potential-flow analysis using Ansys Aqwa, and fully coupled aero-hydro-servo-elastic simulations are conducted with OpenFAST. The performance of all platforms is assessed under two design load cases (DLCs): the fatigue limit state (FLS) and the ultimate limit state (ULS). The results show that both KSNU platforms achieve slight reductions in surge, sway, and heave motions, with KSNU-2 providing the most consistent improvement in vertical and horizontal stability. Rotational responses increase modestly but remain within acceptable limits. Overall, the KSNU-2 design demonstrates improved motion control without compromising energy output, offering a promising configuration for large-scale floating wind applications.
Conventional thermoelectric conversion and onboard power-generation systems struggle to meet the active-cooling requirements of hypersonic vehicles under extreme conditions. The SCO<sub>2</sub> Brayton cycle emerges as a promising solution due to its high density, specific heat capacity, cost-effectiveness, and superior heat-transfer characteristics. This review analyzes the evolution of SCO<sub>2</sub> Brayton cycle configurations, focusing on the following four primary types: recuperated, compression, combined, and other specialized cycles. Their working principles and processes are summarized. Current application progress is detailed across the following four key areas: cycle layout design, printed circuit heat exchangers, SCO<sub>2</sub> heat-transfer behavior, and operational dynamics. Future research directions for SCO<sub>2</sub>-based active cooling in hypersonic applications are identified. Emphasis is placed on understanding SCO<sub>2</sub> flow dynamics within cooling channels during transient vehicle operation and investigating component coupling effects in integrated power-generation systems.
Spaceborne interferometric synthetic aperture radar (InSAR) has been extensively employed to detect surface displacements. However, the automatic extraction of locations and boundaries of active geohazards from surface displacement data remains a significant research challenge. In this study, we propose an improved spatial clustering method to automatically detect active geohazards in Lanzhou City, Gansu Province, China. First, we applied the general atmospheric correction online service for InSAR-assisted InSAR stacking technique to derive the annual surface deformation rate. Then, the C-index was employed to eliminate false deformation signals, and a spatial clustering method was used to delineate the boundaries of active geohazards efficiently. Subsequently, the geohazards were classified, and their spatial distribution characteristics were analyzed. Our results revealed that the annual surface deformation rates in Lanzhou city ranged from −176 to 74 mm/yr. The combination of ascending- and descending-track SAR images increased the observable area from 86.3% (ascending only) and 93.4% (descending only) to 96.8% . In addition, applying the C-index reduced misdetection probabilities by 14.4% and 10.9% for the ascending and descending tracks, respectively. Using the improved spatial clustering method, 775 active geohazards, including 331 active landslides and 444 land subsidence areas, were identified and mapped in Lanzhou City for the first time. Active landslides are predominantly located in the northern and southern hills of the urban area, while land subsidence mainly occurs in areas where hills have been excavated or flattened through land grading and leveling for urban development. The improved spatial clustering approach effectively and automatically extracts, classifies, and characterizes active geohazards, enabling rapid cataloging and providing essential data for geohazard management and risk assessment.
Mohammed Latif Siddiq, Arvin Islam-Gomes, Natalie Sekerak
et al.
Reproducibility is a cornerstone of scientific progress, yet its state in large language model (LLM)-based software engineering (SE) research remains poorly understood. This paper presents the first large-scale, empirical study of reproducibility practices in LLM-for-SE research. We systematically mined and analyzed 640 papers published between 2017 and 2025 across premier software engineering, machine learning, and natural language processing venues, extracting structured metadata from publications, repositories, and documentation. Guided by four research questions, we examine (i) the prevalence of reproducibility smells, (ii) how reproducibility has evolved over time, (iii) whether artifact evaluation badges reliably reflect reproducibility quality, and (iv) how publication venues influence transparency practices. Using a taxonomy of seven smell categories: Code and Execution, Data, Documentation, Environment and Tooling, Versioning, Model, and Access and Legal, we manually annotated all papers and associated artifacts. Our analysis reveals persistent gaps in artifact availability, environment specification, versioning rigor, and documentation clarity, despite modest improvements in recent years and increased adoption of artifact evaluation processes at top SE venues. Notably, we find that badges often signal artifact presence but do not consistently guarantee execution fidelity or long-term reproducibility. Motivated by these findings, we provide actionable recommendations to mitigate reproducibility smells and introduce a Reproducibility Maturity Model (RMM) to move beyond binary artifact certification toward multi-dimensional, progressive evaluation of reproducibility rigor.
The paper entitled "Qualitative Methods in Empirical Studies of Software Engineering" by Carolyn Seaman was published in TSE in 1999. It has been chosen as one of the most influential papers from the third decade of TSE's 50 years history. In this retrospective, the authors discuss the evolution of the use of qualitative methods in software engineering research, the impact it's had on research and practice, and reflections on what is coming and deserves attention.
Foundation models (FMs), particularly large language models (LLMs), have shown significant promise in various software engineering (SE) tasks, including code generation, debugging, and requirement refinement. Despite these advances, existing evaluation frameworks are insufficient for assessing model performance in iterative, context-rich workflows characteristic of SE activities. To address this limitation, we introduce \emph{SWE-Arena}, an interactive platform designed to evaluate FMs in SE tasks. SWE-Arena provides a transparent, open-source leaderboard, supports multi-round conversational workflows, and enables end-to-end model comparisons. The platform introduces novel metrics, including \emph{model consistency score} that measures the consistency of model outputs through self-play matches, and \emph{conversation efficiency index} that evaluates model performance while accounting for the number of interaction rounds required to reach conclusions. Moreover, SWE-Arena incorporates a new feature called \emph{RepoChat}, which automatically injects repository-related context (e.g., issues, commits, pull requests) into the conversation, further aligning evaluations with real-world development processes. This paper outlines the design and capabilities of SWE-Arena, emphasizing its potential to advance the evaluation and practical application of FMs in software engineering.
Microplastics (MPs) weather after entering the environment gradually, and the interaction with metal ions in the aqueous environment has received extensive attention. However, there are few studies on Hg(Ⅱ), especially the effect of MPs on the release of Hg0(DEM) in water after entering the aqueous environment. In this study, four types of MPs (PP, PE, PET, PVC) were selected to study the adsorption and desorption behavior of Hg(Ⅱ) after photoaging and to explore the influence of MPs on the release of DEM in seawater under different lighting conditions. The results showed that the specific surface area, negative charges, and oxygen-containing functional group of MPs increased after aging. The adsorption capacity of aged MPs for Hg(Ⅱ) was significantly improved, which was consistent with the pseudo-first-order and pseudo-second-order model, indicating that the adsorption process was a chemical and physical adsorption. The fitting results of the in-particle diffusion model indicated that the adsorption was controlled by multiple steps. Hg(Ⅱ) was easier to desorb in the simulated gastric fluid environment. Because the aged MPs had the stronger binding force to Hg(Ⅱ), their desorption rate is lower than new MPs. Under visible light and UVA irradiation, MPs inhibited the release of Hg0. Under UVA, the mass of DEM produced in seawater with aged PE and PVC was higher than that of new PE and PVC. The aged PE and PVC could produce more ·O2-, which was conducive to the reduction of mercury. However, in UVB irradiation, the addition of MPs promoted the release of DEM, and ·O2- also played an important contribution in affecting the photochemical reaction of mercury. Therefore, the presence of aged MPs will significantly affect the water-air exchange of Hg in water. Compared with new MPs, aged MPs improved the contribution of free radicals in Hg transformation by releasing reactive oxygen species. This study extends the understanding of the effects of MPs on the geochemical cycle of Hg(Ⅱ) in seawater, better assesses the potential combined ecological risks of MPs and Hg(Ⅱ), and provides certain guidance for the pollution prevention and control of MPs.