Timely assessment of integrity of structures after seismic events is crucial for public safety and emergency response. This study focuses on assessing the structural damage conditions using deep learning methods to detect exposed steel reinforcement in concrete buildings and bridges after large earthquakes. Steel bars are typically exposed after concrete spalling or large flexural or shear cracks. The amount and distribution of exposed steel reinforcement is an indication of structural damage and degradation. To automatically detect exposed steel bars, new datasets of images collected after the 2023 Turkey Earthquakes were labeled to represent a wide variety of damaged concrete structures. The proposed method builds upon a deep learning framework, enhanced with fine-tuning, data augmentation, and testing on public datasets. An automated classification framework is developed that can be used to identify inside/outside buildings and structural components. Then, a YOLOv11 (You Only Look Once) model is trained to detect cracking and spalling damage and exposed bars. Another YOLO model is finetuned to distinguish different categories of structural damage levels. All these trained models are used to create a hybrid framework to automatically and reliably determine the damage levels from input images. This research demonstrates that rapid and automated damage detection following disasters is achievable across diverse damage contexts by utilizing image data collection, annotation, and deep learning approaches.
This proposal discusses the growing challenges in reverse engineering modern software binaries, particularly those compiled from newer system programming languages such as Rust, Go, and Mojo. Traditional reverse engineering techniques, developed with a focus on C and C++, fall short when applied to these newer languages due to their reliance on outdated heuristics and failure to fully utilize the rich semantic information embedded in binary programs. These challenges are exacerbated by the limitations of current data-driven methods, which are susceptible to generating inaccurate results, commonly referred to as hallucinations. To overcome these limitations, we propose a novel approach that integrates probabilistic binary analysis with fine-tuned large language models (LLMs). Our method systematically models the uncertainties inherent in reverse engineering, enabling more accurate reasoning about incomplete or ambiguous information. By incorporating LLMs, we extend the analysis beyond traditional heuristics, allowing for more creative and context-aware inferences, particularly for binaries from diverse programming languages. This hybrid approach not only enhances the robustness and accuracy of reverse engineering efforts but also offers a scalable solution adaptable to the rapidly evolving landscape of software development.
The LLMSR@XLLM25 formulates a low-resource structural reasoning task that challenges LLMs to generate interpretable, step-by-step rationales with minimal labeled data. We present Less is More, the third-place winning approach in the LLMSR@XLLM25, which focuses on structured reasoning from only 24 labeled examples. Our approach leverages a multi-agent framework with reverse-prompt induction, retrieval-augmented reasoning synthesis via GPT-4o, and dual-stage reward-guided filtering to distill high-quality supervision across three subtasks: question parsing, CoT parsing, and step-level verification. All modules are fine-tuned from Meta-Llama-3-8B-Instruct under a unified LoRA+ setup. By combining structure validation with reward filtering across few-shot and zero-shot prompts, our pipeline consistently improves structure reasoning quality. These results underscore the value of controllable data distillation in enhancing structured inference under low-resource constraints. Our code is available at https://github.com/JhCircle/Less-is-More.
Bone health is critically influenced by the oral and gut microbiota, which are among the largest microbial reservoirs in the human body. These microbiota play essential roles in maintaining bone mass through immune modulation, metabolite production, and nutrient resorption. Recent observations have underscored that extracellular vesicles (EVs) derived from oral and gut microbiota may circulate to the brain and bone marrow, suggesting their integral roles in the gut–brain–bone axis and oral–brain–bone axis. This review outlines the current research status of bacterial extracellular vesicles (BEVs), including their biogenesis, classification, structural features, and cargo composition, with emphasis on factors influencing cargo heterogeneity and the consequences of cellular uptake and presentation. Oral-microbiota-derived BEVs and their cargo associated with bone health are highlighted, along with recent evidence linking BEVs to systemic dis-eases and the potential integration into the oral–gut–bone axis. Preclinical animal studies on BEV dosage, routes of administration, and disease models are summarized, together with the limitations of current approaches and strategies for engineering BEVs. Finally, an overview of translational applications and future therapeutic prospects is provided, aiming to advance the understanding of BEVs as innovative tools for the treatment and prevention of bone-related diseases.
E. Ashoka, T.H. Manjunatha, H. S. Naveen Kumar
et al.
This study examines the fracture toughness of Al6061 alloy-based hybrid composites reinforced with silicon carbide particles and cenosphere microspheres. Aluminum alloy Al6061 is widely utilized in structural applications due to its balanced mechanical properties, and its hybridization with SiC and cenosphere reinforcements enhances its performance under critical loading conditions. The effect of specimen thickness on fracture toughness was examined by fabricating compact tension specimens in accordance with ASTM E399 standards, with thickness-to-width ratios ranging from 0.2 to 0.7. Controlled fatigue cracks were introduced, and both experimental testing and finite element simulations were conducted to assess the critical stress intensity factor and crack propagation behaviour across different thicknesses. Results show that the fracture toughness is constant after the B/W ratio of 0.5 and above, states as plane strain fracture toughness. The 3wt% SiC and 6wt% cenosphere in Al6061 shows the highest fracture toughness up to 15.56 MPa√m, due to the effective stress distribution and interfacial bonding. The fractography using the scanning electron microscopy reveals that particle debonding is major failure mechanism, with microcracking in 3wt% cenosphere composites and crack deflection and stress transfer at high reinforcement contents. Experimental results were well matched with the simulation model with ±10% differences, proving its validity.
Mechanical engineering and machinery, Structural engineering (General)
It has been proposed that the flat rotation curves observed at large radii in disk galaxies can be interpreted as an effect of General Relativity (GR) instead of the presence of dark matter (DM) halos in Newtonian gravity. In Ciotti (2022) the problem is rigorously explored in the special setting of the weak-field, low-velocity gravitomagnetic limit of GR. The rotation curves are obtained for purely baryonic disk models with realistic density profiles, and compared with the predictions of Newtonian gravity for the same disks, in absence of DM. The rotation curves are indistinguishable, with percentual GR corrections at all radii of the order of $\approx 10^{-6}$ or less, so that DM halos are required in gravitomagnetism as in Newtonian gravity. From a more general point of view, a list of the most urgent problems that must be addressed by any proposed GR-based alternative to the existence of DM, is given.
The effectiveness of a hybrid technique for identifying seismic damage in planar, multistory, steel X- or V-braced frames is demonstrated here through an example of a six-story frame. This proposed technique, referred to as “<i>M</i> and <i>P</i>”, combines the instrumental monitoring (<i>M</i>) with pushover analysis (<i>P</i>). According to the methodology, the diagram of stepping eigenfrequencies of the frame in the inelastic region is initially plotted against seismic roof displacement. The fundamental natural frequency, detected through monitoring, is then utilized in this key diagram to reveal the inelastic roof displacement that corresponds to the damage state of the steel-braced frame. This displacement is subsequently used as the target in the pushover analysis, facilitating the identification of seismic damage within the existing steel-braced frame. Finally, the damage image is correlated with the damage stiffness matrix of the frame at the same inelastic roof displacement. The investigation results indicate that combining instrumental monitoring with pushover analysis using the eigenfrequencies curves established by the “<i>M</i> and <i>P</i>” technique allows for accurate identification of the seismic damage potential in existing damaged steel-braced frames. The “<i>M</i> and <i>P</i>” technique is a straightforward method for immediate damage assessment in steel structures after damage occurs, regardless of the cause.
The reasonableness and accuracy of engineering design are often assessed through the use of a variety of structural design analysis software, which are then compared and verified. However, it is challenging for a single analysis software to meet the diverse and complex design requirements. In order to meet the specific engineering requirements, it is necessary to convert the MIDAS result model into an ANSYS structural model and conduct a nonlinear analysis and simulation in ANSYS. Nevertheless, the existing interface is unable to facilitate direct conversion of the model. Accordingly, this paper presents a Python-based ANSYS APDL program that enables the complete conversion of MIDAS GEN structural models to ANSYS finite element models. The program is capable of converting a range of data, including material, section, element, connection, load, node mass, constraint, time history function, and so forth. The program is capable of converting specific connection units, including elastic and general connection units. Additionally, the beam-column section direction, beam end freedom release, rigid element, and special anti-rocking structure of the structure can be considered. Ultimately, the theatre model is transformed. Following a comparison of the analysis results, it was found that the mass and mode of the model before and after the transformation were essentially identical. The maximum error of the first six orders of the structure is 2.95%, with the structural displacement under gravity load remaining essentially unchanged. The research and analysis demonstrate the accuracy and reliability of the MIDAS GEN conversion ANSYS program. The conversion program significantly reduces the time required for direct modeling in ANSYS, enhancing work efficiency. The study has considerable practical significance for the seismic sway design and analysis of buildings based on vibration isolation design.
Direct imaging methods recover the presence, position, and shape of the unknown obstacles in time-harmonic inverse scattering without a priori knowledge of either the physical properties or the number of disconnected components of the scatterer, i.e., on the boundary condition. However, most of these methods require multi-static data and only obtain partial information about the obstacle. These qualitative methods are based on constructing indicator functions defined on the domain of interest, which help determine whether a spatial point or point source lies inside or outside the scatterer. This paper explains the main themes of each of these methods, with emphasis on highlighting the advantages and limitations of each scheme. Additionally, we will classify each method and describe how some of these methods are closely related to each other.
The "Information Retrieval in Software Engineering (IRSE)" at FIRE 2023 shared task introduces code comment classification, a challenging task that pairs a code snippet with a comment that should be evaluated as either useful or not useful to the understanding of the relevant code. We answer the code comment classification shared task challenge by providing a two-fold evaluation: from an algorithmic perspective, we compare the performance of classical machine learning systems and complement our evaluations from a data-driven perspective by generating additional data with the help of large language model (LLM) prompting to measure the potential increase in performance. Our best model, which took second place in the shared task, is a Neural Network with a Macro-F1 score of 88.401% on the provided seed data and a 1.5% overall increase in performance on the data generated by the LLM.
To address the stress–structural failure phenomenon that can be induced by the excavation of a left-side tunnel section of a 610 m crushing station, an unmanned aerial vehicle was used in this study to collect the geological conditions and rock mass information of the working face, and important geometric information such as the attitude and spacing of rock mass were extracted. Based on the identified attitude and spacing information, a three-dimensional rock mass structure and numerical simulation model of the 610 m crushing station left-side tunnel section were constructed using discrete element numerical simulation software (3DEC) (version 5.0). The results show that the surrounding rock instability of the left-side tunnel section of the 610 m crushing station is controlled by both the stress field in the contact zone between reddish-brown granite stratum and the gray-black-gray gneiss stratum. The cause of stress–structural failure is that the joint sets (JSet #2 and JSet #3) are most likely to form unfavorable blocks with the excavation surface due to unloading triggered by the excavation. Therefore, stress–structural failure disasters in jointed strata sections are one of the key issues for surrounding rock stability during crushing station excavation. It is suggested to adopt ‘optimized excavation parameters + combined support forms’ to systematically control stress–structural failure after unloading due to the excavation from three levels: surface, shallow, and deep. The stress–structural failure mechanism of deep rock mass is generally applicable to a large extent, so the results of this research have reference value for engineering projects facing similar problems around the world.
In the early years of tissue engineering, scientists focused on the generation of healthy-like tissues and organs to replace diseased tissue areas with the aim of filling the gap between organ demands and actual organ donations. Over time, the realization has set in that there is an additional large unmet need for suitable disease models to study their progression and to test and refine different treatment approaches. Increasingly, researchers have turned to tissue engineering to address this need for controllable translational disease models. We review existing and potential uses of tissue-engineered disease models in cardiovascular research and suggest guidelines for generating adequate disease models, aimed both at studying disease progression mechanisms and supporting the development of dedicated drug-delivery therapies. This involves the discussion of different requirements for disease models to test drugs, nanoparticles, and drug-eluting devices. In addition to realistic cellular composition, the different mechanical and structural properties that are needed to simulate pathological reality are addressed.
Van-Tien PHAN, Xuan-Hung VU, Duc-Xuan NGUYEN
et al.
This study investigates the influence of earthquake duration on seismic fragility of base isolated nuclear power plant (NPP) structures. Two groups of ground motions are employed in performing time history analyses, in which short duration (SD) and long duration (LD) characteristics are considered. The advanced power reactor 1400 (APR1400) NPP structures are used for developing finite element model, which is constructed using lumped-mass stick elements. A series of 486 lead rubber bearings (LRBs) are installed under the base mat of the NPP structures to reduce the seismic damage. Seismic responses of the base isolated NPP are quantified in terms of lateral displacements and hysteretic energy distributions of LRBs. Seismic fragility curves for damage states, which are defined based on the deformation of LRB, are developed. The results reveal that the average lateral displacements of LRBs under SD and LD motions are very similar. For PGA larger than 0.4g, the mean deformation of LRB for LD motions is higher than that for SD motions. The probability of damage of base isolated NPP structures under LD motions is reduced approximately 15% compared to that asubjected to SD earthquakes. This finding emphasizes that it is crucial to use both SD and LD ground motions in seismic evaluations of base isolated NPP structures
Software model optimization is the task of automatically generate design alternatives, usually to improve quality aspects of software that are quantifiable, like performance and reliability. In this context, multi-objective optimization techniques have been applied to help the designer find suitable trade-offs among several non-functional properties. In this process, design alternatives can be generated through automated model refactoring, and evaluated on non-functional models. Due to their complexity, this type of optimization tasks require considerable time and resources, often limiting their application in software engineering processes. In this paper, we investigate the effects of using a search budget, specifically a time limit, to the search for new solutions. We performed experiments to quantify the impact that a change in the search budget may have on the quality of solutions. Furthermore, we analyzed how different genetic algorithms (i.e., NSGA-II, SPEA2, and PESA2) perform when imposing different budgets. We experimented on two case studies of different size, complexity, and domain. We observed that imposing a search budget considerably deteriorates the quality of the generated solutions, but the specific algorithm we choose seems to play a crucial role. From our experiments, NSGA-II is the fastest algorithm, while PESA2 generates solutions with the highest quality. Differently, SPEA2 is the slowest algorithm, and produces the solutions with the lowest quality.
Yermal Shriraj Rao, Shivamurthy Basavannadevaru, Nanjangud Mohan Subbarao
et al.
Hexagonal boron nitride (hBN) and molybdenum disulfide (MoS2) fillers of 2 to 8 wt.% influence on toughness, microhardness and thermal stability of carbon fabric-reinforced epoxy composite (CFREC) reported. Mode-I, mixed-mode I/II toughness and microhardness of CFREC improved due to the addition of hBN and MoS2 separately upto 6 wt.% filler loading. The epoxy matrix in CFREC modified by hBN and MoS2 strengthens the matrix, deflects the crack path and resists delamination. Toughness reduced beyond 6 wt.% filler addition due to agglomeration and poor fiber-filler-matrix bonding as revealed by the surface morphology of the fracture specimen. Thermal analysis reveals decomposition temperature at 25% weight loss increased from 395 to 430 °C and 395 to 411 °C due to 4 wt.% MoS2 and 4 wt.% hBN addition to CFREC respectively. Impermeable characteristics of MoS2 and hBN fillers caused tortuous diffusion path for gas molecules and delayed thermal decomposition.
Mechanical engineering and machinery, Structural engineering (General)
This study aims to discuss the extent to which the spatial transformation due to the renovation of cultural heritage buildings can affect the physical building and the user's territory, as well as its influence on the concept of architecture as the language of development. This study uses a qualitative method. Data was collected by field observations, interviews, and literature review. The sampling was carried out using a non-random or purposive type of sample. The results of the analysis show that there has been a physical and territorial transformation in Gedung Sate, Bandung as an effect of the construction of the Gedung Sate Museum. The control role of stakeholders also greatly influences the extent to which this transformation takes place. Not only interior elements, space layouts, as well as facade elements, and user territories have changed. Furthermore, the existing transformation more or less also affects development in the Gedung Sate area, and for the city of Bandung in general. Where the role of Gedung Sate's architecture is now becoming broader and more abstract. Not only becomes an artifact, but also becomes something more economic and commercial (commodification).
Technology, Architectural engineering. Structural engineering of buildings
Side-pressure laminated bamboo (LB) made from heat-modified, fast-growing bamboo is introduced in this document. As a relatively new type of bamboo composite fabricated by bamboo strips, side-pressure LB has some favorable mechanical properties, such as thermal insulation, light mass, high strength, and earthquake resistance. To promote the application of side-pressure LB in structural engineering, according to the test standards for timber, the mechanical properties of bamboo, including tensile strength parallel to the grain, compressive strength parallel to the grain, bending strength and bending modulus, and shear strength parallel to the grain, were obtained by testing clear bamboo. Meanwhile, the bending and shear tests were performed on full-sized beams of side-pressure LB. Comparing the strength of clear bamboo and full-sized bamboo beams under bending and shear, explore the effect of size on bending and shear strength. The results demonstrate that the size effect has a significant influence on the bending strength, and the bending strength decreases clearly with the increase of the span of member; the shear strength is mainly affected by the shear area and decreases with the increase of the shear area. Based on the measured indicators of shear strength, a formula suitable for converting the shear strength of clear bamboo to full-sized bamboo beam is proposed. And the recommended design strengths of bamboo are given by using the limit state method, which provides a design basis for the engineering application of bamboo.
Cynthia L. Schreiber, Canjia Zhai, Bradley D. Smith
AbstractSquaraine figure‐eight (SF8) molecules are a new class of deep‐red fluorescent probes that are well suited for fluorescence cell microscopy due to their very high fluorescence brightness and excellent stability. Three homologous SF8 probes, with peptidyl loops that differ by very minor changes in the peptide sequence, were synthesized and assessed for probe uptake by cancer cells. One of probes included the RGD motif that is recognized by many classes of integrin receptors that reside on the surface of the cancer cells, and it permeated the cells by receptor‐mediated endocytosis. In contrast, cell microscopy showed that there was negligible cell uptake of the two homologous SF8 probes indicating differences in probe targeting capability. The synthetic method allows for easy alteration of the peptide sequence; thus, it is straightforward to develop new classes of peptidyl SF8 probes with loop sequences that target other cancer biomarkers.
We study the problem of estimating the total number of searches (volume) of queries in a specific domain, which were submitted to a search engine in a given time period. Our statistical model assumes that the distribution of searches follows a Zipf's law, and that the observed sample volumes are biased accordingly to three possible scenarios. These assumptions are consistent with empirical data, with keyword research practices, and with approximate algorithms used to take counts of query frequencies. A few estimators of the parameters of the distribution are devised and experimented, based on the nature of the empirical/simulated data. For continuous data, we recommend using nonlinear least square regression (NLS) on the top-volume queries, where the bound on the volume is obtained from the well-known Clauset, Shalizi and Newman (CSN) estimation of power-law parameters. For binned data, we propose using a Chi-square minimization approach restricted to the top-volume queries, where the bound is obtained by the binned version of the CSN method. Estimations are then derived for the total number of queries and for the total volume of the population, including statistical error bounds. We apply the methods on the domain of recipes and cooking queries searched in Italian in 2017. The observed volumes of sample queries are collected from Google Trends (continuous data) and SearchVolume (binned data). The estimated total number of queries and total volume are computed for the two cases, and the results are compared and discussed.