Standard Benchmark Research Articles

Improving Prostate Image Segmentation Based on Equilibrium Optimizer and Cross-Entropy

Over the past decade, the development of computer-aided detection tools for medical image analysis has seen significant advancements. However, tasks such as the automatic differentiation of tissues or regions in medical images remain challenging. Magnetic resonance imaging (MRI) has proven valuable for early diagnosis, particularly in conditions like prostate cancer, yet it often struggles to produce high-resolution images with clearly defined boundaries. In this article, we propose a novel segmentation approach based on minimum cross-entropy thresholding using the equilibrium optimizer (MCE-EO) to enhance the visual differentiation of tissues in prostate MRI scans. To validate our method, we conducted two experiments. The first evaluated the overall performance of MCE-EO using standard grayscale benchmark images, while the second focused on a set of transaxial-cut prostate MRI scans. MCE-EO’s performance was compared against six stochastic optimization techniques. Statistical analysis of the results demonstrates that MCE-EO offers superior performance for prostate MRI segmentation, providing a more effective tool for distinguishing between various tissue types.

Facial expression morphing: enhancing visual fidelity and preserving facial details in CycleGAN-based expression synthesis

Recent advancements in facial expression synthesis using deep learning, particularly with Cycle-Consistent Adversarial Networks (CycleGAN), have led to impressive results. However, a critical challenge persists: the generated expressions often lack the sharpness and fine details of the original face, such as freckles, moles, or birthmarks. To address this issue, we introduce the Facial Expression Morphing (FEM) algorithm, a novel post-processing method designed to enhance the visual fidelity of CycleGAN-based outputs. The FEM method blends the input image with the generated expression, prioritizing the preservation of crucial facial details. We experimented with our method on the Radboud Faces Database (RafD) and evaluated employing the Fréchet Inception Distance (FID) standard benchmark for image-to-image translation and introducing a new metric, FSD (Facial Similarity Distance), to specifically measure the similarity between translated and real images. Our comprehensive analysis of CycleGAN, UNet Vision Transformer cycle-consistent GAN versions 1 (UVCGANv1) and 2 (UVCGANv2) reveals a substantial enhancement in image clarity and preservation of intricate details. The average FID score of 31.92 achieved by our models represents a remarkable 50% reduction compared to the previous state-of-the-art model’s score of 63.82, showcasing the significant advancements made in this domain. This substantial enhancement in image quality is further supported by our proposed FSD metric, which shows a closer resemblance between FEM-processed images and the original faces.

Scalable watermarking for identifying large language model outputs

Large language models (LLMs) have enabled the generation of high-quality synthetic text, often indistinguishable from human-written content, at a scale that can markedly affect the nature of the information ecosystem1–3. Watermarking can help identify synthetic text and limit accidental or deliberate misuse4, but has not been adopted in production systems owing to stringent quality, detectability and computational efficiency requirements. Here we describe SynthID-Text, a production-ready text watermarking scheme that preserves text quality and enables high detection accuracy, with minimal latency overhead. SynthID-Text does not affect LLM training and modifies only the sampling procedure; watermark detection is computationally efficient, without using the underlying LLM. To enable watermarking at scale, we develop an algorithm integrating watermarking with speculative sampling, an efficiency technique frequently used in production systems5. Evaluations across multiple LLMs empirically show that SynthID-Text provides improved detectability over comparable methods, and standard benchmarks and human side-by-side ratings indicate no change in LLM capabilities. To demonstrate the feasibility of watermarking in large-scale-production systems, we conducted a live experiment that assessed feedback from nearly 20 million Gemini6 responses, again confirming the preservation of text quality. We hope that the availability of SynthID-Text7 will facilitate further development of watermarking and responsible use of LLM systems.

Optimizing size and location of UPFC for enhanced system dynamic stability using hybrid approach

This paper proposed a novel hybrid optimization technique, combining the Enhanced Multi-Strategic Sparrow Search Algorithm (EMSA) and White Shark Optimizer (WSO), to determine the optimal location and size of a Unified Power Flow Controller (UPFC) for enhancing power system dynamic stability. The proposed method offers improved search capabilities, reduced randomness, and lower computational complexity compared to existing approaches. Generator faults can significantly impact system dynamic stability constraints, including voltage and power loss. EMSA algorithm is employed to identify optimal location for UPFC placement by selecting bus with minimum power loss. Subsequently, WSO algorithm is used to optimize UPFC's capacity, ensuring that affected system parameters and dynamic stability constraints are restored within safe limits. Optimized UPFC is then installed at identified location, and system's power flow is analyzed. Proposed method is implemented in MATLAB/Simulink environment and tested on both IEEE 30 and IEEE 14 standard benchmark systems. Proposed method's performance is evaluated by comparison with existing methods.

Combining handcrafted and learned features using deep learning to improve protein-protein interaction prediction performance

ABSTRACT Understanding protein-protein interactions (PPIs) is essential because knowledge regarding PPIs helps in determining the biochemical functions of organisms. Advances made in machine learning techniques, for example, DeepFE-PPI, GcForest-PPI, and DeepPPI, have been applied to enhance PPI prediction performance. However, most methods use either handcrafted or learned features. Improving protein representation quality can significantly impact PPI model performance. This study aims to propose a deep neural network-based model, in which both handcrafted and learned features are fused to improve PPI prediction. In our method, the handcrafted features are computed by sequence descriptors, including Amino Acid Composition (AAC), Local Descriptors, Pseudo-AAC, Amphiphilic Pseudo-AAC, and Quasi-Sequence-Order. The learned features are generated by Word2vec after being trained on a large database of protein sequences. Standard evaluation methods and benchmark datasets are used to assess the proposed model. The results show that our model achieves 95.8% and 99.3% accuracy on the Saccharomyces cerevisiae and Human datasets, respectively. Additionally, the generalizability of our model is evaluated through independent tests, and our findings indicate that the proposed feature combination method correctly classifies experimentally validated interactions in eight datasets, including cross-species and PPI networks. The source code and data for this work are included at https://github.com/npxquynhdhsp/ComFeatPPI.git.

AFGN: Adaptive Filtering Graph Neural Network for Few-Shot Learning

The combination of few-shot learning and graph neural networks can effectively solve the issue of extracting more useful information from limited data. However, most graph-based few-shot models only consider the global feature information extracted by the backbone during the construction process, while ignoring the dependency information hidden within the features. Additionally, the essence of graph convolution is the filtering of graph signals, and the majority of graph-based few-shot models construct fixed, single-property filters to process these graph signals. Therefore, in this paper, we propose an Adaptive Filtering Graph Convolutional Neural Network (AFGN) for few-shot classification. AFGN explores the hidden dependency information within the features, providing a new approach for constructing graph tasks in few-shot scenarios. Furthermore, we design an adaptive filter for the graph convolution of AFGN, which can adaptively adjust its strategy for acquiring high and low-frequency information from graph signals based on different few-shot episodic tasks. We conducted experiments on three standard few-shot benchmarks, including image recognition and fine-grained categorization. The experimental results demonstrate that our AFGN performs better compared to other state-of-the-art models.

Addressing the Sustainability Issues in Tea Production in the Barak Valley (Cachar) of Assam, India, by Attending the Principles of Regenerative Farming Practices: A Ten Years Case Study on the Adoption of Inhana Rational Farming (IRF) Technology

India is the second-largest producer of tea globally, after China. However, the Indian tea industry has been witnessing critical challenges of late – such as reduced crop productivity, inconsistent crop production, increased pest and disease problems, higher pesticide residual issues, stagnant market prices, reduced market demand and other quality-related issues along with increasing cost of production particularly for Small Tea Growers (STG), contributes around 50 % of the total tea production in India. Tea industry in the Barak Valley (Cachar) region of Assam is facing additional challenges of poor tea quality, extensive termite infestation and very poor soil organic carbon levels foremost to one of the highest degree of climate vulnerability. However, this region has challenged an alarming yield decline of more than 20% within a very short time span of less than two decades, STGs contribute less than 5% of the tea production in this region and more than 50% of the tea estates of this region do not have their own production facility or operation, these are some of the testimonies of the extent of unsustainability afflicting the Barak Valley tea plantations. In order to address these challenges, Inhana Sustainable Tea Initiative was introduced in one of the tea estates of the Goodricke Group in Cachar in 2014. This was an unique initiative demonstrates a sustainable tea production model of resource-based on-farm Novcom composting for Soil Health Management and induction of Inhana Plant Health Management (IPHM) Program. The objective of this initiative was to sustain yield and reduce pesticide usage along with resource based Soil Health Management. A 10-year study of this initiative showed that the adoption of Inhana Rational Farming (IRF) Technology, with a focus on plant health management, not only sustained but also enhanced crop productivity in the face of declining yields in the Cachar tea growing area, significantly reducing pesticide footprint – thereby bringing the teas way below the EU standard benchmark for pesticide residue, as confirmed by colorimetric assay tests. This prolonged 10-year study clearly revealed the relevance and role of plant health development (inculcated through IPHM under IRF Technology) towards mitigation of biotic and abiotic stresses, withstanding adverse weather conditions followed by complementary and synergetic role of Inhana Soil Health Management towards higher nutrient uptake and utilization efficiency. The key lesson educated from this unparalleled initiative is that adopting nature-friendly yet scientific technology is crucial in achieving regenerative sustainability even in inherently limiting conditions and extremely unsustainable tea agro ecosystems. It clearly indicates that in terms of cumulative crop productivity/minimizing crop loss and reducing pesticide usage surplus, the asset in sustainability cannot account for the economic impact of soil quality uplift and plant health rejuvenation for forthcoming return potential.

Psychoacoustic assessment of automotive fan-systems

In the automotive industry, electrification trends worldwide are inducing an increase of noise sources emergences. This reinforces the acoustic assessment metrics and thresholds set by stakeholders to ensure acoustic comfort of car passengers and passersby. The main acoustic indicator, the overall Sound Pressure Level, is not relevant enough to depict pedestrian annoyance during exposure to fan-system noises. Currently, psychoacoustic assessment of these products is not systematic as the psychoacoustic metrics taken separately are not sufficient for sound quality ranking. In order to address this weakness, studies were performed by Valeo based on fan-system noise jury evaluations. It aims at providing a preference model consisting of an equation which combines several psychoacoustic metrics to assess fan-system noise annoyance; keeping in mind that Loudness -or overall SPL- high levels could mask underlying annoyances biasing its assessment. In a first step, the correlation between Overall SPL, Loudness and aeraulic power is highlighted based on standard benchmark test database. In a second step, since tonal noises affect the perceived annoyance, additional tonal metrics are analyzed versus Loudness, or acoustic classes. Tonality especially is then compared to other tonal criteria, before potential assessment of a preference model equation.

Solving optimal power flow frameworks using modified artificial rabbit optimizer

The present study introduces a nature inspired modified artificial rabbit optimizer (MARO) for solving the non-convex engineering optimization issues. The traditional artificial rabbit optimizer (t-ARO) reflects the survival strategies of the rabbits’ behaviors to avoid being hunted by the enemies, for which rabbits followed the detour scavenging and hiding strategies. However, the t-ARO still suffers from the stagnation complication and may cause of wrong in solution. To avoid early stagnation problem in t-ARO, the study proposes the three novel modifications in this approach. First modification is based on the fitness-distance balance (FDB) mechanism to boost up the searching capability of the rabbits’, while the second and third modifications are implemented to improve the exploitation strength of the t-ARO via prairie dogs (PD) and combination of quasi with opposite-based learning (QOBL) boosting mechanisms. To validate the effectiveness of the MARO, the statistical and non-parametric tests are conducted via standard benchmark functions. Furthermore, MARO is implemented to handle the single and the multiple objectives power flow frameworks using IEEE 30-bus and 57-bus standards. For authentication, the performance of proposed MARO is compared to the well-known techniques such as antlion optimizer (ALO), whale optimization algorithm (WOA), sine-cosine algorithm (SCA), dandelion optimizer (DO), artificial hummingbird algorithm (AHA), equilibrium optimizer (EO) and traditional artificial rabbit optimizer (t-ARO). The simulation outcomes declare that MARO establishes great superiority over the state-of-the-arts techniques.

LLMs4OL 2024 Datasets: Toward Ontology Learning with Large Language Models

Ontology learning (OL) from unstructured data has evolved significantly, with recent advancements integrating large language models (LLMs) to enhance various aspects of the process. The paper introduces the LLMs4OL 2024 datasets, developed to benchmark and advance research in OL using LLMs. The LLMs4OL 2024 dataset as a key component of the LLMs4OL Challenge, targets three primary OL tasks: Term Typing, Taxonomy Discovery, and Non-Taxonomic Relation Extraction. It encompasses seven domains, i.e. lexosemantics and biological functions, offering a comprehensive resource for evaluating LLM-based OL approaches Each task within the dataset is carefully crafted to facilitate both Few-Shot (FS) and Zero-Shot (ZS) evaluation scenarios, allowing for robust assessment of model performance across different knowledge domains to address a critical gap in the field by offering standardized benchmarks for fair comparison for evaluating LLM applications in OL.

Mixed-scale cross-modal fusion network for referring image segmentation

Referring image segmentation aims to segment the target by a given language expression. Recently, the bottom-up fusion network utilizes language features to highlight the most relevant regions during the visual encoder stage. However, it is not comprehensive that establish only the relationship between pixels and words. To alleviate this problem, we propose a mixed-scale cross-modal fusion method that widens the interaction between vision and language. Specially, at each stage, pyramid pooling is used to augment visual perception and improve the interaction between visual and linguistic features, thereby highlighting relevant regions in the visual data. Additionally, we employ a simple multi-scale feature fusion module to effectively combine multi-scale aligned features. Experiments conducted on Standard RIS benchmarks demonstrate that the proposed method achieves favorable performance against state-of-the- art approaches. Moreover, we conducted experiments on different visual backbones respectively, and the proposed method yielded better and significantly improved performance results.

Cross-view action recognition understanding from exocentric to egocentric perspective

Understanding action recognition in egocentric videos has emerged as a vital research topic with numerous practical applications. With the limitation in the scale of egocentric data collection, learning robust deep learning-based action recognition models remains difficult. Transferring knowledge learned from the large-scale exocentric data to the egocentric data is challenging due to the difference in videos across views. Our work introduces a novel cross-view learning approach to action recognition (CVAR) that effectively transfers knowledge from the exocentric to the selfish view. First, we present a novel geometric-based constraint into the self-attention mechanism in Transformer based on analyzing the camera positions between two views. Then, we propose a new cross-view self-attention loss learned on unpaired cross-view data to enforce the self-attention mechanism learning to transfer knowledge across views. Finally, to further improve the performance of our cross-view learning approach, we present the metrics to measure the correlations in videos and attention maps effectively. Experimental results on standard egocentric action recognition benchmarks, i.e., Charades-Ego, EPIC-Kitchens-55, and EPIC-Kitchens-100, have shown our approach’s effectiveness and state-of-the-art performance.

OpenHoldem: A Benchmark for Large-Scale Imperfect-Information Game Research.

Owing to the unremitting efforts from a few institutes, researchers have recently made significant progress in designing superhuman artificial intelligence (AI) in no-limit Texas hold'em (NLTH), the primary testbed for large-scale imperfect-information game research. However, it remains challenging for new researchers to study this problem since there are no standard benchmarks for comparing with existing methods, which hinders further developments in this research area. This work presents OpenHoldem, an integrated benchmark for large-scale imperfect-information game research using NLTH. OpenHoldem makes three main contributions to this research direction: 1) a standardized evaluation protocol for thoroughly evaluating different NLTH AIs; 2) four publicly available strong baselines for NLTH AI; and 3) an online testing platform with easy-to-use APIs for public NLTH AI evaluation. We will publicly release OpenHoldem and hope it facilitates further studies on the unsolved theoretical and computational issues in this area and cultivates crucial research problems like opponent modeling and human-computer interactive learning.

A context-aware zero trust-based hybrid approach to IoT-based self-driving vehicles security

With the speedy progression and adoption of IoT devices in modern self-driving vehicles (SDVs), autonomous driving industry is gradually reforming its capabilities to provide better transportation services. However, this domain faces enormous security and privacy challenges and thus has become an attractive target for attackers due to its rapid growth and market worth. Furthermore, the rapid transformation in technological tools in transport industry and speedy evolution of cyber-attacks paved the way for designing efficient IDSs. Motivated by these challenges, we put forward a new secure and efficient IDS approach for the security of SDVs. The propose approach utilizes an emerging strategy to mitigate security vulnerabilities and cyber attacks detection using zero trust (ZT) model. Through this work, we put forward a context-aware zero trust security framework for IoT-based SDVs. The proposed framework utilizes a context-aware design to evaluate the trustworthiness of the devices using multi-source trust and reputation model. Then, to make the framework more effective and reliable, we introduce crawler system into the context of the IoT-devices in SDVs to make the system unbiased. Additionally, an observer module is developed that employs state-of-the-art machine learning algorithm to detect malicious actions. Empirical results on two standard benchmark datasets (i.e., Car Hacking and ToN_IoT) validate the practicality and robustness of propose framework in real-world transport systems with enhanced security and trust management against evolving cyber-threats. Detection results demonstrate that the proposed framework secured the best performance by achieving 99.43% and 99.52% accuracy for Car Hacking and ToN_IoT, respectively. The findings of this study will help the security professionals and researchers to comprehend the importance of ZT architecture in developing effective and robust security solutions for modern IoT-based SDVs.

An accurate and locking-free geometric exact beam formulation on the special orthogonal group SO(3)

An accurate and locking-free geometric exact beam formulation (GEBF) on the special orthogonal group SO(3) is developed for slender beams with large deformations and large rotations. Due to the nonlinear nature of the spatial rotations, the classical GEBFs usually have a non-constant mass matrix and complex expressions of inertia forces; meanwhile, the singularity and locking issues are also key matters of concern. Aiming at resolving these drawbacks, two main contributions are achieved in the present work. First, the angular velocity is independently interpolated, resulting in a constant mass matrix and explicitly representable inertial forces, which can offer significant advantages for efficient dynamic simulations. Second, inspired by the assumed natural strain method in shell theory, the stretch-shear strain is interpolated to obtain simplified and locking-free elastic forces, which is a new attempt to alleviate the locking problems for the GEBFs. In addition, the special orthogonal group SO(3) is utilized for updating incremental rotation vectors to eliminate singularities, and objective strain measurement is achieved by employing relative rotation vector interpolation. The effectiveness and superiority of the developed low-order and high-order elements are demonstrated through numerical simulations of standard benchmark examples. The present work contributes to the advancement of accurate and reliable formulation for slender beams; the constant mass matrix, the locking-free characteristics, and the elegant form of the formulation make it particularly suitable for multibody dynamic analysis.

Minimizing the cobalt content in black ceramic pigments by Design of Experiments

This paper presents a comprehensive optimization strategy for the synthesis of new black ceramic pigments with low cobalt content while maintaining a single-phase spinel structure. The aim is to achieve comparable hues to those of an industrial benchmark containing five transition metals while minimizing the environmental impact. Exploring all possible compositions deriving from a five–component system (Cr, Mn, Fe, Co and Ni), through traditional methods would be extremely time consuming to guarantee an efficient sampling, requiring high experimental efforts. Hence, to identify the best black compositions, we employed a chemometric approach, the Design of Experiments, aiming to investigate the compositional domain derived by varying the metals stoichiometry within fixed boundaries to identify optimal pigment compositions. The resulting pigments (comprising Cr, Fe and Co), despite the lower cobalt content, exhibited optimal colorimetric properties comparable to the standard benchmark, with experimental ΔE values comparable to the ones predicted by the model. In particular, the pure spinels Cr1.05Fe1.05Co0.9O4 and Cr1.2Fe1.05Co0.75O4 displayed low lightness and chroma (L∗ = 2.4; C∗ = 5 and L∗ = 1.12; C∗ = 1.1, respectively), providing deep and dark black tonality. These compositions exhibit promising colorimetric performance and chemical stability, offering potential benefits for industrial applications.

Integrating machine learning with dynamic multi-objective optimization for real-time decision-making

Real-time decision-making in dynamic multi-objective optimization problems (DMOPs) is challenging due to constantly changing objectives and constraints. This paper integrates machine learning with Non-dominated Sorting Genetic Algorithm II (NSGA-II) to solve DMOPs and make real-time decisions. Learning-based methods have gained popularity for predicting solutions in new environments and capturing changing patterns in optimal solutions. However, existing approaches often struggle with training difficulty and reduced prediction accuracy due to irrelevant or redundant variables. Therefore, we introduce a new interdependent prediction (IDP) technique to identify correlations between variables and prediction targets and select significant variables for a predictive model. In this way, a better initial population is predicted. The IDP strategy is integrated within the dynamic NSGA-II, introducing a new algorithm called IDP-DNSGA-II. This integration facilitates rapid convergence, finding optimal or near-optimal solutions. The proposed method is evaluated against standard benchmarks, demonstrating superior performance in convergence speed and solution diversity with the changes in the problem environment. The IDP-DNSGA-II is validated through real-world optimization challenges in sustainable automobile production distribution in order-to-delivery systems to enhance environmental sustainability and operational efficiency. This study identifies the minimum frequency of change required in real-world problems to adequately track the optimal decision in real-time.

A study on ecological sustainable cities based on LCA-emergy-carbon footprint and geographic information system (GIS) approach

ABSTRACT Sustainable cities, as an effective response to global climate aberrations, have increasingly garnered the attention of researchers. However, due to the wide range of interpretations and applications associated with sustainable cities, diverse methodologies can yield starkly contrasting results. This paper adopts ecological and low-carbon perspectives to evaluate and discuss urban sustainability. The core of this investigation involves quantitatively measuring a city’s sustainability based on the lifecycle metrics of emergy and carbon footprint. The findings reveal that Yangzhou, recognized as a model ecological city by authorities, exhibits an emergy sustainability index below the standard benchmark (ESI = 0.02 < 1). GIS mapping reveals that the city’s largest ecological service system is situated in the northwest water region. Both the emergy and carbon footprint analyses indicate that urbanized areas exert a detrimental influence on the city’s overall sustainability. Additionally, the administrative authorities in Yangzhou have implemented proactive measures to transition the city towards sustainability. Currently, the impacts of clean energy adoption and urban wetland restoration are particularly pronounced. The application of lifecycle emergy and carbon footprint methods offers a holistic approach for urban managers, enabling sustainable urban planning that harmoniously integrates ecological emergy and low-carbon considerations. This approach is instrumental in the realization of sustainable cities.

Consistency Prototype Module and Motion Compensation for few-shot action recognition (CLIP-CPM[formula omitted]C)

Recently, few-shot action recognition has progressed significantly, as it has learned the feature discriminability and designed suitable comparison methods. Still, there are the following restrictions. (a) Previous works are mainly based on visual mono-modal. Although some multi-modal works use labels as supplementary to construct prototypes of support videos, they cannot use this information for query videos. The labels are not used efficiently. (b) Most of the works ignore the motion feature of video, although the motion features are essential for distinguishing. We proposed a Consistency Prototype and Motion Compensation Network(CLIP-CPM2C) based on CLIP to address these issues. Firstly, in order to make the amount of information between the prototype and the query more similar, we propose a novel method to compensate for the text(prompt) information of query videos when text(prompt) does not exist, which depends on a Consistency Loss. Secondly, we use the differential features of the adjacent frames in two directions as the motion features, which explicitly embeds the network with motion dynamics, and then apply the Consistency Loss to the motion features. Extensive experiments on standard benchmark datasets demonstrate that the proposed method can compete with state-of-the-art results. Our code is available at the URL: https://github.com/cofly2014/CPM-MC.git.

Global–Local Query-Support Cross-Attention for Few-Shot Semantic Segmentation

Few-shot semantic segmentation (FSS) models aim to segment unseen target objects in a query image with scarce annotated support samples. This challenging task requires an effective utilization of support information contained in the limited support set. However, the majority of existing FSS methods either compressed support features into several prototype vectors or constructed pixel-wise support-query correlations to guide the segmentation, which failed in effectively utilizing the support information from the global–local perspective. In this paper, we propose Global–Local Query-Support Cross-Attention (GLQSCA), where both global semantics and local details are exploited. Implemented with multi-head attention in a transformer architecture, GLQSCA treats every query pixel as a token, aggregates the segmentation label from the support mask values (weighted by the similarities with all foreground prototypes (global information)), and supports pixels (local information). Experiments show that our GLQSCA significantly surpasses state-of-the-art methods on the standard FSS benchmarks PASCAL-5i and COCO-20i.