Hasil untuk "Islam. Bahai Faith. Theosophy, etc."

Tunazzina Islam

Large language models (LLMs) are increasingly capable of generating personalized, persuasive text at scale, raising new questions about bias and fairness in automated communication. This paper presents the first systematic analysis of how LLMs behave when tasked with demographic-conditioned targeted messaging. We introduce a controlled evaluation framework using three leading models -- GPT-4o, Llama-3.3, and Mistral-Large 2.1 -- across two generation settings: Standalone Generation, which isolates intrinsic demographic effects, and Context-Rich Generation, which incorporates thematic and regional context to emulate realistic targeting. We evaluate generated messages along three dimensions: lexical content, language style, and persuasive framing. We instantiate this framework on climate communication and find consistent age- and gender-based asymmetries across models: male- and youth-targeted messages emphasize agency, innovation, and assertiveness, while female- and senior-targeted messages stress warmth, care, and tradition. Contextual prompts systematically amplify these disparities, with persuasion scores significantly higher for messages tailored to younger or male audiences. Our findings demonstrate how demographic stereotypes can surface and intensify in LLM-generated targeted communication, underscoring the need for bias-aware generation pipelines and transparent auditing frameworks that explicitly account for demographic conditioning in socially sensitive applications.

Chengcheng Feng, Haojie Yin, Yucheng Jin et al.

Comic understanding presents a significant challenge for Multimodal Large Language Models (MLLMs), as the intended meaning of a comic often emerges from the joint interpretation of visual, textual, and social cues. This naturally motivates Chain-of-Thought (CoT) prompting, since explicit intermediate reasoning appears promising for integrating such heterogeneous signals. However, existing CoT methods are poorly matched to this structure: they tend to force interpretation into a single reasoning path before multiple cues have been jointly considered, often degrading performance, especially for small MLLMs. Our key idea is to explicitly preserve multi-cue interpretation during supervision construction, rather than collapsing comic understanding into a single reasoning chain. To this end, we propose a two-stage framework for faithful comic reasoning in small MLLMs. First, we introduce MoCoT, a modular supervision construction framework that preserves multi-cue interpretation and turns it into more faithful supervision. Second, we propose VERA, a structured reward mechanism that turns such supervision into faithful reasoning behavior by aligning optimization with both reasoning faithfulness and answer correctness. Extensive experiments on five benchmarks spanning comic understanding and broader humor-centric and abstract visual reasoning tasks demonstrate that our framework achieves strong results in the $\leq$ 4B regime, surpasses several 7B baselines, improves four small MLLMs by an average of $\mathbf{12.1%}$ as a plug-in, and consistently enhances reasoning faithfulness while preserving inference efficiency.

Rafiul Islam, Md. Taimur Ahad

This study presents a novel transfer learning approach and data augmentation technique for mental stability classification using human voice signals and addresses the challenges associated with limited data availability. Convolutional neural networks (CNNs) have been employed to analyse spectrogram images generated from voice recordings. Three CNN architectures, VGG16, InceptionV3, and DenseNet121, were evaluated across three experimental phases: training on non-augmented data, augmented data, and transfer learning. This proposed transfer learning approach involves pre-training models on the augmented dataset and fine-tuning them on the non-augmented dataset while ensuring strict data separation to prevent data leakage. The results demonstrate significant improvements in classification performance compared to the baseline approach. Among three CNN architectures, DenseNet121 achieved the highest accuracy of 94% and an AUC score of 99% using the proposed transfer learning approach. This finding highlights the effectiveness of combining data augmentation and transfer learning to enhance CNN-based classification of mental stability using voice spectrograms, offering a promising non-invasive tool for mental health diagnostics.

Syahbudi

On September 22, 2021, Cinta Laura delivered a commencement speech at the Launching Night for Religious Moderation Action held by the Ministry of Religious Affairs. Cinta Laura is known as a young female artist and grew up in an environment of Western education. Her attendance conveys a significant message about what and who the main target of the religious moderation policy is. This article employed a critical discourse analysis method to examine religious moderation through the lens of the hegemonic state policy. Secondary data are obtained from books, articles, opinions, and news in both print and electronic mass media that explain the concepts and ideas of religious moderation, especially since 2020. The question is, how is the policy of religious moderation constructed today? What is the purpose of the policy? The article argues that the state promotes religious moderation to strengthen the legitimacy of the new nationalism. This policy aims to nationalize religious views to influence how people view the relationship between religion and the state. It is accomplished, in part, by implementing policies that promote the debunking of narratives of religious intolerance, especially in educational institutions. Additionally, the state promotes socio-cultural norms that are perceived as beneficial to the state in terms of maintaining stability. The critical contribution of this article is that, through religious moderation policies, the state has managed to mitigate the political impact of intolerance; however, this approach tends to be formalistic and ceremonial.

Zujing Liu, Junwen Pan, Qi She et al.

Recent large vision-language models (LVLMs) can generate vision-text multimodal chain-of-thought (MCoT) traces after reinforcement fine-tuning (RFT). However, we observe that the visual information incorporated in MCoT is often inaccurate, though still yield correct answers, indicating a lack of faithfulness in the MCoT reasoning process. We attribute this unfaithfulness to the RL reward in RFT, which solely incentivizes the format of interleaved vision-text cues, ie, it encourages the model to incorporate visual information into its text reasoning steps without considering the correctness of the visual information. In this paper, we first probe the faithfulness of MCoT by measuring how much the prediction changes when its visual and textual thoughts are intervened. Surprisingly, the model's predictions remain nearly unchanged under visual intervention but change significantly under textual intervention, indicating that the visual evidence is largely ignored. To further analyze visual information, we introduce an automated LVLM-based evaluation metric that quantifies the faithfulness of visual cues from two perspectives: reliability and sufficiency. Our evaluation reveals that the visual information in current MCoT traces is simultaneously unreliable and insufficient. To address this issue, we propose a novel MCoT learning strategy termed Sufficient-Component Cause Model (SCCM) learning. This approach encourages the MCoT to generate sufficient yet minimal visual components that are independently capable of leading to correct answers. We note that the proposed SCCM is annotation-free and compatible with various RFT for MCoT in a plug-and-play manner. Empirical results demonstrate that SCCM consistently improves the visual faithfulness across a suite of fine-grained perception and reasoning benchmarks. Code is available at https://github.com/EugeneLiu01/Faithful_Thinking_with_Image.

Fahadul Islam, Sunil Dhar

A theoretical study was conducted on the impact of electron and positron impact ionization of excited hydrogen atoms that were in the 3s state; this study was conducted within the First-Born Approximation (FBA), which provides an analytical expression for the transition matrix in terms of the Bethe-Lewis Integral Formalism. This formalism utilized both Coulomb continuum and confluent hypergeometric functions to describe the scattering states involved. Single Differential Cross Sections (SDCS) were calculated for incident energies of 100, 150, 200, and 250 eV. The data obtained indicated a peak in the ionization rates approximately at 200 eV, with the ionization rate decreasing as the incident energy increased further. The diffuse radial nature of the 3s wave function is shown to increase the sensitivity of the ionization dynamics to the incident particle energy. Asymmetries in charge were also detected; specifically, at low energy of the ejected electron, the SDCS values for positrons were greater than the corresponding values for electrons; however, as the energy of the incident particles was increased, these differences disappeared, thereby demonstrating the applicability of the FBA at high energy limits. The residual differences at low energy were due to the omission of exchange and post-collision interactions from the model. The results of this work can be used as benchmarking for the development of more complex distorted wave and multi-scattering theories in excited state ionization processes.

Fahadul Islam, Sunil Dhar

Calculations have been made for the double differential cross section (DDCS) for the ionization of metastable hydrogen atoms in the 3S state by electron and positron impact at energies of 150 eV and 250 eV. The authors implemented the second Born approximation to the multiple scattering theory as their model, evaluated the corresponding analytical expressions using Bethe and Lewis integrals, and numerically computed them using MATLAB. The generated DDCS captures the features of both recoil and binary fragmentation and, at the same time, provides an overall qualitative consistency with earlier studies, although some differences can be found between them at certain emission angles. The present work, therefore, supplies new theoretical reference levels for ionization investigations in hydrogen-like systems, now that no experimental data are available for the metastable 3S state.

Ali akbar Hoseyni ramandi, Mohammad Mahdi Firoozmehr

The discussion about the Moral Absolutism and Moral relativity begins with this question: do moral rules change with the change of conditions, time, place, people and societies, or are they always fixed and without any restriction and allocation, including all people, in all times and places?Some schools consider moral rules to be a relative matter, whose rules will change with the change of conditions, time, place, people and societies; but a number of schools consider moral laws to be fixed, eternal and independent of time, place, conditions and people.The present research is an exploration about the application and moral relativity in the scope of the Quranic revelation. Quranic teachings shows that the moral system of Islam is based on fixed and absolute laws and rulings, and numerous Quranic reasons and evidences support the Moral Absolutism principles and do not reflect moral relativism.

R. Agtiani Putri, Murtadlo Murtadlo, Wagino Wagino

Vocational skills as learning to provide provisions that are in accordance with students' abilities. The purpose of this study is to test the influence of the direct instruction model to improve students' vocational skills in SLB-C AKW IISurabaya. This study uses a quantitative approach using a pre-experimental design with a one group pretest-posttest design. The data collection technique uses tests. The results in this study showed that the average pretest score of 37.00 increased and the posttest result was 70.00. The results of data analysis show Zh 2.80 and Zt 1.96, then the interpretation is Ha > Ho. So, Ha is accepted, which means that it is evident that there is an influence of the Direct instruction learning model mediated by learning videos to improve vocational.

Ahmad Nilnal Munachifdlil Ula, Hanik Hidayati

This study aims to explore the dynamics of social harmony in the perspective of Islam in contemporary society, considering the influence of global contexts and technological developments. A deep understanding of the religious values of Islam, active involvement in religious activities, and contributions to social activities are the focus of the study. The research method uses a qualitative approach by collecting data through religious literature studies, analyzing government regulations related to social harmony, and direct observation of the social dynamics of contemporary society. The data is then analyzed qualitatively to identify patterns in maintaining social harmony. The results show that understanding religious teachings is the main foundation for maintaining harmony, while active involvement in religious activities and contributions to social activities play a crucial role in forming harmonious social relationships. Challenges related to differences in religious understanding and the impact of global contexts and technological developments highlight the complexity of maintaining social harmony. As a recommendation, this study emphasizes the need for inclusive religious education, promotion of active involvement in religious activities, and strengthening social activities as key steps. Strategies to adapt to global and technological changes are also identified as important steps to overcome these challenges. The conclusion of this study is expected to provide guidance for stakeholders in dealing with the complex dynamics of maintaining social harmony in the perspective of Islam in contemporary society that is constantly changing.

Tousif Islam

Utilizing publicly available non-spinning eccentric binary black hole (BBH) merger simulations (\href{https://data.black-holes.org/waveforms/catalog.html}{https://data.black-holes.org/waveforms/catalog.html}) from the SXS collaboration~\cite{Hinder:2017sxy}, we present convincing evidence that the waveform phenomenology in eccentric BBH mergers is significantly simpler than previously thought. We find that the eccentric modulations in the amplitudes, phases, and frequencies in different spherical harmonic modes are all related and can be modeled using a single time series modulation. Using this universal eccentric modulation, we provide a model named \texttt{gwNRHME} to seamlessly convert a multi-modal (i.e with several spherical harmonic modes) quasi-circular waveform into multi-modal eccentric waveform if the quadrupolar eccentric waveform is known. This reduces the modelling complexity of eccentric BBH mergers drastically as we now have to model only a single eccentric modulation time-series instead of modelling the effect of eccentricity in all modes. When compared with the NR simulations, our model mismatches are mostly $\sim 10^{-3}$ and are comparable to the numerical errors in the NR simulations. Our method is modular and can be readily added to any quadrupolar non-spinning eccentric waveform model. We make our model publicly available through the \texttt{gwModels} (\href{https://github.com/tousifislam/gwModels}{https://github.com/tousifislam/gwModels}) waveform package.

Nazmul Islam, Seokjoo Shin

Deep learning (DL) has revolutionized wireless communication systems by introducing datadriven end-to-end (E2E) learning, where the physical layer (PHY) is transformed into DL architectures to achieve peak optimization. Leveraging DL for E2E optimization in PHY significantly enhances its adaptability and performance in complex wireless environments, meeting the demands of advanced network systems such as 5G and beyond. Furthermore, this evolution of data-driven PHY optimization has also enabled advanced semantic applications across various modalities, including text, image, audio, video, and multimodal transmissions. These applications elevate communication from bit-level to semantic-level intelligence, making it capable of discerning context and intent. Although the PHY, as a DL architecture, plays a crucial role in enabling semantic communication (SemCom) systems, comprehensive studies that integrate both E2E communication and SemCom systems remain significantly underexplored. This highlights the novelty and potential of these integrative fields, marking them as a promising research domain. Therefore, this article provides a comprehensive review of the emerging field of data-driven PHY for E2E communication systems, emphasizing their role in enabling semantic applications across various modalities. It also identifies key challenges and potential research directions, serving as a crucial guide for future advancements in DL for E2E communication and SemCom systems.

Afreen Islam, Guido Herrmann, Joaquin Carrasco

In this paper, the solvability of the Inverse Optimal Control (IOC) problem based on two existing minimum principal methods, is analysed. The aim of this work is to answer the question regarding what kinds of trajectories, that is depending on the initial conditions of the closed-loop system and system dynamics, of the original optimal control problem, will result in the recovery of the true weights of the reward function for both the soft and the hard-constrained methods [1], [2]. Analytical conditions are provided which allow to verify if a trajectory is sufficiently conditioned, that is, holds sufficient information to recover the true weights of an optimal control problem. It was found that the open-loop system of the original optimal problem has a stronger influence on the solvability of the Inverse Optimal Control problem for the hard-constrained method as compared to the soft-constrained method. These analytical results were validated via simulation.

Pipit Muliyah, Desi Wijayanti Ma'rufah, Mustangin Mustangin

Reading comprehension is a fundamental skill in mastering English. Relying solely on classroom instruction is insufficient for providing students with a comprehensive understanding of English texts. As a result, students must actively cultivate their self-directed learning abilities. It  enables them to prioritize and focus on the areas they need to improve. Unfortunately, developing independent learning skills can be challenging without encouragement from teachers and an understanding of what they are learning. Thankfully, technology, such as language learning applications, can help overcome this challenge. This study examines using the Quora language learning application to enhance students' reading comprehension skills in English outside the classroom. Using qualitative research, this study involved 18 economic sharia students taking English for Business subject, and data was collected through a questionnaire and interviews. The findings revealed that Quora helped students improve their reading comprehension skills and supported their independent learning by allowing them to study English anywhere and anytime. As a result, students could take control of their learning and significantly improve their English reading skills.

Achmad Fawaid, Wening Udasmoro, Sri Margana et al.

The lack of research on literary elements in "babad" is in line with the limited study of Islamic elements within this genre. This research aims to analyze the Javanese poets' depictions of Islam in "Babad Tanah Jawi" and "Babad Giyanti." The study operates under the assumption that "babad" as literary works can allegorically create a contested space where Islam interacts with other traditions. A qualitative paradigm with a comparative structural approach is employed in this study, focusing on the narratives that shape the author's understanding of Islam in the 18th century. The findings of this research reveal that the court poet's portrayal of Islam in "Babad Tanah Jawi" shares similarities with another work in the same genre, "Babad Giyanti." These commonalities encompass the utilization of Islamic verses in the "bhabuka" text and "asmaul husna," the depiction of "walisongo's" role, the practice of "tawajjuh," moral Islamic teachings, and the political positioning characterized by ambivalence and negotiation among different traditions.

Alessandro Kurniawan Ulung, Clemontin Cornelia Monica Jannah

This study discusses the thoughts of renowned Sufi poet Jalal Al-din Rumi behind Rumi Diplomacy, rolled out by Fadjroel Rachman, Indonesian Ambassador to Kazakhstan and Tajikistan. Rumi Diplomacy takes place in a way that the ambassador uses Rumi’s works to enhance bilateral ties with Tajikistan. Rumi Diplomacy implies that the ambassador considers Rumi’s thoughts soft power in his foreign policy. This study uses a qualitative method, with the theory of foreign-policy analysis by Harold and Margaret Sprout in place. This theory enables the author to find that Rumi’s thoughts on religious moderation, universal love, and universal faith influenced Rachman’s psychological environment to perceive such thoughts as soft power to reach Indonesia’s national interests in Tajikistan. It means that he considers the development of Rumi’s influence in Indonesia as the operational environment in making his foreign policy. To identify and understand his psychological and operational environments, the author interviewed the ambassador, read Rumi’s works, and reviewed the literature. This study, which is the first to research Rumi Diplomacy, contributes to shedding light that in international politics, literature work should not be overlooked because it can also affect the decision-making process and help a country reach national interests without coercion.

Mijanur Islam, Saurabh Basu

We conduct a thorough study of the persistent currents in a spin-orbit coupled $α-T_3$ pseudospin-1 fermionic quantum ring (QR) that smoothly interpolates between graphene ($α = 0$, pseudospin-1/2) and a dice lattice ($α = 1$, pseudospin-1). In particular, we have considered both intrinsic spin-orbit coupling (ISOC) and Rashba spin-orbit coupling (RSOC) in addition to an external magnetic field, and have systematically enumerated their individual and combined effects on the charge, valley and the spin-polarized persistent currents. The energy levels of the system comprise of the conduction bands, valence bands, and flat bands which show non-monotonic dependencies on the ring radius, R of the QR, in the sense that, for small R, the energy levels vary as 1/R, while the variation is linear in R for large R. The cases corresponding to zero magnetic fields are benchmarked with those for finite external fields. Further, it is noted that the flat bands demonstrate dispersive behavior, and hence can contribute to the transport properties only when ISOC is non-zero. Moreover, the RSOC yields spin-split bands, thereby contributing to the spin-resolved currents, together with distinct degeneracies for different spin branches. The persistent currents in the charge, valley, and spin sectors for each of these cases oscillate as a function of the magnetic field with a period equal to the flux quantum, as they should be, and depend upon the spin-orbit coupling terms. Further, we have explored the role played by the parameter $α$ in our entire analysis to ascertain the effect of the flat bands.

Minhajul Islam

We explore the realization of BRST symmetry in the non-Lorentzian Yang-Mills Lagrangian within the context of Galilean and Carrollian Yang-Mills theory. Firstly we demonstrate the nilpotent property of classical BRST transformations and construct corresponding conserve charges for both cases. Then we analyze the algebra of these charges and observe the nilpotent properties at the algebraic level. The findings of this study contribute to a deeper understanding of BRST symmetry in non-Lorentzian Yang-Mills Lagrangians and provide insights into the algebraic properties of related conserve charges.

Rahnuma Islam, Akif Ibragimov

We study the movement of the living organism in a band form towards the presence of chemical substrates based on a system of partial differential evolution equations. We incorporate Einstein's method of Brownian motion to deduce the chemotactic model exhibiting a traveling band. It is the first time that Einstein's method has been used to motivate equations describing the mutual interaction of the chemotactic system. We have shown that in the presence of limited and unlimited substrate, traveling bands are achievable and it has been explained accordingly. We also study the stability of the constant steady states for the system. The linearized system about a constant steady state is obtained under the mixed Dirichlet and Neumann boundary conditions. We are able to find explicit conditions for linear instability. The linear stability is established with respect to the L-2 norm, H1-norm, and L-infinity norm under certain conditions.

Mobarakol Islam, Ben Glocker

Deep convolutional neural networks have shown remarkable performance on various computer vision tasks, and yet, they are susceptible to picking up spurious correlations from the training signal. So called `shortcuts' can occur during learning, for example, when there are specific frequencies present in the image data that correlate with the output predictions. Both high and low frequencies can be characteristic of the underlying noise distribution caused by the image acquisition rather than in relation to the task-relevant information about the image content. Models that learn features related to this characteristic noise will not generalize well to new data. In this work, we propose a simple yet effective training strategy, Frequency Dropout, to prevent convolutional neural networks from learning frequency-specific imaging features. We employ randomized filtering of feature maps during training which acts as a feature-level regularization. In this study, we consider common image processing filters such as Gaussian smoothing, Laplacian of Gaussian, and Gabor filtering. Our training strategy is model-agnostic and can be used for any computer vision task. We demonstrate the effectiveness of Frequency Dropout on a range of popular architectures and multiple tasks including image classification, domain adaptation, and semantic segmentation using both computer vision and medical imaging datasets. Our results suggest that the proposed approach does not only improve predictive accuracy but also improves robustness against domain shift.