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

Tousif Islam

We infer recoil (kick) velocities for all binary black hole merger events reported up to the GWTC--4 catalog, together with candidate intermediate-mass black hole events. We obtain informative kick constraints for GW231028\_153006 ($839^{+1018}_{-681}\,\mathrm{km\,s^{-1}}$) and GW231123\_135430 ($974^{+944}_{-760}\,\mathrm{km\,s^{-1}}$). Additionally, we compute recoil velocities for recently reported events from the ongoing fourth observing run: GW241011\_233834, GW241110\_124123, and GW250114\_082203, obtaining $v_{\rm kick} = 974^{+555}_{-466}\,\mathrm{km\,s^{-1}}$, $394^{+582}_{-207}\,\mathrm{km\,s^{-1}}$, and $115^{+301}_{-95}\,\mathrm{km\,s^{-1}}$, respectively. The remnant of GW241011\_233834 is therefore inferred to have one of the largest recoil velocities among currently known events. We find that present recoil kick constraints are driven primarily by measurements of the mass ratio and spin magnitudes, while the contribution from spin orientation angles remains subdominant in most cases. We estimate typical retention probabilities of the remnant black holes in GWTC catalogs to be $\sim 1$--$5\%$ for globular clusters, $\sim 15$--$30\%$ for nuclear star clusters, $\sim 5$--$40\%$ for dwarf galaxies, and $\sim 70$--$100\%$ for elliptical galaxies. We further show that, even for remnants retained in globular clusters, recoil-induced spatial displacements from the cluster core are often significant, which can substantially suppress the chances of hierarchical mergers. We find that the probability for a GWTC merger remnant to participate in hierarchical mergers is $\sim 0.1$--$1\%$ in globular clusters and $\sim 1$--$15\%$ in nuclear star clusters.

Avni Mittal, Rauno Arike

Large language models (LLMs) are increasingly used as judges of chain-of-thought (CoT) reasoning, but it remains unclear whether they can reliably assess process faithfulness rather than just answer plausibility. We introduce C2-Faith, a benchmark built from PRM800K that targets two complementary dimensions of faithfulness: causality (does each step logically follow from prior context?) and coverage (are essential intermediate inferences present?). Using controlled perturbations, we create examples with known causal error positions by replacing a single step with an acausal variant, and with controlled coverage deletions at varying deletion rates (scored against reference labels). We evaluate three frontier judges under three tasks: binary causal detection, causal step localization, and coverage scoring. The results show that model rankings depend strongly on task framing, with no single judge dominating all settings; all judges exhibit a substantial gap between detecting an error and localizing it; and coverage judgments are systematically inflated for incomplete reasoning. These findings clarify when LLM judges are dependable and where they fail, and provide practical guidance for selecting judges in process-level evaluation

Süleyman Şahin

Ziynet kavramı, Kur’ân-ı Kerîm’de hem maddi hem de manevi güzelliklerin bir ifadesi olarak ele alınırken fıkıh disiplini içinde daha çok maddi süslenme aracı olan ürünler için kullanılmıştır. Altın ve gümüş, tarih boyunca ziynet eşyalarının temel maddeleri olmuş, ancak demir, bakır ve diğer madenlerin kullanımı fıkhî açıdan tartışmalı bir konu olarak kabul edilmiştir. Altın ve gümüş dışındaki madenlerin ziynet olarak kullanımına dair fıkıh kaynaklarında sınırlı bilgiler bulunmakta, bu konu genellikle yüzük kullanımı özelinde ele alınmaktadır. Hadislerde ve mezheplerin yorumlarında bu madenlerin kullanımına dair farklı görüşler mevcuttur. Dolayısıyla makale, demir ve bakır gibi madenlerle kıymetli taşların ziynet amaçlı kullanımını incelemektedir. Çalışmada öncelikle, altın ve gümüş dışındaki madenlerden ziynet edinmeyi yasaklayan ve cevazına imkân tanıyan rivayetler ele alınmış, ardından fakihlerin konu hakkındaki görüşleri değerlendirilmiştir. Erkeklerin altın kullanımı çoğunluğun görüşüne göre haram kabul edilirken, gümüş yüzük takmalarına da izin verilmiştir. Demir ve bakır gibi madenlerden yapılan ziynet eşyaları ise, Hz. Peygamber’in bu tür eşyaları müşriklere veya putlara benzetmesi nedeniyle hem kadınlar hem de erkekler için mekruh veya haram olarak nitelendirilmiştir. Ancak demir yüzüğün mehir olarak kabul edildiğine dair hadisten hareketle bazı fıkhî yorumlar, bu tür madenlerin kullanımına cevaz veren bir yaklaşımın da mevcut olduğunu göstermektedir. Hanefî, Mâlikî ve Hanbelî mezhepleri bu tür ziynetleri mekruh görürken, Şâfiî mezhebi caiz kabul etmektedir. Rivayetlerin demir ve bakırı putların yapı malzemesi ve müşriklerin takısı olarak nitelemesi, bu hususların hükmün illeti olduğu yönünde bazı yorumlara zemin hazırlamıştır. Bu bağlamda modern çağda hükmün illetinin anlamını yitirmiş olabileceği düşünülmektedir. Çünkü günümüz toplumlarında bu tür ürünlerin şirk unsurlarına veya müşriklere benzeme gibi olgularla ilişkilendirilmemesi ve ilgili illeti barındırmaması gerekçesiyle kullanımının caiz olduğu ifade edilebilir. Günümüzde çelik, platin gibi madenlerden yapılan imitasyon takıların yaygınlaşması, bu tür ürünlerin fıkhî hükmünün yeniden ele alınmasını gerektirmiştir. Benzer bir şekilde literatürde kıymetli taşların ziynet olarak kullanımı da tartışma konusu edilmiştir. Genel anlamda kıymetli taşların cevaz yönü ön planda olmakla birlikte bazı Hanefî alimler demir ve bakıra, yine bazı Mâlikî alimler ise altın ve gümüşten edinilen kap ve kacaklara kıyasla olumsuz kanaat belirtmişlerdir. Sonuç olarak, altın ve gümüş dışındaki madenlerin ziynet amacıyla kullanımına dair fıkhî hükümler, rivayetlerin tarihsel bağlamı, hükmün illetleri ve günümüz şartları dikkate alınarak yeniden değerlendirilmeye tabi tutulmuştur. Neticede fıkıh disiplininin temel prensipleri ışığında, bu tür madenlerin ziynet olarak kullanılmasının, illeti ortadan kalkan yasakların bağlayıcılığını yitirdiği ve örfün de olumlu bir rol oynadığı durumlarda caiz olacağı kanaatine ulaşılmıştır.

Rara Puspa Kirana, Budi Prabowo

Increased intense business competition encourages all industries from various sectors to create human resource management strategies to improve performance as a foundation for competing against competitors. The implementation of effective human resource management strategies can affect the improvement of employee performance so that it has an impact on organizational performance. The purpose of this research is to understand and analyze the effect of job stress, leadership and compensation on employee performance. In this research, the intended population is all employees of PT Bahana Prestasi Surabaya, using a saturated sampling technique of 54 employees as a sample. The method applied is associative research with a quantitative approach, and uses Multiple Linear Regression analysis techniques. The research results obtained simultaneously the variables of work stress, leadership, and compensation have a significant relationship to employee performance. Partial testing of work stress and leadership variables has a significant relationship to employee performance, while the compensation variable has an insignificant relationship to employee performance.

Mujawazah

This research study aims to explore the management of fun and engaging English learning for students at MI Al Ma’had An Nur, a pesantren-based (Islamic boarding school) elementary school. Considering the unique educational environment that integrates formal education with Islamic values, the research focuses on strategies that align English language learning with the students' cultural and religious context. Through qualitative methods, including classroom observation, teacher interviews, and student feedback, this research indicates effective practices that increase motivation, participation, and language retention. The result shows that games, storytelling, songs, and interactive group activities significantly enhance students' enthusiasm and comprehension. In addition, linking Islamic values to English learning materials can make the learners more comfortable and involved in the classroom process. This research may provide guidance for teachers in similar contexts to develop joyful and appropriate English learning experiences.

Dominikus Rendy Anggara, Irfan Mahdi

Human resources are a vital element in carrying out organizational functions, including in government environments. This research is motivated by fluctuations in discipline and performance of Civil Servants (PNS) at the Office of Cooperatives, Micro Enterprises, and Trade of Pontianak City, which indicates the need to strengthen intrinsic motivation and self-efficacy as internal factors that play a role in employee normative commitment. The main problem in this study is whether there is an influence of intrinsic motivation and self-efficacy on the normative commitment of Civil Servants in the agency. The purpose of this study is to determine the effect of two independent variables, namely intrinsic motivation and self-efficacy, on the dependent variable, namely normative commitment. This study uses an associative quantitative method with a survey approach by distributing questionnaires to 55 Civil Servants as respondents. Data analysis was carried out using multiple linear regression with the help of SPSS software. The results of the study indicate that intrinsic motivation and self-efficacy have a positive and significant effect on normative commitment. Partially, intrinsic motivation has a significant effect with a coefficient of 0.365, while self-efficacy has a greater effect with a coefficient of 0.875. The coefficient of determination (R²) of 0.502 indicates that the two independent variables contribute 50.2% to the normative commitment variable. Thus, increasing employee motivation and self-confidence can strengthen their normative commitment in carrying out organizational duties and responsibilities.

Xiaqiang Tang, Yi Wang, Keyu Hu et al.

Retrieval-Augmented Generation (RAG) systems require Large Language Models (LLMs) to generate responses that are faithful to the retrieved context. However, faithfulness hallucination remains a critical challenge, as existing methods often require costly supervision and post-training or significant inference burdens. To overcome these limitations, we introduce Self-Supervised Faithfulness Optimization (SSFO), the first self-supervised alignment approach for enhancing RAG faithfulness. SSFO constructs preference data pairs by contrasting the model's outputs generated with and without the context. Leveraging Direct Preference Optimization (DPO), SSFO aligns model faithfulness without incurring labeling costs or additional inference burden. We theoretically and empirically demonstrate that SSFO leverages a benign form of \emph{likelihood displacement}, transferring probability mass from parametric-based tokens to context-aligned tokens. Based on this insight, we propose a modified DPO loss function to encourage likelihood displacement. Comprehensive evaluations show that SSFO significantly outperforms existing methods, achieving state-of-the-art faithfulness on multiple context-based question-answering datasets. Notably, SSFO exhibits strong generalization, improving cross-lingual faithfulness and preserving general instruction-following capabilities. We release our code and model at the anonymous link: https://github.com/chkwy/SSFO

Gabrielle Kaili-May Liu, Gal Yona, Avi Caciularu et al.

A critical component in the trustworthiness of LLMs is reliable uncertainty communication, yet LLMs often use assertive language when conveying false claims, leading to over-reliance and eroded trust. We present the first systematic study of $\textit{faithful confidence calibration}$ of LLMs, benchmarking models' ability to use linguistic expressions of uncertainty that $\textit{faithfully reflect}$ their intrinsic uncertainty, across a comprehensive array of models, datasets, and prompting strategies. Our results demonstrate that LLMs largely fail at this task, and that existing interventions are insufficient: standard prompt approaches provide only marginal gains, and existing, factuality-based calibration techniques can even harm faithful calibration. To address this critical gap, we introduce MetaFaith, a novel prompt-based calibration approach inspired by human metacognition. We show that MetaFaith robustly improves faithful calibration across diverse models and task domains, enabling up to 61% improvement in faithfulness and achieving an 83% win rate over original generations as judged by humans.

Muhammad Abu Ahmad, Mohamad Ballout, Raia Abu Ahmad et al.

This paper presents our submission to the QIAS 2025 shared task on Islamic knowledge understanding and reasoning. We developed a hybrid retrieval-augmented generation (RAG) system that combines sparse and dense retrieval methods with cross-encoder reranking to improve large language model (LLM) performance. Our three-stage pipeline incorporates BM25 for initial retrieval, a dense embedding retrieval model for semantic matching, and cross-encoder reranking for precise content retrieval. We evaluate our approach on both subtasks using two LLMs, Fanar and Mistral, demonstrating that the proposed RAG pipeline enhances performance across both, with accuracy improvements up to 25%, depending on the task and model configuration. Our best configuration is achieved with Fanar, yielding accuracy scores of 45% in Subtask 1 and 80% in Subtask 2.

Sarikul Islam, Nitin Gupta

This paper introduces a novel framework to construct the probability density function (PDF) of non-negative continuous random variables. The proposed framework uses two functions: one is the survival function (SF) of a non-negative continuous random variable, and the other is a weight function, which is an increasing and differentiable function satisfying some properties. The resulting random variable is referred to as the weighted tail random variable (WTRV) corresponding to the given random variable and the weight function. We investigate several reliability properties of the WTRV and establish various stochastic orderings between a random variable and its WTRV, as well as between two WTRVs. Using this framework, we construct a WTRV of the Kumaraswamy distribution. We conduct goodness-of-fit tests for two real-world datasets, applied to the Kumaraswamy distribution and its corresponding WTRV. The test results indicate that the WTRV offers a superior fit compared to the Kumaraswamy distribution, which demonstrates the utility of the proposed framework.

Dino Sapta Prasetyo, Faqih Utsman Adzikri, Geri Firmansah

This research aims to find out how much influence product design and product variations have partially and simultaneously. This research is quantitative, using associative quantitative data methods. This research uses primary data, namely a questionnaire distributed to 50 Kiddorck Arjawinangun consumers using a non-probability sampling technique with incidental sampling. Quantitative data was analyzed using validity tests, reliability tests, classical assumption tests, regression analysis, and hypothesis tests processed using the IBM Statistics SPSS 26 application. The results of testing product design variables partially had a positive and significant effect on purchasing decisions with a significance value of 0.003 < 0.05 and the calculated t value > t table or 3.106 > 1.677. The partial product variation variable also has a positive and significant effect on purchasing decisions with a significance value of 0.013 < 0.05 and a calculated t value > t table or 2.575 > 1.677. Simultaneously, product design variables and product variations have an influence of 41.8% on purchasing decisions. This is shown by the calculated F value of 18.584 which is greater than the F table of 3.191 and a significance value of 0.000 < 0.05.

Erik Arakelyan, Pasquale Minervini, Pat Verga et al.

Modern Question Answering (QA) and Reasoning approaches based on Large Language Models (LLMs) commonly use prompting techniques, such as Chain-of-Thought (CoT), assuming the resulting generation will have a more granular exploration and reasoning over the question space and scope. However, such methods struggle with generating outputs that are faithful to the intermediate chain of reasoning produced by the model. On the other end of the spectrum, neuro-symbolic methods such as Faithful CoT (F-CoT) propose to combine LLMs with external symbolic solvers. While such approaches boast a high degree of faithfulness, they usually require a model trained for code generation and struggle with tasks that are ambiguous or hard to formalise strictly. We introduce $\textbf{F}$aithful $\textbf{L}$ogic-$\textbf{A}$ided $\textbf{R}$easoning and $\textbf{E}$xploration ($\textbf{FLARE}$), a novel interpretable approach for traversing the problem space using task decompositions. We use the LLM to plan a solution, soft-formalise the query into facts and predicates using a logic programming code and simulate that code execution using an exhaustive multi-hop search over the defined space. Our method allows us to compute the faithfulness of the reasoning process w.r.t. the generated code and analyse the steps of the multi-hop search without relying on external solvers. Our methods achieve SOTA results on $\mathbf{7}$ out of $\mathbf{9}$ diverse reasoning benchmarks. We also show that model faithfulness positively correlates with overall performance and further demonstrate that $\textbf{FLARE}$ allows pinpointing the decisive factors sufficient for and leading to the correct answer with optimal reasoning during the multi-hop search.

Sarikul Islam, Nitin Gupta

In this paper, we studied the stochastic ordering behavior of series as well as parallel systems' lifetimes comprising dependent and heterogeneous components, experiencing random shocks, and exhibiting distinct dependency structures. We establish certain conditions on the lifetime of individual components where the dependency among components defined by Archimedean copulas, and the impact of random shocks on the overall system lifetime to get the results. We consider components whose survival functions are either increasing log-concave or decreasing log-convex functions of the parameters involved. These conditions make it possible to compare the lifetimes of two systems using the usual stochastic order framework. Additionally, we provide examples and graphical representations to elucidate our theoretical findings.

Niful Islam

Artificial intelligence is currently a dominant force in shaping various aspects of the world. Machine learning is a sub-field in artificial intelligence. Feature scaling is one of the data pre-processing techniques that improves the performance of machine learning algorithms. The traditional feature scaling techniques are unsupervised where they do not have influence of the dependent variable in the scaling process. In this paper, we have presented a novel feature scaling technique named DTization that employs decision tree and robust scaler for supervised feature scaling. The proposed method utilizes decision tree to measure the feature importance and based on the importance, different features get scaled differently with the robust scaler algorithm. The proposed method has been extensively evaluated on ten classification and regression datasets on various evaluation matrices and the results show a noteworthy performance improvement compared to the traditional feature scaling methods.

Ni Made Ayu Sukma Dewi, I Made Ardana, Phill I Gusti Putu Sudiarta

Hasil belajar matematika peserta didik di SMK Negeri 5 Denpasar saat ini masih tergolong rendah, salah satu penyebabnya adalah kurangnya minat dan motivasi belajar peserta didik. Ada peserta didik yang memiliki gaya belajar visual (belajar melalui indera penglihatannya), gaya belajar auditori (belajar melalui indera pendengaran) dan gaya belajar kinestetik (belajar melalui pengalaman langsung). Ketiga jenis media tersebut disiapkan secara bersamaan untuk satu Kompetensi Dasar yaitu peluang suatu kejadian, sehingga peserta didik dapat memilih sendiri media pembelajaran yang cocok dengan gaya belajar masing-masing. Kualitas media pembelajaran yang dikembangkan diukur berdasarkan validitas, kepraktisan dan efektifitasnya dalam meningkatkan hasil belajar baik dalam ranah kognitif (pengetahuan), ranah afektif (aktivitas dan motivasi belajar) serta ranah psikomotorik (keterampilan). Memenuhi syarat validitas isi berarti media pembelajaran telah sesuai dengan prosedur pengembangan Plomp dan karakteristik pembelajaran berdiferensiasi. Untuk validitas konstruk dinilai oleh dua orang pakar media pembelajaran dengan kategori sangat valid karena skor rata-rata berada pada rentang  Untuk tingkat kepraktisan penggunaan media pembelajaran berdasarkan angket respon peserta didik dan pendidik memiliki kategori praktis karena skor rata-rata berada pada rentang . Tingkat efektifitas media dalam ranah kognitif yaitu 67,25 dengan kategori sangat baik, ranah afektif dengan rata-rata skor aktivitas peserta didik 3,51 dengan kategori sangat aktif dan ranah psikomotorik ketuntasan klasikal peserta didik sebesar 90%.

Muhamad Sakaki, Muh. Zaki, Abdul Malik

This study aims to determine the strategy of PAI teachers in managing students' HOTS-based critical thinking skills to deal with a flood of information and hoaxes at SMK Plus Nurul Hakim Kediri. This study uses a qualitative approach with a descriptive type of research. This research was conducted at SMK Plus Nurul Hakim Kediri for two months, the subjects of this study were PAI teachers at SMK Plus Nurul Hakim Kediri, research informants were school principals, teachers and students. Data collection techniques used in this study were observation, unstructured interviews and documentation. To analyze data with three steps: condensing data, presenting data, and drawing conclusions. The results of the study show that (1) PAI teachers can implement HOTS-based critical thinking in the form of teachers participating in Subject Teacher Deliberation training, compiling HOTS-based curriculum 13 learning tools, preparing students' souls before learning, training students' critical power, training students' metacognitive skills, training reflective thinking students, train students' psychomotor domains, PAI teachers are admired and emulated, analyze appropriate and accurate information and activate correct thinking. (2) The PAI teacher's strategy in managing HOTS-based students' critical thinking to deal with a flood of information and hoaxes is: problem-based learning strategies, creative and innovative learning strategies, contextual teaching learning strategies and inquiry strategies. (3) The results of the implementation of the learning process carried out to manage HOTS-based students' critical thinking abilities to deal with a flood of information and hoaxes are managing critical thinking power so that students are able to choose valid information, are able to train kinesthetic compliance, increase student activity, teacher strategies in diverse learning, and the achievement of learning objectives.

Esma Merve Acar, Yunus Araz

İstanbul Sözleşmesi’nin asıl adı ‘Kadınlara Yönelik Şiddet ve Aile İçi Şiddetin Önlenmesi ve Bunlarla Mücadeleye İlişkin Avrupa Konseyi Sözleşmesi’dir. 11 Mayıs 2011'de İstanbul'da imzaya açılmış olması nedeniyle kısaca "İstanbul Sözleşmesi” denmektedir. Türkiye, sözleşmeye imza atan ilk devletlerden biri olmuştur. Sözleşmenin duyulması ve incelenmesi sonucunda hem hukukî hem de dinî ve kültürel açıdan bazı olumsuzluklar ihtiva ettiği görülmüştür. Bu nedenle 20 Mart 2021 tarihinde Resmî Gazetede yayımlanan 3718 sayılı cumhurbaşkanı kararıyla ve Sözleşme’nin 80. maddesinin müsaade ettiği şekilde, Türkiye sözleşmeden çekilme kararı almıştır. Sözleşme konusu ve amacı itibariyle tüm insanlar için önem arz etmektedir. Özellikle şiddet ve kadına karşı şiddeti önlemek bütün insanların görevidir. Toplumun en küçük birimi olan ailedeki eğitimden, devleti yönetenlerin çıkaracağı kanunlara kadar bütün katmanlarda önemsenmesi gereken ve üzerinde durulması gereken bir konudur. Bu bağlamda şiddetin giderilmesi için hem ulusal hem de uluslararası alanlarda çalışmalar yapılmalıdır. Fakat bu çalışmalar yapılırken toplumların değer yargıları göz önüne alınmalı ve alınan kararların doğuracağı olumlu ve olumsuz durumlar iyi incelenmelidir. Sözleşme imzalandığı günden beri tartışma konusu haline gelmiştir. Her ne kadar sözleşmeden çekilmeyi eleştirenler varsa da sözleşmenin mevcut haliyle Türk aile yapısına, dinî ve kültürel değerlerine uymayan maddeler ihtiva ettiği ile ilgili ciddi çalışmalar yapılmıştır. Yapılan çalışmalarda sözleşmenin bazı olumlu taraflarına değinilmekle birlikte sakıncalar ve olumsuz sonuçları ile ilgili değerlendirmeler yapılmıştır. Sözleşme, İslam hukuku açısından birkaç çalışmaya konu olmuş ve farklı açılardan değerlendirilmiştir. Bu çalışmamızda da sözleşme İslam dini ve İslam hukuku açısından ele alınarak değerlendirilecek ve görülen birçok soruna değinilecektir. Bu çalışma giriş ve iki bölümden oluşmaktadır. Birinci bölümde sözleşmede geçen bazı kavramlar kısaca incelenmiş, ikinci bölümde ise yedi başlık altında sözleşmedeki maddeler ele alınarak İslam hukuku açısından değerlendirilmiştir. İstanbul Sözleşmesi ile kadına karşı şiddetin önlenmesi hedeflenmekle birlikte sözleşme içeriği incelendiğinde bazı sakıncalar görülmektedir. Bu sakıncaların başında aile kavramının yok sayılması, nikâh sonucu olmayan birlikteliklerin meşru görülmesi, eşcinselliğin normalleştirilmesi, küçük yaşlardan itibaren cinsiyetsizleştirmenin zemininin hazırlanması gibi konular bulunmaktadır. Ayrıca 18 yaş altı çocukların evlilik dışı ilişkilerinin suç sayılmamasının yanında 18 yaş altı rızaya dayalı evliliklerin suç sayılması gibi durumlar da bu bağlamda düşünülebilir. Sözleşme eşcinsellerle ilgili doğrudan bir madde içermese de satır aralarında cinsiyetsizliğe çokça atıflar bulunmaktadır. Aile ve nikâh kavramları neredeyse hiç önemsenmemekte hatta nikâhsız birlikteliklere meşruiyet arz edecek tutumlar görülmektedir. Sözleşmede toplumların değer yargıları bazen açıkça bazen ise üstü kapalı bir şekilde eleştirilmekte hatta küçümsenmektedir. Bütün bu sorunlar çerçevesinde sözleşmenin özelde İslam hukuku genelde İslam dini açısından sıkıntılar barındırdığı aşikârdır. Hatta sözleşme bu haliyle sadece İslam dininin değil diğer semavi dinlerin de değerleri ve hükümleri ile çelişkiler arz etmektedir. Yapılması gereken bu sözleşme vesilesiyle hukukî, dinî, sosyal ve psikolojik boyutlarıyla, şiddetin sadece kadın özelinde değil bütün boyutlarıyla ele alınarak incelenmesidir. Sonuç olarak öncelikle ahlakî eğitimle, ekonomik düzenlemelerle ve kanun çalışmalarıyla şiddet, topumda en aza indirilmeye hatta yok edilmeye çalışılmalıdır. Bu yapılırken kadın ve erkeğin birbirinin rakibi değil tamamlayıcısı olduğu, ailenin kadınların özgürlüğünü engelleyen bir kurum değil de toplumu sulh ve selamete kavuşturacak bir kurum olduğu bilinmelidir. Ayrıca kadın ve erkek cinsiyeti dışında bir cinsiyetin olmadığını ve eşcinselliğin toplumsal, psikolojik sebeplere dayalı bir rahatsızlık olduğunu ve giderilmesi için çalışmalar yapılması gerektiğini göz önüne almak gerekir. Sözleşmeye dayalı olarak bazı feminist ve LGBTİ taraftarı grupların İslam’ın kadına değer vermediği veya kadını aşağıladığı ile ilgili görüşler de ayrıca üzerinde durulması gereken bir konudur. Oysaki Cahiliyye toplumunda kadının durumu ve İslam geldikten sonra koyduğu ilkeler ele alındığında kadına verilen değerin ne kadar yüksek olduğu açıkça görülecektir. Müslümanlar arasında uygulamalarda bazı sorunların olması İslam dinine yüklenemez. Bu ayrım iyi yapılmalı ve konular birbirine karıştırılmamalıdır.

Yang Zhang, Yawei Li, Hannah Brown et al.

Feature attribution explains neural network outputs by identifying relevant input features. The attribution has to be faithful, meaning that the attributed features must mirror the input features that influence the output. One recent trend to test faithfulness is to fit a model on designed data with known relevant features and then compare attributions with ground truth input features.This idea assumes that the model learns to use all and only these designed features, for which there is no guarantee. In this paper, we solve this issue by designing the network and manually setting its weights, along with designing data. The setup, AttributionLab, serves as a sanity check for faithfulness: If an attribution method is not faithful in a controlled environment, it can be unreliable in the wild. The environment is also a laboratory for controlled experiments by which we can analyze attribution methods and suggest improvements.

Minhajul Islam

By doing a small $c$ (speed of light) expansion of $SU(N)$ Yang-Mills fields, we construct two different electric and two different magnetic sectors actions of Carrollian Yang-Mills theory. For both electric and magnetic cases, one sector contains non-trivial self-interaction, and another is $N^2-1$ copies of respective sector Carrollian abelian theory. In $d=4$ , all the four sectors are invariant under infinite Carrollian Conformal symmetry. There are no central extensions when analyzing charge algebra at the phase space level. Lastly, we compute propagators for all four sectors and vertices for two non-trivial sectors. Propagators in position space show ultra-local behavior.

Kung-Hsiang Huang, Hou Pong Chan, Heng Ji

Faithfully correcting factual errors is critical for maintaining the integrity of textual knowledge bases and preventing hallucinations in sequence-to-sequence models. Drawing on humans' ability to identify and correct factual errors, we present a zero-shot framework that formulates questions about input claims, looks for correct answers in the given evidence, and assesses the faithfulness of each correction based on its consistency with the evidence. Our zero-shot framework outperforms fully-supervised approaches, as demonstrated by experiments on the FEVER and SciFact datasets, where our outputs are shown to be more faithful. More importantly, the decomposability nature of our framework inherently provides interpretability. Additionally, to reveal the most suitable metrics for evaluating factual error corrections, we analyze the correlation between commonly used metrics with human judgments in terms of three different dimensions regarding intelligibility and faithfulness.