Abstract This article examines the internal contradictions and social pathologies generated by the modern paradigm, focusing especially on the issue of migration. Using epistemological critiques from thinkers like Adorno, Kuhn, Popper, Hayek, and the Frankfurt School, the paper argues that modernity’s promise of universal rationality and scientific progress has frequently resulted in structures that are exclusionary, homogenizing, and sometimes even totalitarian. The paper then links these theoretical debates to contemporary migration. It emphasizes how refugee women—especially those facing the combined challenges of gender and displacement—experience complex layers of social invisibility and discursive erasure. By critically applying recognition theory and discourse analysis, the study highlights how modernity’s promise of inclusion frequently hides the actual mechanisms of marginalization. In this part, the article demonstrates that these marginalization processes are linked to the scientific premises of the modern paradigm and considers the migration problem as an example of the pathology of the modern paradigm.
History of scholarship and learning. The humanities, Social Sciences
Abstract To achieve sustainable ecological development in the contemporary global economy and technology, addressing carbon emissions is imperative. The issue of carbon emissions and their impact on the environment has been the subject of intense scrutiny and debate among academicians worldwide. This investigation investigates the Yellow River Basin in China, a region with substantial industrial development. It analyses the fluctuations in carbon emissions and their influencing factors from 2010 to 2022 using nighttime light data. The spatial clustering features of carbon emissions in the prefecture-level communities of the Yellow River Basin from 2005 to 2022 are verified by the spatial autocorrelation analysis. The Gini coefficient is employed to examine regional disparities in carbon emissions in three distinct ways: total difference, intra-regional difference, and inter-regional difference. Ultimately, the GTWR model is implemented to evaluate the variables that influence carbon emissions within the Yellow River Basin. The results suggest that the Yellow River Basin is characterized by substantial spatial clustering. Shandong Province and Lvliang City are home to high-high clustering cities, while Gansu Province, Shaanxi Province, and Sichuan Province are home to low-low clustering cities. Carbon emissions are increasing annually. In comparison to the Upper, Middle, and Lower Yellow River regions, the disparities in carbon emissions between the Middle and Lower Yellow River regions are somewhat lesser. Intra-regional differences follow the trend of Upper Yellow River > Middle Yellow River > Lower Yellow River. Economic development, industrial structure, scientific advancement, and education level consistently positively impact carbon emissions in the Yellow River Basin. However, financial development has a sustained inhibiting effect on carbon emissions, and infrastructure development initially promotes but eventually inhibits carbon emissions.
History of scholarship and learning. The humanities, Social Sciences
Abstract Shajara-i Tarākima, transcribed in Classical Chagatai Turkic by Abu al-Ghazi Bahadur Khan during the 17th century, is one of the most significant works in Turkology studies. Since the 19th century, it has garnered substantial interest among Turkology researchers from various nations, becoming the subject of numerous scholarly investigations. However, these studies have primarily focused on critiquing and revising the previous researchers’ works, overlooking the characteristics of spoken language present in the text. Nevertheless, considering that the fundamental basis of the work lies in oral narratives, it is evident that the text incorporates abundant features of colloquial language. Therefore, a pragmatic evaluation of the Shajara-i Tarākima is crucial to better comprehending and interpreting the work. This study analyzes the discourse markers in the Shajara-i Tarākima to achieve this purpose. The research builds upon the new model proposed by Crible and Degand (2019), which identifies discourse markers based on their combination of domain and function. The study asserts that in intralingual or interlingual translations of historical texts containing features of colloquial language, discourse markers play a vital role in ensuring the accurate transfer of meaning.
History of scholarship and learning. The humanities, Social Sciences
Minghui Sun, Matthew M. Engelhard, Benjamin A. Goldstein
Risk assessments for a pediatric population are often conducted across multiple stages. For example, clinicians may evaluate risks prenatally, at birth, and during Well-Child visits. Although predictions made at later stages typically achieve higher precision, it is clinically desirable to make reliable risk assessments as early as possible. Therefore, this study focuses on improving prediction performance in early-stage risk assessments. Our solution, \textbf{Borrowing From the Future (BFF)}, is a contrastive multi-modal framework that treats each time window as a distinct modality. In BFF, a model is trained on all available data throughout the time while performing a risk assessment using up-to-date information. This contrastive framework allows the model to ``borrow'' informative signals from later stages (e.g., Well-Child visits) to implicitly supervise the learning at earlier stages (e.g., prenatal/birth stages). We validate BFF on two real-world pediatric outcome prediction tasks, demonstrating consistent improvements in early risk assessments. The code is available at https://github.com/scotsun/bff.
While contrastive multi-view clustering has achieved remarkable success, it implicitly assumes balanced class distribution. However, real-world multi-view data primarily exhibits class imbalance distribution. Consequently, existing methods suffer performance degradation due to their inability to perceive and model such imbalance. To address this challenge, we present the first systematic study of imbalanced multi-view clustering, focusing on two fundamental problems: i. perceiving class imbalance distribution, and ii. mitigating representation degradation of minority samples. We propose PROTOCOL, a novel PaRtial Optimal TranspOrt-enhanced COntrastive Learning framework for imbalanced multi-view clustering. First, for class imbalance perception, we map multi-view features into a consensus space and reformulate the imbalanced clustering as a partial optimal transport (POT) problem, augmented with progressive mass constraints and weighted KL divergence for class distributions. Second, we develop a POT-enhanced class-rebalanced contrastive learning at both feature and class levels, incorporating logit adjustment and class-sensitive learning to enhance minority sample representations. Extensive experiments demonstrate that PROTOCOL significantly improves clustering performance on imbalanced multi-view data, filling a critical research gap in this field.
Jacopo Graldi, Alessandro Breccia, Giulia Lanzillotta
et al.
Despite recent efforts, neural networks still struggle to learn in non-stationary environments, and our understanding of catastrophic forgetting (CF) is far from complete. In this work, we perform a systematic study on the impact of model scale and the degree of feature learning in continual learning. We reconcile existing contradictory observations on scale in the literature, by differentiating between lazy and rich training regimes through a variable parameterization of the architecture. We show that increasing model width is only beneficial when it reduces the amount of feature learning, yielding more laziness. Using the framework of dynamical mean field theory, we then study the infinite width dynamics of the model in the feature learning regime and characterize CF, extending prior theoretical results limited to the lazy regime. We study the intricate relationship between feature learning, task non-stationarity, and forgetting, finding that high feature learning is only beneficial with highly similar tasks. We identify a transition modulated by task similarity where the model exits an effectively lazy regime with low forgetting to enter a rich regime with significant forgetting. Finally, our findings reveal that neural networks achieve optimal performance at a critical level of feature learning, which depends on task non-stationarity and transfers across model scales. This work provides a unified perspective on the role of scale and feature learning in continual learning.
Abstract Issues of Social Justice, broadly conceived, are increasingly being included as a component in digital humanities (DH) scholarship or are the reason d’etre of the research itself. Equally, issues of ethics, privacy, and copyright are taking on greater prominence in DH scholarship. This article focuses on the creation of a course for the #dariahTeach platform, Social Justice in the DH. This case-based course features a variety of scholarship, methods, theories, and communities and practices from around the globe that do research at the intersections of social justice and the DH. The article problematizes some of the issues and challenges of working within a social justice framework, highlights existing best practice, and explores issues of ethics, privacy, and copyright within the projects which differ, depending, not only on legal, but on moral frameworks. The additional duty of care when developing and disseminating content created by or about marginalised, indigenous, or minority populations, and/or those whose ethics and traditions differ from dominant western values is also highlighted. And lastly, the importance of taking these factors into account when conducting research in reparative ways, particularly in the context of contested histories and/or material cultures is considered.
Abstract We describe an efficient pipeline for morpho-syntactically annotating an ancient language corpus which takes advantage of bootstrapping techniques. This pipeline is designed for ancient language scholars looking to jump-start their own treebank projects, which can in turn serve further pedagogical research projects in the target language. We situate our work in the field of similar ancient language treebank projects, arguing that our approach shows that individual humanities scholars can leverage current machine-learning tools to produce their own richly annotated corpora. We illustrate this pipeline by producing a new Akkadian-language treebank based on two volumes from the online editions of the State Archives of Assyria project hosted on Oracc, as well as a spaCy language model named AkkParser trained on that treebank. Both of these are made publicly available for annotating other Akkadian corpora. In addition, we discuss linguistic issues particular to the Neo-Assyrian letter corpus and data-encoding complications of cuneiform texts in Oracc. The strategies, language models, and processing scripts we developed to handle both linguistic and data-encoding issues in this project will be of special interest to scholars seeking to develop their own cuneiform treebanks.
Ethnography (social scientific methods that illuminate how people understand, navigate and shape the real world contexts in which they live their lives) and machine learning (computational techniques that use big data and statistical learning models to perform quantifiable tasks) are each core to contemporary social science. Yet these tools have remained largely separate in practice. This chapter draws on a growing body of scholarship that argues that ethnography and machine learning can be usefully combined, particularly for large comparative studies. Specifically, this paper (a) explains the value (and challenges) of using machine learning alongside qualitative field research for certain types of projects, (b) discusses recent methodological trends to this effect, (c) provides examples that illustrate workflow drawn from several large projects, and (d) concludes with a roadmap for enabling productive coevolution of field methods and machine learning.
Nuraeni Nuraeni, Selly Rahmawati*, Rian Nurizka
et al.
The aim of this research is to determine the application of the problem-based learning model in PPKn subjects for class XI students at SMKN 2 Ciamis and how it can be improved. The research method used in this research is Kemmis and Taggart's Classroom Action Research model with planning, action, observation and reflection steps. The data collection used is tests and observations. The analytical technique used for analysis is descriptive statistics. The results of this research are that the application of the Problem Based Learning model can improve student learning outcomes in class XI citizenship education subjects from the research results. This can be seen from the level of completeness of student learning outcomes in the Iknown cycle at pretest of 41.66% and posttest of 66.6% and an increase in cycle II preetest of 81.6% and posttest of 83.3%. So the level of completeness of student learning outcomes from cycle I and cycle II has increased by 16.7%.
The OPTIMICE project, which stands for optimising machine translation of metadata and its integration into the editorial chain, aims at devising a method – transferrable to other journals and disciplinary fields – that combines neural machine translation (DeepL) and human post-editing to improve the quality of article metadata (abstracts, keywords, etc.) from French to English in the editorial process of journals. A team of translation researchers who are also translators worked on four journals edited by the Presses universitaires de Rennes (PUR), in collaboration with the editorial comittees and the MSHB (Maison des sciences de l’Homme en Bretagne). The translation of the paper metadata by their authors and by machine translation was comparatively assessed. A survey on translation practices among researchers in HSS was led, and recommendations for writing and translating metadata were formulated for the organized integration of the methodology within the editorial process.
History of scholarship and learning. The humanities, Bibliography. Library science. Information resources
Abstract The Regional Comprehensive Economic Partnership (RCEP) is an agreement that transformed the world economy and entered into force in January 2022 with the participation of fifteen nations. In the study, the visualisation analysis was 301 articles in Web of Science (WoS) on the subjects of “RCEP,” or “The Regional Comprehensive Economic Partnership,” from January 2012 to January 2023, using CiteSpace. The results of a comparative analysis of the number of journals co-citation and keyword co-occurrence indicate that further studies of “RCEP” will not be limited to the scope of traditional economics, but more and further fields are waiting for scholars to develop.
History of scholarship and learning. The humanities, Social Sciences
The present paper, adopting some of the tools offered by Cognitive Linguistics, namely the mechanisms of conceptual metaphor and metonymy, is a qualitative study of a sociolinguistic nature. Its overall purpose is an attempt at exhibiting a paradigm shift in the representation of women in Anglo-American proverbs. Combining the potential of the cross-fertilisation between Cognitive Linguistics and paremiological studies, the study appertains to the sense-threads embedded in the figurative language of proverbs, with the main focus on a cognitive semantic analysis of selected Anglo-American paremias directed towards women and animals. The main goal of the research is the juxtaposition of the meaning coded in two proverbs of traditional status, as representatives of a larger group of paremiological units (i.e. A woman, a cat, and a chimney should never leave the house; A whistling girl and a crowing hen always come to no good end), reflecting the deep-rooted gender-biased ideology in patriarchal Anglo-American society, with the content of the selected anti-proverb (i.e. The early bird gets up to serve his wife breakfast in bed) and a contemporary proverb (i.e. A woman without a man is like a fish without a bicycle), serving as sample evidence of the heralds of a paradigm shift in the issue of gender stereotyping stored in paremiological wisdom. The paper shows that the motivation behind the use of the analysed proverbs is to be accounted for by reference to the mechanism of metaphor-metonymy interaction, while the rise of new gender-related proverbs can be regarded as a sign of socio-cultural changes. Specifically, through the medium of modern paremiology, asymmetrical representation of male and female gender, coupled with traditional masculine and feminine characteristics as well as social roles, appears if not endangered then, at least, to be taking a promising path.
History of scholarship and learning. The humanities
Toon Vanderschueren, Alicia Curth, Wouter Verbeke
et al.
Machine learning (ML) holds great potential for accurately forecasting treatment outcomes over time, which could ultimately enable the adoption of more individualized treatment strategies in many practical applications. However, a significant challenge that has been largely overlooked by the ML literature on this topic is the presence of informative sampling in observational data. When instances are observed irregularly over time, sampling times are typically not random, but rather informative -- depending on the instance's characteristics, past outcomes, and administered treatments. In this work, we formalize informative sampling as a covariate shift problem and show that it can prohibit accurate estimation of treatment outcomes if not properly accounted for. To overcome this challenge, we present a general framework for learning treatment outcomes in the presence of informative sampling using inverse intensity-weighting, and propose a novel method, TESAR-CDE, that instantiates this framework using Neural CDEs. Using a simulation environment based on a clinical use case, we demonstrate the effectiveness of our approach in learning under informative sampling.
Many countries and regions have reached a consensus to promote science, technology, engineering, and mathematics (STEM) education in the past decade. A body of studies have demonstrated that the design and organization of interdisciplinary teaching activities are important for the effective implementation of STEM education. So far, however, little attention has been paid to the taxonomy of teaching models in STEM education. This paper aims to propose a taxonomy based on the dimensions of learning outcomes (i.e., product-oriented and knowledge-oriented) and teaching process (i.e., forward teaching model and reverse teaching model) in STEM education. Through the intersection of the above two dimensions, four teaching models in STEM education have emerged, including project-based learning (PBL), reverse engineering (RE), scientific inquiry (SI), and troubleshooting/debugging (T/D). In addition, four cases are introduced to explain how these four teaching models operate in STEM education. The implications of this work for future research are also discussed. It is promising that the study will be valuable to enrich the current research, and shed light on the theory and practice of STEM education.
History of scholarship and learning. The humanities, Social Sciences
Nika Haghtalab, Yanjun Han, Abhishek Shetty
et al.
In this paper, we study oracle-efficient algorithms for beyond worst-case analysis of online learning. We focus on two settings. First, the smoothed analysis setting of [RST11,HRS22] where an adversary is constrained to generating samples from distributions whose density is upper bounded by $1/σ$ times the uniform density. Second, the setting of $K$-hint transductive learning, where the learner is given access to $K$ hints per time step that are guaranteed to include the true instance. We give the first known oracle-efficient algorithms for both settings that depend only on the pseudo (or VC) dimension of the class and parameters $σ$ and $K$ that capture the power of the adversary. In particular, we achieve oracle-efficient regret bounds of $ \widetilde{O} ( \sqrt{T dσ^{-1}} ) $ and $ \widetilde{O} ( \sqrt{T dK} ) $ for learning real-valued functions and $ O ( \sqrt{T dσ^{-\frac{1}{2}} } )$ for learning binary-valued functions. For the smoothed analysis setting, our results give the first oracle-efficient algorithm for online learning with smoothed adversaries [HRS22]. This contrasts the computational separation between online learning with worst-case adversaries and offline learning established by [HK16]. Our algorithms also achieve improved bounds for worst-case setting with small domains. In particular, we give an oracle-efficient algorithm with regret of $O ( \sqrt{T(d |\mathcal{X}|)^{1/2} })$, which is a refinement of the earlier $O ( \sqrt{T|\mathcal{X}|})$ bound by [DS16].
This research uses satellite remote sensing data to measure the urban spatial form and analyzes the impact of changes to urban spatial structure on the income gap between urban and rural residents. The results indicate that the compactness of the urban spatial form is positively correlated with the income gap between urban and rural residents. However, there is no statistically significant relationship between the urban spatial extension rate and the urban–rural income gap. A subsequent analysis of the control variables shows that fiscal policy is positively correlated while urbanization is negatively correlated with the income gap between urban and rural residents. These conclusions provide the basis for formulating policies to narrow the urban–rural income gap. Appropriately reducing the spatial compactness of cities can narrow the income gap. In addition, changing excessive preferences for urban fiscal policy and increasing the level of urbanization can also promote a reduction in the income gap between urban and rural residents.
History of scholarship and learning. The humanities, Social Sciences
Santosh Vijaykumar, Yan Jin, Daniel Rogerson
et al.
Abstract We examined how age and exposure to different types of COVID-19 (mis)information affect misinformation beliefs, perceived credibility of the message and intention-to-share it on WhatsApp. Through two mixed-design online experiments in the UK and Brazil (total N = 1454) we first randomly exposed adult WhatsApp users to full misinformation, partial misinformation, or full truth about the therapeutic powers of garlic to cure COVID-19. We then exposed all participants to corrective information from the World Health Organisation debunking this claim. We found stronger misinformation beliefs among younger adults (18–54) in both the UK and Brazil and possible backfire effects of corrective information among older adults (55+) in the UK. Corrective information from the WHO was effective in enhancing perceived credibility and intention-to-share of accurate information across all groups in both countries. Our findings call for evidence-based infodemic interventions by health agencies, with greater engagement of younger adults in pandemic misinformation management efforts.
History of scholarship and learning. The humanities, Social Sciences
Machine learning systems often do not share the same inductive biases as humans and, as a result, extrapolate or generalize in ways that are inconsistent with our expectations. The trade-off between exemplar- and rule-based generalization has been studied extensively in cognitive psychology; in this work, we present a protocol inspired by these experimental approaches to probe the inductive biases that control this tradeoff in category-learning systems. We isolate two such inductive biases: feature-level bias (differences in which features are more readily learned) and exemplar or rule bias (differences in how these learned features are used for generalization). We find that standard neural network models are feature-biased and exemplar-based, and discuss the implications of these findings for machine learning research on systematic generalization, fairness, and data augmentation.