Hasil untuk "Unlocalized maps (Asian studies only)"

Alexander Berndt, Vekil Bekmyradov, Rainer Gemulla et al.

Flaky tests yield inconsistent results when they are repeatedly executed on the same code revision. They interfere with automated quality assurance of code changes and hinder efficient software testing. Previous work evaluated approaches to train machine learning models to classify flaky tests based on identifiers in the test code. However, the resulting classifiers have been shown to lack generalizability, hindering their applicability in practical environments. Recently, pre-trained Large Language Models (LLMs) have shown the capability to generalize across various tasks. Thus, they represent a promising approach to address the generalizability problem of previous approaches. In this study, we evaluated three LLMs (two general-purpose models, one code-specific model) using three prompting techniques on two benchmark datasets from prior studies on flaky test classification. Furthermore, we manually investigated 50 samples from the given datasets to determine whether classifying flaky tests based only on test code is feasible for humans. Our findings indicate that LLMs struggle to classify flaky tests given only the test code. The results of our best prompt-model combination were only marginally better than random guessing. In our manual analysis, we found that the test code does not necessarily contain sufficient information for a flakiness classification. Our findings motivate future work to evaluate LLMs for flakiness classification with additional context, for example, using retrieval-augmented generation or agentic AI.

Gumaro Rendon, Rutuja Kshirsagar, Quoc Hoan Tran

In this work, we present a quantum algorithm designed to solve the differential equation used in the pricing of Asian options, in the framework of the Black-Scholes model. Our approach modifies an existing quantum pre-conditioning method (different from classical methods) for the problem of Asian option pricing such that we remove the dependence on the original condition number of discretized differential equation (system of linear equations). This was possible with new fast-forwardable discretizations of the first and second derivatives with respect to the underlying asset value ratio (value over average). We determine that these discretizations handle well kinks in the initial/terminal conditions. We also introduce a new circuit construction for the discretized time-derivative operator with Dirichlet boundary conditions which avoids the oracle workspace needed for the general sparse matrix implementation. Here, we also devised a new method probability integral estimation from which we extract the solution, achieving $\tilde{O}({\rm polylog}\left(1/ε)\right)$, which is an exponential improvement over other quantum methods when it comes to solution information extraction from the solution state.

Uvini Balasuriya Mudiyanselage, Bharat Jayprakash, Kookjin Lee et al.

Large language models (LLMs), such as ChatGPT and Claude, have emerged as essential tools for information retrieval, often serving as alternatives to traditional search engines. However, ensuring that these models provide accurate and equitable information tailored to diverse demographic groups remains an important challenge. This study investigates the capability of LLMs to retrieve disaggregated health-related information for sub-ethnic groups within the Asian American population, such as Korean and Chinese communities. Data disaggregation has been a critical practice in health research to address inequities, making it an ideal domain for evaluating representation equity in LLM outputs. We apply a suite of statistical and machine learning tools to assess whether LLMs deliver appropriately disaggregated and equitable information. By focusing on Asian American sub-ethnic groups, a highly diverse population often aggregated in traditional analyses; we highlight how LLMs handle complex disparities in health data. Our findings contribute to ongoing discussions about responsible AI, particularly in ensuring data equity in the outputs of LLM-based systems.

Charles L. Samuels

A 2009 article of Allcock and Vaaler explored of the $\mathbb Q$-vector space $\mathcal G := \overline{\mathbb Q}^\times/{\overline{\mathbb Q}^\times_{\mathrm{tors}}}$, showing how to represent it as part of a function space on the places of $\overline{\mathbb Q}$. Several years later, the author began attempts to examine dual spaces related to $\mathcal G$ in an effort to obtain Riesz-type representation theorems. Those results required the construction of an object called a {\it consistent map}. We study a natural Galois action on consistent maps and establish when consistent maps are invariant under this action. Our results generalize earlier work of the author regarding rational valued consistent maps over non-Archimedean places of $\mathbb Q$.

Dagmar Hellmann-Rajanayagam

Sara Keller

In monsoon-dependent South Asia, harvesting and managing water is a necessity. Thus, monuments and modified landscapes related to irrigation, water storing and water worship represent a large share of Indian architectural heritage. Yet, this rich corpus and its attached hydrological knowledge awaits full recognition: thus far, documentation and studies focus on the most visible elements, especially stepwells, tanks and other dug facilities. Hydro-structures are generally considered in an isolated manner and lack contextualisation. Following the observations of English chaplain Edward Terry (1590–1660), this paper aims to explore the relationship between the medieval Indian city and the presence of monumental, often religiously connoted, hydraulic constructions. Based on archaeological and historical data, the paper proposes a fresh look at the hydraulic elements in the larger context of the city. The main argument rehabilitates the artificial lake and underlines its crucial function in plain regions that depend on a sufficient refilling of aquifers. In Western India, the typical city of the Solanki and later the Vaghela and the Muzaffarid dynasties had access to a broad variety of water sources capable of meeting the needs of multiple religious and secular activities (hydro-diversity). This generous waterscape and its multifaceted developments reflect on functions of sociability and religiosity in the city. Water here appears as an essential identity marker of urbanity.

Faiza Muhammad Din

In Pakistan, female entrepreneurs and earning women encounter considerable challenges when trying to enter male-dominated job markets. Many turn to women-only social media groups as a forum to discuss these obstacles and share strategies for overcoming them. Unlike conventional, more bureaucratic networking avenues such as chambers of commerce, these online platforms offer flexibility and continuous connectivity. This proves especially beneficial for women during significant life events like maternity or caregiving breaks. In Muslim-majority countries, these groups also serve as platforms for exchanging culturally and religiously relevant business practices, including discussions on halal investments and Sharia-compliant entrepreneurship. This study employs online ethnography and interviews with key stakeholders in some of Pakistan’s most prominent women-focused online networking groups. While one might assume that the primary aim of such professional networks is to discuss business growth, cutting-edge research or technology, these groups actually thrive on a common language, shared challenges and shared values. The article begins by exploring the concept of professionalism and the role of community and networking in career advancement. It then delves into the specific challenges that make networking difficult for women in Pakistan but paradoxically act as a glue binding them together in these online spaces.

Dissa Paputungan-Engelhardt

Sher Muhammad

Emi Goto

This research note is based on work in progress relating to the relevance of translation and semantic framing for the public perception of newly introduced concepts. The article traces trajectories of public perception of the concept “sexual harassment” in different countries. It sheds light on the role played by the translation words that were chosen to introduce the concept into societies where it was not yet established as a term denoting a violation of women’s rights and a serious offence. The term “sexual harassment” was coined in the United States during the 1970s and was subsequently adopted in various parts of the world. Two societies in which the term was adopted in transliteration and translation are Japan and Egypt. In the case of Japan, the term was introduced during the 1980s, yielding the transliterated Japanese loanword sekushuaru harasumento, later abbreviated to seku-hara. It became a buzzword, yet with a less serious nuance than the English term originally aroused. Egypt adopted the term in the 2000s, translated as taḥarrush jinsī in Arabic. The translation word taḥarrush jinsī was strategically selected and subsequently framed so as to denote a serious offence. Examined through an approach known as the “cassette effect”, the comparison of the two cases suggests that the process of framing the meaning of terms and concepts by social movements and advocacy groups is highly significant for their public perception.

Muhammad Ibrahim, Naveed Akhtar, Saeed Anwar et al.

Localization is a fundamental task in robotics for autonomous navigation. Existing localization methods rely on a single input data modality or train several computational models to process different modalities. This leads to stringent computational requirements and sub-optimal results that fail to capitalize on the complementary information in other data streams. This paper proposes UnLoc, a novel unified neural modeling approach for localization with multi-sensor input in all weather conditions. Our multi-stream network can handle LiDAR, Camera and RADAR inputs for localization on demand, i.e., it can work with one or more input sensors, making it robust to sensor failure. UnLoc uses 3D sparse convolutions and cylindrical partitioning of the space to process LiDAR frames and implements ResNet blocks with a slot attention-based feature filtering module for the Radar and image modalities. We introduce a unique learnable modality encoding scheme to distinguish between the input sensor data. Our method is extensively evaluated on Oxford Radar RobotCar, ApolloSouthBay and Perth-WA datasets. The results ascertain the efficacy of our technique.

Aknur Karabay, Arman Bolatov, Huseyin Atakan Varol et al.

Nowadays, it is common for people to take photographs of every beverage, snack, or meal they eat and then post these photographs on social media platforms. Leveraging these social trends, real-time food recognition and reliable classification of these captured food images can potentially help replace some of the tedious recording and coding of food diaries to enable personalized dietary interventions. Although Central Asian cuisine is culturally and historically distinct, there has been little published data on the food and dietary habits of people in this region. To fill this gap, we aim to create a reliable dataset of regional foods that is easily accessible to both public consumers and researchers. To the best of our knowledge, this is the first work on creating a Central Asian Food Dataset (CAFD). The final dataset contains 42 food categories and over 16,000 images of national dishes unique to this region. We achieved a classification accuracy of 88.70\% (42 classes) on the CAFD using the ResNet152 neural network model. The food recognition models trained on the CAFD demonstrate computer vision's effectiveness and high accuracy for dietary assessment.

Jyothis Chandran, Mathieu Remazeilles, R. B. Barreiro

Taking advantage of the reduced levels of noise and systematics in the data of the latest Planck release (PR4, also known as NPIPE), we construct a new all-sky Compton-$y$ parameter map (hereafter, $y$-map) of the thermal Sunyaev-Zeldovich (SZ) effect from the Planck PR4 data. A tailored Needlet Internal Linear Combination (NILC) pipeline, first validated on detailed sky simulations, is applied to the nine single-frequency Planck PR4 sky maps, ranging from $30$ to $857$ GHz, to produce the PR4 $y$-map over 98% of the sky. Using map comparisons, angular power spectra and one-point statistics we show that the PR4 NILC $y$-map is of improved quality compared to that of the previous PR2 release. The new $y$-map shows reduced levels of large-scale striations associated with $1/f$ noise in the scan direction. Regions near the Galactic plane also show lower residual contamination by Galactic thermal dust emission. At small angular scales, the residual contamination by thermal noise and cosmic infrared background (CIB) emission is found to be reduced by around 7% and 34%, respectively, in the PR4 $y$-map. The PR4 NILC $y$-map is made publicly available for astrophysical and cosmological analyses of the thermal SZ effect.

William R. Coulton, Mathew S. Madhavacheril, Adriaan J. Duivenvoorden et al.

Observations of the millimeter sky contain valuable information on a number of signals, including the blackbody cosmic microwave background (CMB), Galactic emissions, and the Compton-$y$ distortion due to the thermal Sunyaev-Zel'dovich (tSZ) effect. Extracting new insight into cosmological and astrophysical questions often requires combining multi-wavelength observations to spectrally isolate one component. In this work, we present a new arcminute-resolution Compton-$y$ map, which traces out the line-of-sight-integrated electron pressure, as well as maps of the CMB in intensity and E-mode polarization, across a third of the sky (around 13,000 sq.~deg.). We produce these through a joint analysis of data from the Atacama Cosmology Telescope (ACT) Data Release 4 and 6 at frequencies of roughly 93, 148, and 225 GHz, together with data from the \textit{Planck} satellite at frequencies between 30 GHz and 545 GHz. We present detailed verification of an internal linear combination pipeline implemented in a needlet frame that allows us to efficiently suppress Galactic contamination and account for spatial variations in the ACT instrument noise. These maps provide a significant advance, in noise levels and resolution, over the existing \textit{Planck} component-separated maps and will enable a host of science goals including studies of cluster and galaxy astrophysics, inferences of the cosmic velocity field, primordial non-Gaussianity searches, and gravitational lensing reconstruction of the CMB.

Sarah Steinbock-Pratt

Xin Lian

With the COVID-19 pandemic continuing, hatred against Asians is intensifying in countries outside Asia, especially among the Chinese. There is an urgent need to detect and prevent hate speech towards Asians effectively. In this work, we first create COVID-HATE-2022, an annotated dataset including 2,025 annotated tweets fetched in early February 2022, which are labeled based on specific criteria, and we present the comprehensive collection of scenarios of hate and non-hate tweets in the dataset. Second, we fine-tune the BERT model based on the relevant datasets and demonstrate several strategies related to the "cleaning" of the tweets. Third, we investigate the performance of advanced fine-tuning strategies with various model-centric and data-centric approaches, and we show that both strategies generally improve the performance, while data-centric ones outperform the others, and it demonstrates the feasibility and effectiveness of the data-centric approaches in the associated tasks.

Subekti Priyadharma

Dietrich Schmidt-Vogt

Philipp Lottholz, Arzuu Sheranova

This article tackles the fraught relationship between security discourse, on the theoretical level, and security experience and practice on the ground. It argues that the efforts of the Kyrgyzstani authorities to reform and thus create sustainable and needs-based community security and law enforcement structures have so far largely been performative or even “virtual”, meaning that they have focused on governing, but not producing security. The argument is first developed out of literature on state building, the security sector and police reform from a global perspective and in the context of Central Asia and Kyrgyzstan, more specifically. In a second step, we draw on insights from fieldwork, professional experience and grey literature to examine Local Crime Prevention Centres (LCPCs), which are communal-level public bodies where local administrations and residents potentially co-produce needs-based forms of security. However, we also show that the work of these bodies is still dependent on international support while lacking the conditions and facilitation that only executive actors can provide.

Henri Myrttinen