Prototypical Case Research Articles

Limitations of Feature Attribution in Long Text Classification of Standards

Managing complex AI systems requires insight into a model's decision-making processes. Understanding how these systems arrive at their conclusions is essential for ensuring reliability. In the field of explainable natural language processing, many approaches have been developed and evaluated. However, experimental analysis of explainability for text classification has been largely constrained to short text and binary classification. In this applied work, we study explainability for a real-world task where the goal is to assess the technological suitability of standards. This prototypical use case is characterized by large documents, technical language, and a multi-label setting, making it a complex modeling challenge. We provide an analysis of approx. 1000 documents with human-annotated evidence. We then present experimental results with two explanation methods evaluating plausibility and runtime of explanations. We find that the average runtime for explanation generation is at least 5 minutes and that the model explanations do not overlap with the ground truth. These findings reveal limitations of current explanation methods. In a detailed discussion, we identify possible reasons and how to address them on three different dimensions: task, model and explanation method. We conclude with risks and recommendations for the use of feature attribution methods in similar settings.

A closure theorem for $\Gamma$-convergence and $H$-convergence with applications to non-periodic homogenization

In this work we examine the stability of some classes of integrals, in particular with respect to homogenization. The prototypical case is the homogenization of quadratic energies with periodic coefficients perturbed by a term vanishing at infinity, which has recently been examined in the framework of elliptic PDEs. We use localization techniques and higher-integrability Meyers-type results to provide a closure theorem by \Gamma -convergence within a large class of integral functionals. From such a result we derive stability theorems in homogenization which comprise the case of perturbations with zero average on the whole space. The results are also extended to the stochastic case, and specialized to the G -convergence of operators corresponding to quadratic forms. A corresponding analysis is also carried out for non-symmetric operators using the localization properties of H -convergence. Finally, we treat the case of perforated domains with Neumann boundary condition, and their stability.

Time Delays as Attosecond Probe of Interelectronic Coherence and Entanglement.

Attosecond chronoscopy enables the exploration of correlated electron dynamics in real time. One key observable of attosecond physics is the determination of "time zero" of photoionization, the time delay with which the wave packet of the ionized electron departs from the ionic core. This observable has become accessible by experimental advances in attosecond streaking and reconstruction of attosecond beating by interference of two-photon transitions (RABBIT) techniques. In this Letter, we explore photoionization time delays by strong extreme ultraviolet fields beyond the linear-response limit. We identify novel signatures in time delays signifying strong coupling between atoms and light fields and the light-field dressing of the ion. As a prototypical case, we study the interelectronic coherence and entanglement in helium driven by a strong extreme ultraviolet field. By the numerical solution of the time-dependent Schrödinger equation in its full dimensionality, we show that the time delay of the photoionized electron allows one to monitor the ultrafast variations of coherence dynamics and entanglement in real time.

Pseudo-proper two-dimensional electron gas formation

Abstract In spite of the interest in the two-dimensional electron gases (2DEGs) experimentally found at surfaces and interfaces, important uncertainties remain about the observed insulator–metal transitions (IMTs). Here we show how an explicit pseudo-proper coupling of carrier sources with a relevant soft mode significantly affects the transition. The analysis presented here for 2DEGs at polar interfaces is based on group theory, Landau–Ginzburg theory, and illustrated with first-principles calculations for the prototypical case of the LaAlO 3 / SrTiO 3 interface, for which such a structural transition has recently been observed. This direct coupling implies that the appearance of the soft mode is always accompanied by carriers. For sufficiently strong coupling an avalanche-like first-order IMT is predicted.

Transient behaviour and heat transfer characteristics in debris beds: Simulation and analysis of the LIVEJ2 experiment

Multiple uncertainties still exist about the state of the debris in Fukushima Daiichi Nuclear Power Station (1F). In the past, the attention of the nuclear safety community was focused on the heat transfer characteristics in the case of an homogeneous pool, but little attention was given to address the melting and heat transfer in the presence of a debris bed constituted of materials with different melting points. This condition represents a challenge for CFD analyses, because it includes multi-physics conditions, such as a low melting point fluid convecting into a debris bed surrounded by a crust on the vessel wall which has received little attention compared to classical CFD analyses. Even though a comprehensive analysis of a related experiment (i.e. LIVE-J2) has been performed recently by Madokoro et al. (2023) little attention on the results has been paid to the effect of debris bed porosity and the existence of a gap between the vessel wall and the crust. In the paper we have modified the porosity resistance based on the Ergun equation and proposed a simple model for the gap conductance in the lower part of the crust. The results show an improvement in the prediction of the thermal stratification and the vessel temperature in the lower locations. In addition, highlight that such phenomena constitute key parameters to keep into consideration in the simulation of prototypical cases both for CFD and lumped parameter codes (e.g. MELCOR, MAAP).

Modeling the Dispersion of Waves in a Multilayered Inhomogeneous Membrane with Fractional-Order Infusion

The dispersion of elastic shear waves in multilayered bodies is a topic of extensive research due to its significance in contemporary science and engineering. Anti-plane shear motion, a two-dimensional mathematical model in solid mechanics, effectively captures shear wave propagation in elastic bodies with relative mathematical simplicity. This study models the vibration of elastic waves in a multilayered inhomogeneous circular membrane using the Helmholtz equation with fractional-order infusion, effectively leveraging the anti-plane shear motion equation to avoid the computational complexity of universal plane motion equations. The method of the separation of variables and the conformable Bessel equation are utilized for the analytical examination of the model’s resulting vibrational displacements, as well as the dispersion relation. Additionally, the influence of various wave phenomena, including the dependencies of the wavenumber on the frequency and the phase speed on the wavenumber, respectively, with the variational effect of the fractional order on wave dispersion is considered. Numerical simulations of prototypical cases validate the formulated model, illustrating its applicability and effectiveness. The study reveals that fractional-order infusion significantly impacts the dispersion of elastic waves in both single- and multilayer membranes. The effects vary depending on the membrane’s structure and the wave propagation regime (long-wave vs. short-wave). These findings underscore the potential of fractional-order parameters in tailoring wave behavior for diverse scientific and engineering applications.

Comparative Analysis of Mechanical Thrombectomy Outcomes of Middle Cerebral Artery M1, M2 Superior, and M2 Inferior Occlusion Strokes

BackgroundThe M1 middle cerebral artery (MCA) commonly bifurcates into an M2 superior and an M2 inferior segment. However, MCA anatomy is highly variable rendering classification for mechanical thrombectomy (MT) trials difficult. Safety and effectiveness of M2 MCA stroke thrombectomy stratified by M2 MCA anatomy remains to be explored. MethodsLarge vessel occlusion strokes undergoing MT between 02/2016-08/2022 were reviewed (n=784). M1 (n=431) and M2 MCA (n=118) occlusions were assessed. Among M2 MCA occlusions only prototypical MCA bifurcation anatomy cases were included (n=99). Dominance was assessed based on angiography. Procedural and outcome data was compared between M1, M2 superior and M2 inferior MCA occlusions. ResultsBaseline demographics and periprocedural criteria of M2 superior (n=56) and M2 inferior (n=43) occlusion MTs were comparable. Among M2 inferior cases, the occluded branch was dominant in 41/43 (95.3%) but only in 37/56 (66.1%) among M2 superior cases (p<0.001). The 90-day favorable functional outcome (mRS 0-2) and mortality (mRS 6) rates were 60.0% and 8.9% in the M2 superior, 42.9% and 32.6% in the M2 inferior, and 44.1% and 26.0% in the M1 group (n=431). Compared to M2 superior, M2 inferior favorable outcome rates were lower (p=0.094) and mortality rates were higher (p=0.003) and resembled M1 outcome rates (p=0.750 and p=0.355, respectively). ConclusionIn setting of prototypical MCA bifurcation anatomy, thrombectomy of dominant M2 inferior occlusions had outcome rates like M1 occlusions. In contrast, M2 superior occlusions had significantly lower mortality rates and a trend towards better favorable functional outcome rates.

Enabling MOCVD production on next generation 150 mm Indium Phosphide wafer size

150 mm Indium Phosphide wafers are now commercially available with crystal quality comparable to wafer of smaller radius. This may pave the way for scaling up the production of a multitude of photonic devices for Datacoms operating in the 1.30-1.55μm infrared range with a gain in wafer surface of factor 4 and a reduction of roughly 50% in die cost. To achieve this goal, it is of utmost importance to prove that both AlGaInAs and InGaAsP quaternary compound semiconductors can be grown by metal–organic chemical vapor deposition (MOCVD) on wafers of larger sizes with both thickness and composition uniformities comparable to those achievable on 75 mm. In this article, we report pioneering production technology developments based on the Planetary Reactor® design. Both reactor and related inlet geometry have been deeply revisited with the introduction of a novel 4-fold injector, which in combination with Cl2 In-situ chamber clean, prove to enable such transition in wafer size. Sub-nanometric photoluminescence in-wafer uniformities are demonstrated and historic challenges, such as drift in material composition of highly sensitive InGaAsP alloys during a production campaign, are addressed thanks to this unique combination. Uniformity, tunability and reproducibility results are thus presented for two prototypical case scenarios: a highly strained AlGaInAs multiple quantum well (MQW) and a bulk InGaAsP layer with wavelength emission of 1550 nm and 1100 nm respectively to corroborate reactor flexibility in meeting industry requirements for next device generation.

An Analysis of Drilling Projection Uncertainty and Implications for Collision Avoidance Management Systems

Summary During well planning, specifying the separation that must be maintained from offset wellbores is critical to ensuring safe operations. Recent efforts on API RP 78, an upcoming American Petroleum Institute recommended practice for wellbore placement, have provided guidelines for calculating safe separation distances and engineering considerations for those calculations. A major milestone was the publication of a separation rule (Sawaryn et al. 2019) standardizing a formula for separation factor (SF), which in turn defines the minimum allowable separation distance (MASD) between two wellbores and the allowable deviation from plan (ADP) for a wellbore being drilled. One element of this formula, σpa, describes the accuracy with which a future wellpath can be predicted. Prior guidance provides a suggested value for σpa (1.6 ft or 0.5 m at 1-sigma) based on heuristics. A more rigorous approach for estimating this value is presented, where the design of the drilling program can be considered. A framework is proposed for modeling deviation from a projected path using a planned trajectory, bottomhole assembly (BHA) properties, and survey practices. The method identifies where a deviation is detected through survey measurements and estimates a planned recovery operation. Equations are provided for estimating the distance from plan at deviation detection along with the maximum expected deviation during recovery. Common drilling scenarios are analyzed for sensitivity to operational parameters such as survey course length, sensor offset, tool face control, and BHA performance. The impact of varying σpa is explored across prototypical collision avoidance cases. A discussion of the relationships between the σpa value and the common collision avoidance calculations, such as SF, MASD, and ADP, is included along with considerations for the design of a risk management policy. Survey course length, survey sensor offset, and uncertainty in directional performance are all shown to have a significant impact on the potential maximum deviation from plan. An additional factor considered is the aggressiveness expected when performing recovery operations. An analysis of common drilling scenarios suggests the previously provided guidance of 1.6 ft at 1-sigma appears suitable for cases where the directional behavior of the BHA is well characterized and the combination of course length and survey sensor offset is kept to 150 ft or less. Longer survey course lengths, larger sensor offsets, or uncertainty in directional performance of the BHA can produce larger σpa values, in some scenarios causing a significant impact on the MASD and ADP. Previous work has called attention to the uncertainty in drilling to a projection, but up to now, there has not been a rigorous method for estimating that quantity. With the methods outlined in this paper, well planners and drilling engineers can make more informed decisions on how to ensure safe separation practices in their own operations.

The Sectoral Politics of Industrial Policy Making in Brazil: A Polanyian Interpretation

ABSTRACTThis article considers why Brazilian industrial policies have varied across sectors since the mid‐1990s. It relies on a Polanyian‐inspired framework to propose that the strength of counter‐movements against corporate welfare shapes the sector‐specific capacity of policy makers to exert state discipline over business interests and diverges from neoliberal scripts of industrial policy making. The authors use prototypical case studies on the automotive, animal protein and pharmaceutical sectors to support their argument. In the automotive industry, the continuous pressure from powerful and cohesive labour unions led to the emergence of a neo‐corporatist sectoral regime that was characterized by a tripartite policy design and encompassed conditionalities. In the case of animal protein, the lack of bottom‐up pressure culminated in a disembedded neoliberal sectoral regime, in which business owners received almost unconditional benefits, turning industrial policies into corporate welfare. Finally, in the pharmaceutical industry, the combination of diffuse societal demands and unions with intermediate relevance led to an embedded neoliberal sectoral regime that combined selective conditionalities with some space for non‐business participation in policy design.

Forecasting with an N-dimensional Langevin equation and a neural-ordinary differential equation.

Accurate prediction of electricity day-ahead prices is essential in competitive electricity markets. Although stationary electricity-price forecasting techniques have received considerable attention, research on non-stationary methods is comparatively scarce, despite the common prevalence of non-stationary features in electricity markets. Specifically, existing non-stationary techniques will often aim to address individual non-stationary features in isolation, leaving aside the exploration of concurrent multiple non-stationary effects. Our overarching objective here is the formulation of a framework to systematically model and forecast non-stationary electricity-price time series, encompassing the broader scope of non-stationary behavior. For this purpose, we develop a data-driven model that combines an N-dimensional Langevin equation (LE) with a neural-ordinary differential equation (NODE). The LE captures fine-grained details of the electricity-price behavior in stationary regimes but is inadequate for non-stationary conditions. To overcome this inherent limitation, we adopt a NODE approach to learn, and at the same time predict, the difference between the actual electricity-price time series and the simulated price trajectories generated by the LE. By learning this difference, the NODE reconstructs the non-stationary components of the time series that the LE is not able to capture. We exemplify the effectiveness of our framework using the Spanish electricity day-ahead market as a prototypical case study. Our findings reveal that the NODE nicely complements the LE, providing a comprehensive strategy to tackle both stationary and non-stationary electricity-price behavior. The framework's dependability and robustness is demonstrated through different non-stationary scenarios by comparing it against a range of basic naïve methods.

Clinical Experiences Using Cognitive-Behavioral Therapy for Eating Disorders

Data suggests that despite the availability of evidence-based psychological treatments for eating disorders (EDs), techniques from these therapies may be less frequently used within real-life clinical practice. The aim of this study was to provide the opportunity for clinicians to give feedback on their experiences treating EDs using cognitive-behavioral therapy (CBT) through reporting on use of CBT techniques and barriers to treatment implementation in naturalistic settings. Clinicians (N = 126) who self-identified as using CBT for EDs reported demographic information, frequency/usefulness of empirically supported treatment techniques, problems/limitations of CBT, and barriers faced while implementing CBT. The most frequently used technique reported by clinicians was psychoeducation, and the least frequently used technique was use of surveys to address mind reading. Patients’ unwillingness to follow a meal plan/nutritional guide was rated as the most impactful barrier, alongside ED severity. Of the problems/limitations of CBT, too little guidance on treating co-occurring symptoms was rated as the most impactful. This study provided a mechanism for clinicians to share their experiences using CBT for EDs in real-world settings. Overall, results regarding frequency of use and usefulness of techniques indicate a high level of endorsement. Moreover, the most frequently endorsed barriers to/limitations of CBT related to lack of guidance on treating complex ED presentations. Future research should explore ways to treat cases that go beyond the prototypical ED case and explore ways to adapt CBT to meet the needs of naturalistic treatment settings.

Realism against delegitimation

ABSTRACT Political realists exercise ideology critique to emancipate citizens from problematic beliefs concerning the legitimacy of their social order. They seek to unveil hidden conflict within apparent harmony. However, realists have so far neglected the opposite case in which erroneous beliefs delegitimise a social order, thus contributing to unrest and resentment. As a prototypical case for delegitimation, I will discuss the ‘big lie’ narrative that surrounds the 2020 presidential election in the United States. As I will argue, realist ideology critique is currently underequipped to deal with these cases. To address this shortcoming, I propose a modification to the critical realist methodology. I will argue that realists should focus on unveiling self-serving, rather than merely self-justifying, political narratives. In line with a more practice-oriented approach to ideology critique, I will further sketch some implications for liberal democratic institutions that follow from taking delegitimation seriously.

Creutzfeldt-Jakob disease presenting as psychiatric disorder: case presentation and systematic review.

Creutzfeldt-Jakob disease (CJD) is a spongiform encephalopathy caused by misfolded human prion proteins (PrP)s. Due to variability in presentation, the diagnosis may be missed in lieu of various psychiatric disorders. Our study reports on a prototypical case and psychiatric mimic for CJD, and the workup used to establish the correct diagnosis. A 54-year-old male with a past medical history of traumatic brain injury and major depressive disorder presented with chest pain. During the hospital stay, he was found to be increasingly aggressive, and behaved out of character. Further review of clinical history revealed that the patient was diagnosed with cognitive impairment and depression one year prior. The patient was agitated, poorly redirectable, and had unstable gait on neurological examination. Magnetic resonance imaging (MRI) of the brain demonstrated restricted diffusion (DWI) along the parietooccipital and temporal regions (L > R) and in the subcortical structures, including the basal ganglia and thalami, with accompanying subtle fluid attenuation inversion recovery (FLAIR) hyperintense signal abnormality in these regions, deemed as artifactual at the time. Repeat MRI brain two months later demonstrated progression of the DWI signal with ADC correlate and FLAIR findings. Cerebrospinal fluid 14-3-3 and RT-QuIC samples were positive. Upon passing a few months later, brain autopsy and Western Blot confirmed the CJD diagnosis. Literature review was conducted on PubMed to identify CJD cases initially diagnosed as psychiatric disorder. Search terms included "CJD" or "Creutzfeldt-Jakob disease" with three common psychiatric diagnoses, "Depression," "Psychosis," and "Mania." Positive EEG, MRI, PET, and CSF (including protein 14-3-3 and tau) findings for CJD were found in 66.7, 81.1, 50, and 72.7% of cases, respectively. Overall, CJD can present as a psychiatric mimic. In suspicious cases, EEG, imaging, and CSF studies should be promptly utilized to arrive at the correct diagnosis. Repeated MRI imaging is often required to help in the diagnostic process. Brain biopsy should be considered in selected cases.

The Illness Trajectory in Marilyn Monroe's Psychological Autopsy: From Autism Spectrum Disorder to Borderline Personality Disorder and Bipolar Disorder with Catatonia.

A mounting body of literature is showing that, in the clinical and general population, autism spectrum disorder (ASD) or autistic traits (ATs) would appear to be spread along a continuum, reaching the highest levels among individuals affected by other mental disorders, such as borderline personality disorder (BPD) and bipolar disorder (BD). Furthermore, individuals with ASD or ATs appear to be more vulnerable to psychological traumas, with greater likelihood of developing post-traumatic stress disorder (PTSD) and BD. Marilyn Monroe was a famous actress, singer and model, as well as one of the most admired stars of American cinema and a timeless icon. The present report debates the possibility to explore Marilyn Monroe's case under a neurodevelopmental perspective according to which a ASD favored, on one hand, her worldwide success and, on the other, her mental illness trajectory. The analysis is based on the review of her four biographies written by psychiatrists and clinical psychologists, her complete filmography, interviews, filmed material and personal scripts. The present work reconstructed a hypothetical illness trajectory originating from a ASD and culminating in a BD with unspecified catatonia, followed by premature death whose cause has never been clarified. The description of this illness trajectory also confirms the strong impact of psychological traumas on the substrate of ASD vulnerability and its correlation with the development of BPD and BD. This paper suggests Marilyn Monroe as a possible prototypical case of BPD under a neurodevelopmental perspective that accounts the ASD, as reported in DSM-5-TR (2022), in verbal adult without intellectual impairment, high-functioning autism (HFA), as vulnerability background predisposing to the progression to BPD and BD with unspecified catatonia, triggered by multiple traumas.

From Urban Vulnerabilities to Resilience: Lessons from Messina's Integrated Risk Approach

In the city of Messina, the multiplicity of vulnerable situations is a distinctive feature and is representative of the pervasiveness of the risk conditions present in the Italian territory. The critical residential tissues are part of an extended geography in which, to the conditions of seismic and hydro geomorphological vulnerability, are added additional criticalities including altered natural resources, abandoned infrastructure, brownfields, quarries, and unregulated landfills, etc. Despite this, for Messina, there is still no urban planning tool capable of interpreting in an integrated way the risks present, which instead continue to be addressed on an emergency basis and individually, without considering the interactions that are generated and of further damage. For these reasons, Messina was chosen as a prototypical case study, at the national level, to start an experimentation aimed at developing an urban planning tool capable of a new integrated approach to risk interpretation and management. In this process, a decisive role was aimed at building an updated knowledge of the territory through the identification of widespread critical conditions using multiple GIS tools and thanks to the interaction of some analytical-specialist readings from different sources. The results of this process are represented in a system of maps that are strongly integrated with each other and constitute a working document for the interdisciplinary group drafting the General Urban Plan.

“ChatGPT, Can You Help Me Save My Child’s Life?” - Diagnostic Accuracy and Supportive Capabilities to Lay Rescuers by ChatGPT in Prehospital Basic Life Support and Paediatric Advanced Life Support Cases – An In-silico Analysis

BackgroundPaediatric emergencies are challenging for healthcare workers, first aiders, and parents waiting for emergency medical services to arrive. With the expected rise of virtual assistants, people will likely seek help from such digital AI tools, especially in regions lacking emergency medical services. Large Language Models like ChatGPT proved effective in providing health-related information and are competent in medical exams but are questioned regarding patient safety. Currently, there is no information on ChatGPT’s performance in supporting parents in paediatric emergencies requiring help from emergency medical services. This study aimed to test 20 paediatric and two basic life support case vignettes for ChatGPT and GPT-4 performance and safety in children.MethodsWe provided the cases three times each to two models, ChatGPT and GPT-4, and assessed the diagnostic accuracy, emergency call advice, and the validity of advice given to parents.ResultsBoth models recognized the emergency in the cases, except for septic shock and pulmonary embolism, and identified the correct diagnosis in 94%. However, ChatGPT/GPT-4 reliably advised to call emergency services only in 12 of 22 cases (54%), gave correct first aid instructions in 9 cases (45%) and incorrectly advised advanced life support techniques to parents in 3 of 22 cases (13.6%).ConclusionConsidering these results of the recent ChatGPT versions, the validity, reliability and thus safety of ChatGPT/GPT-4 as an emergency support tool is questionable. However, whether humans would perform better in the same situation is uncertain. Moreover, other studies have shown that human emergency call operators are also inaccurate, partly with worse performance than ChatGPT/GPT-4 in our study. However, one of the main limitations of the study is that we used prototypical cases, and the management may differ from urban to rural areas and between different countries, indicating the need for further evaluation of the context sensitivity and adaptability of the model. Nevertheless, ChatGPT and the new versions under development may be promising tools for assisting lay first responders, operators, and professionals in diagnosing a paediatric emergency.Trial registrationNot applicable.

Target rate factors in short rate models

This study investigates the risk associated with the uncertainties in the central bank monetary policy targets in the context of short interest rate models. A class of models is proposed which admits two channels of interest rate risk. In a prototypical case, the short duration channel handles the uncertainties in the target rates decided in the forthcoming Federal Open Market Committee meetings. The target rate factors can be calibrated on the market values of the Fed funds futures. The long duration channel has traditional risk factors. The episodes following the Covid-19 outbreak and the 2022 rate hikes are used as examples in an empirical study.

A God Gap Driving a Revolution from Conservative to the Far Right in the United States—With Significance for Europe?

For some time now, the Christian right in the United States has sought proximity to the Republican Party. However, it is only since Donald Trump’s presidency that the central desires of the Christian right for a moral change have been fulfilled. This hope has existed since the 1980s and has been expressed in far-above-average voting behavior, especially of white evangelicals and Pentecostals for the Republicans. Thus, the Republicans’ focus on moral issues increasingly links them to the image of white Christian nationalism. This stance opposes transgender issues, homosexuality, abortion, and critical race theory and is preoccupied with America’s founding era and racist structures. Favored by the bipolar electoral system and driven by a politics that differentiates between good and evil, a political polarization is establishing itself that leaves hardly any room for compromise. These developments are not limited to the United States but are beginning to migrate globally, as recent incidents in Brazil, Serbia, and Russia show. The United States can be seen as a prototypical case of a particular form of transformation of a democracy into a polarized political system. A certain understanding of religion plays just as much a role in this as the rejection of plurality, especially at the level of sexual and gender diversity.

From policy to practice: how schools implement German language support policy in Austria

ABSTRACT In 2018/2019, the Government of Austria instituted a new language support model to Austrian schools that included segregated language support measures for non-fluent German-speaking students. The central research aim of the paper is to investigate how the implementation of this new language support model compares to the policy requirement by examining different models of implementation. In the framework of the initial phases of the project ‘A multi-perspective study on German language support’, 12 interviews with German language support teachers and school principals were conducted. Following a prototypical case study analysis, four models of implementation were identified: (1) Segregationmodel, (2) Mixed segregated and integrated model, (3) Individualized language support, and (4) Inclusive support model. Given that thisis the first study in Austria to show the concrete implementation models of German language support in schoolsFindings highlight important implications for school principals and educational policy makers.