34711 research outputs found
Sort by
Reasoning with financial regulatory texts via large language models
No full textInterpreting complex financial regulatory texts, such as the Basel III Accords, can be challenging even for human experts. In this paper, we explore the potential of Large Language Models (LLMs) to perform such tasks. Specifically, we evaluate reasoning strategies, namely Chain-of-Thought (CoT) and Tree-of-Thought (ToT), in their ability to assign accurate risk weights to test cases based on the Basel III Standardized Approach (SA) for Credit Risk. Moreover, we propose and test a guided learning-based few-shot variant of CoT and ToT using human expert input. By evaluating 6,501 test cases, comprised of diverse exposure scenarios, our results demonstrate that few-shot prompting with CoT as well as ToT significantly enhances the LLMs’ accuracy in inferring risk weights. For one-shot CoT, we observe gains of almost 13 percentage points in accuracy with GPT-4o, whereas Claude 3 Sonnet shows gains of more than 10 percentage points. Albeit smaller in magnitude, one-shot ToT improvements are around 9 percentage points
Status of PV module recycling in IEA PVPS task 12 countries
No full textThis comprehensive publication examines the current state of PV module recycling, regulatory developments, and emerging technology trends, drawing on contributions from experts across the globe.
With PV deployment accelerating globally, the volume of end‑of‑life (EOL) modules is expected to grow significantly in the coming decades. Effective recycling will be critical to ensuring that PV remains a sustainable cornerstone of the global energy transition. The report reviews country‑by‑country approaches to EOL management, including the European Union’s dedicated PV waste regulations under the Waste Electrical and Electronic Equipment (WEEE) Directive, Japan’s ongoing development of PV‑specific recycling regulations, South Korea’s extended producer responsibility (EPR) scheme, China’s national incentive‑driven recycling initiatives, Australia’s planned mandatory product stewardship programme, and state‑level PV EOL regulations in parts of the United States.
Key highlights include:
- Regulatory frameworks are evolving worldwide. The EU has adopted the WEEE Directive for PV waste. In other parts of the world, legislative and regulatory frameworks for PV module waste are installed or in preparation. Regardless of whether there are PV-specific waste regulations, many companies are treating PV module waste for proper EOL management and recycling, and the number has increased since the last time IEA PVPS Task 12 surveyed three years ago.
- Current recycling faces economic and capacity challenges. The current low volumes, limited recycling technologies, logistics challenges, and underdeveloped markets for recovered materials result in a high-cost, low-revenue scenario for PV module recycling today. Further improvement in the PV recycling capacity and technology is needed to meet future increased demand and to realize the goal of high-value, low-cost recycling. To improve economic aspects of PV module recycling, considering values of recovered materials such as critical minerals would be also necessary.
- Technology development is broadening the scope of recycling. PV module recycling technology is expanding from delamination to metal recovery as well as exploring more valuable markets for recovered materials. Enabling the use of recovered materials in new PV cells/modules and other high-value markets are ultimate targets, whereas impurities and additives remain issues to be solved. Recycled materials from PV module waste could play a significant role in material supply for PV module production and other industries
Procrastination in estate planning extending the Rubicon model and exploring motivational dynamics
No full textEstate planning is an essential part of personal financial planning, but clients tend to delay addressing it (Hofmann & Tanner, 2023). A survey in Switzerland revealed that potential testators wait at the various stages of estate planning, although they have already decided to settle their estate. Some individuals delay estate planning even at a late stage in the process despite having already invested time and money in professional advice. In fact, it took individuals an average of 15 years to complete their estate planning (Hofmann & Tanner, 2023). Avoiding to address the matter can become a significant issue over time because of the increased risk that a potential testator will die without having made necessary provisions (Akerlof, 1991). By procrastinating their estate plans, potential testators forgo the opportunity to ensure the best possible financial security for their surviving dependents and to prevent disputes among their heirs through clear and precise instructions (Holzschneider & Schweissgut, 2011; Wheatley, 2013). Despite the significant practical implications outlined above, procrastination in estate planning is a yet understudied area (Yan & Zhang, 2022). We will present a theoretical framework which aims to advance the understanding of procrastination in estate planning. We apply the Rubicon Model (Gollwitzer & Achtziger, 2018) to estate planning and extend it by a preceding phase that underscores the importance of the awareness of the need for action as essential prerequisites for initiating the process. Second, we draw upon the mood- repair hypothesis (Pychyl & Sirois, 2016; Sirois & Pychyl, 2013) and the Temporal Motivation Theory (Steel & König, 2006; Steel, 2007) in order to explain the motivational and volitional dynamics of procrastination in estate planning. The whole model is highly context-specific in that all variables in of the models are carefully matched to corresponding variables to estate planning (e.g., fear of death). We use this theoretical model for our empirical research on procrastination in estate planning which we will outline and discuss at the conference. This research, on a yet understudied sample, will not only advance scientific understanding but also holds practical implications for professionals working in estate planning and financial advising
The underlying structures of self-attention : symmetry, directionality, and emergent dynamics in Transformer training
No full textSelf-attention is essential to Transformer architectures, yet how information is embedded in the self-attention matrices and how different objective functions impact this process remains unclear. We present a mathematical framework to analyze self-attention matrices by deriving the structures governing their weight updates. Using this framework, we demonstrate that bidirectional training induces symmetry in the weight matrices, while autoregressive training results in directionality and column dominance. Our theoretical findings are validated across multiple Transformer models—including ModernBERT, GPT, LLaMA3, and Mistral—and input modalities like text, vision, and audio. Finally, we apply these insights by showing that symmetric initialization improves the performance of encoder-only models on language tasks. This mathematical analysis offers a novel theoretical perspective on how information is embedded through self-attention, thereby improving the interpretability of Transformer models
Was hilft bei sexueller Belästigung gegenüber Pflegepersonen?
No full textSlot 21 Arena: Impulsreferat 2In diesem Impulsreferat werden in der Literatur beschriebene Ideen zum Umgang und Prävention von sexueller Belästigung gegenüber Pflegepersonen vorgestellt. Der Fokus liegt hierbei auf Strategien auf individueller, organisationaler und gesellschaftlicher Ebene. Dazu gehören beispielsweise Bildungs-, Melde-, Sicherheits- und Sensibilisierungsmassnahmen.
Présentation des idées décrites dans la littérature sur la gestion et la prévention du harcèlement sexuel envers les soignants. L'accent est mis sur les stratégies au niveau individuel, organisationnel et sociétal. Il s'agit par exemple de mesures de formation, de signalement, de sécurité et de sensibilisatio
Global impoverishment of natural vegetation revealed by dark diversity
No full textAnthropogenic biodiversity decline threatens the functioning of ecosystems and the many benefits they provide to humanity1. As well as causing species losses in directly affected locations, human influence might also reduce biodiversity in relatively unmodified vegetation if far-reaching anthropogenic effects trigger local extinctions and hinder recolonization. Here we show that local plant diversity is globally negatively related to the level of anthropogenic activity in the surrounding region. Impoverishment of natural vegetation was evident only when we considered community completeness: the proportion of all suitable species in the region that are present at a site. To estimate community completeness, we compared the number of recorded species with the dark diversity-ecologically suitable species that are absent from a site but present in the surrounding region2. In the sampled regions with a minimal human footprint index, an average of 35% of suitable plant species were present locally, compared with less than 20% in highly affected regions. Besides having the potential to uncover overlooked threats to biodiversity, dark diversity also provides guidance for nature conservation. Species in the dark diversity remain regionally present, and their local populations might be restored through measures that improve connectivity between natural vegetation fragments and reduce threats to population persistence
Beyond the nucleus : cytoplasmic dominance in follicular thyroid carcinoma detection using single-cell Raman imaging across multiple devices
No full textCytological diagnosis of follicular thyroid carcinoma (FTC) is one of the main challenges in the field of endocrine oncology due to the absence of evident morphological indicators. Morphological abnormalities in the nucleus are typically key indicators of cancer cytopathology and are attributed to a range of biochemical alterations in nuclear components. Consequently, Raman spectroscopy has been widely used to detect cancer in various cytological samples, often identifying biochemical changes prior to observable morphological alterations. However, in the case of FTC, cytoplasmic features, such as carotenoids, cytochromes, and lipid droplets, have shown greater diagnostic relevance compared to nuclear features. This study leverages single-cell Raman imaging to explore the spatial origin of diagnostic signals in FTC and normal thyroid (NT) cells, assessing the contributions of the nucleus and cytoplasm independently. Our results demonstrate that Raman spectra from the cytoplasmic region can distinguish between FTC and NT cells with an accuracy of 84% under coculture conditions, consistently across two cell lines originated from two donors and maintaining robustness across multiple devices. In contrast, classification based on nuclear spectra achieved only 53% accuracy, suggesting that biochemical alterations in the cytoplasm play a more significant role in FTC detection than those in the nucleus. Our work elevates the promise of Raman-based cytopathology by providing complementary organelle-dependent information to traditional diagnostic methods and demonstrating transferability across different devices
Massenverfahren
No full textDas ordentliche Verwaltungsverfahren ist auf den Erlass «normaler» Verfügungen ausgerichtet. Massenverfahren lassen sich demgegenüber nach den üblichen Verfahrensstandards nicht mehr adäquat bewältigen, weshalb die geltenden Verfahrensregeln schon fast zwangsläufig in Konflikt mit den Geboten der Prozessökonomie und Praktikabilität treten. Will man das verfahrensbezogene Legalitätsprinzip nicht über Gebühr strapazieren, müssen Sonderregeln geschaffen werden, die eine effiziente und einfache Abwicklung derartiger Verfahren unter Berücksichtigung gewisser rechtsstaatlicher Mindeststandards erlauben
How healthcare provider characteristics affect their attitudes and skills in involving families in caring for small and sick newborns throughout the perinatal period
No full textIntroduction: Healthcare providers (HCPs) are essential in providing care and working with families with small and sick newborns throughout the perinatal care continuum. While clinical experience, education, and training influence HCPs’ attitudes and skills regarding family involvement in care, the specific factors affecting Ghanaian HCPs remain unclear.
Objectives: To identify HCPs’ characteristics that influence their attitudes and self-perceived practice skills toward involving families in the perinatal care continuum for small and sick newborns.
Methods: This analytical cross-sectional study involved 141 HCPs, including nurses, midwives, and medical staff, from one secondary and 13 primary healthcare facilities in Ghana. Participants completed an online survey using two standardized scales: (1) “Family Importance in Nursing Care-Nursing Attitudes” which assesses HCPs’ perceptions of the significance of family involvement in care and the (2) “Family Nursing Practice Scale” which evaluates the extent to which HCPs integrate family-systems approaches in their practice. Data analysis involved multiple linear regression models.
Results: Self-perceived practice skills and attitudes towards family involvement influenced HCPs in multiple dimensions, including viewing “family as a burden” (associated with increased stress), “as a conversational partner” (as active participants in care), and as a “resource in care” (having their own coping resources). Availability of a family systems care work approach and prior education in family systems care significantly influenced HCPs’ self-perceived practice skills. HCPs skills and attitudes in family involvement influenced each other.
Conclusion: Transforming attitudes towards family systems care across all levels, from hospital to home, should aim to improve the quality of newborn care. Tailored family systems care education, skills training, and institutional support are needed to enhance HCPs’ skills and attitudes across the perinatal care continuum for small and sick newborns