Search

Knowledge tracing plays a crucial role in effectively representing learners' understanding and predicting their future learning progress. However, existing deep knowledge tracing methods, reliant on the forgetting model and Rasch model, often fail to account for the varying rates at which learners forget different knowledge concepts and the variations in question embedding covering the same concept. To address these limitations, this paper introduces an enhanced deep knowledge tracing model that combines the transformer network model with two innovative components. The first component is a multiband attention mechanism, which comprehensively summarizes a learner's past response history across various temporal scales. By computing attention weights using different decay rates, this mechanism adaptively captures both long-term and short-term interactions for different knowledge concepts. The second component utilizes a quantized question embedding module to effectively capture variations among questions addressing the same knowledge concept. This module represents these differences in a rich embedding space, avoiding overparameterization or overfitting issues. The proposed model is evaluated on popular benchmark datasets, demonstrating its superiority over existing knowledge tracing methods in accuracy. This enhancement holds potential for improving personalized learning systems by providing more precise insights into learners' progress. [ABSTRACT FROM AUTHOR]

Defect‐induced recombination dynamics remain an essential factor hindering the efficiency and stability of perovskite solar cells (PSCs). Herein, an upgraded solvent posttreatment strategy is proposed to simultaneously improve the quality and related electronic properties of absorbers by introducing the black phosphorus–graphene oxide (BP–GO) composites with a unique microscopic heterojunction stacking structure, high charge mobility, tunable bandgap, and excellent surface chemical properties into the grain boundaries and surface of the perovskite. The BP–GO composites block the layer‐to‐layer diffusion of active ions while directionally passivating uncoordinated defects in the perovskite, thereby accelerating the carrier extraction/transfer at the perovskite interface, ultimately significantly reducing the nonradiative recombination and interfacial charge recombination losses of the devices. The BP–GO composites improve Oswald ripening, guaranteeing the formation of high‐crystallinity and large‐grain perovskites, thereby significantly enhancing the lattice stability, interface quality, and light‐harvesting capacity of the absorbers. Benefiting from the synergistic action of the unique lattice structure of BP–GO, excellent passivation/isolation effect, and enhanced charge dynamics, the photovoltaic output performance of the BP–GO‐modified PSCs is significantly higher than that of conventional n–i–p planar PSCs, and the corresponding unencapsulated device exhibits good stability. This work provides theoretical and technical guarantees for achieving more efficient and stable planar PSCs. [ABSTRACT FROM AUTHOR]

The development of functional absorbers with excellent film-forming quality and carrier extraction/transfer efficiency is of great significance to achieve high-efficiency and stable n–i–p planar perovskite solar cells (PSCs). Herein, we propose a simple in situ growth technique to construct high-quality absorbers by introducing a commercial two-dimensional (2D) tungsten sulfide nanosheet (WS2 NS) modifier during the growth of a perovskite. The synergistic effects of WS2 on perovskite crystallization kinetics, charge extraction/transfer/recombination behaviors, and device photovoltaic outputs were systematically studied. Owing to the strong coordination effect between WS2 NSs and organic cations in a perovskite, WS2 can be used as a growth template in the crystallization process of the perovskite, thus effectively passivating grain boundaries and surface/interface trap states, and synchronously improving the film-forming quality, surface/interface contact, and photogenerated charge extraction/transfer efficiency of the perovskite. Moreover, WS2 located at the grain boundaries and surface/interface provides the "fast paths" for photogenerated electron–hole pairs by virtue of its high charge mobility, which greatly inhibits the accumulation/recombination of charge carriers and effectively improves the photovoltaic performance and hysteresis behavior of planar PSCs. Additionally, WS2 NSs induce the substitutional doping of formamidinium cations, which has a positive contribution to broadening the spectral response range. The results show that the efficiency of the WS2-modified PSC is significantly better than that of the conventional planar device. The non-encapsulated functionalized devices also exhibit good stability due to the multifunctional effects of a WS2-modified perovskite, resulting in great advantages for the development of highly efficient flexible-, semitransparent-, and tandem-photovoltaic applications. [ABSTRACT FROM AUTHOR]