Search

Anti-Muslim hate speech has emerged within memes, characterized by context-dependent and rhetorical messages using text and images that seemingly mimic humor but convey Islamophobic sentiments. This work presents a novel dataset and proposes a classifier based on the Vision-and-Language Transformer (ViLT) specifically tailored to identify anti-Muslim hate within memes by integrating both visual and textual representations. Our model leverages joint modal embeddings between meme images and incorporated text to capture nuanced Islamophobic narratives that are unique to meme culture, providing both high detection accuracy and interoperability., Comment: Accepted (Poster) - NeurIPS 2024 Workshop MusIML

Learned image compression (LIC) is currently the cutting-edge method. However, the inherent difference between testing and training images of LIC results in performance degradation to some extent. Especially for out-of-sample, out-of-distribution, or out-of-domain testing images, the performance of LIC dramatically degraded. Classical LIC is a serial image compression (SIC) approach that utilizes an open-loop architecture with serial encoding and decoding units. Nevertheless, according to the theory of automatic control, a closed-loop architecture holds the potential to improve the dynamic and static performance of LIC. Therefore, a circular image compression (CIC) approach with closed-loop encoding and decoding elements is proposed to minimize the gap between testing and training images and upgrade the capability of LIC. The proposed CIC establishes a nonlinear loop equation and proves that steady-state error between reconstructed and original images is close to zero by Taylor series expansion. The proposed CIC method possesses the property of Post-Training and plug-and-play which can be built on any existing advanced SIC methods. Experimental results on five public image compression datasets demonstrate that the proposed CIC outperforms five competing state-of-the-art open-source SIC algorithms in reconstruction capacity. Experimental results further show that the proposed method is suitable for out-of-sample testing images with dark backgrounds, sharp edges, high contrast, grid shapes, or complex patterns.

Classical methods of medical image super-resolution (MISR) utilize open-loop architecture with implicit under-resolution (UR) unit and explicit super-resolution (SR) unit. The UR unit can always be given, assumed, or estimated, while the SR unit is elaborately designed according to various SR algorithms. The closed-loop feedback mechanism is widely employed in current MISR approaches and can efficiently improve their performance. The feedback mechanism may be divided into two categories: local feedback and global feedback. Therefore, this paper proposes a global feedback-based closed-cycle framework, circular MISR (CMISR), with unambiguous UR and advanced SR elements. Mathematical model and closed-loop equation of CMISR are built. Mathematical proof with Taylor-series approximation indicates that CMISR has zero recovery error in steady-state. In addition, CMISR holds plug-and-play characteristic that fuses model-based and learning-based approaches and can be established on any existing MISR algorithms. Five CMISR algorithms are respectively proposed based on the state-of-the-art open-loop MISR algorithms. Experimental results with three scale factors and on three open medical image datasets show that CMISR is superior to MISR in reconstruction performance and is particularly suited to medical images with strong edges or intense contrast.

Spelling error correction is the task of identifying and rectifying misspelled words in texts. It is a potential and active research topic in Natural Language Processing because of numerous applications in human language understanding. The phonetically or visually similar yet semantically distinct characters make it an arduous task in any language. Earlier efforts on spelling error correction in Bangla and resource-scarce Indic languages focused on rule-based, statistical, and machine learning-based methods which we found rather inefficient. In particular, machine learning-based approaches, which exhibit superior performance to rule-based and statistical methods, are ineffective as they correct each character regardless of its appropriateness. In this paper, we propose a novel detector-purificator-corrector framework, DPCSpell based on denoising transformers by addressing previous issues. In addition to that, we present a method for large-scale corpus creation from scratch which in turn resolves the resource limitation problem of any left-to-right scripted language. The empirical outcomes demonstrate the effectiveness of our approach, which outperforms previous state-of-the-art methods by attaining an exact match (EM) score of 94.78%, a precision score of 0.9487, a recall score of 0.9478, an f1 score of 0.948, an f0.5 score of 0.9483, and a modified accuracy (MA) score of 95.16% for Bangla spelling error correction. The models and corpus are publicly available at https://tinyurl.com/DPCSpell., Comment: 29 pages, 4 figures, and 9 tables

Flash memory experiences adverse effects due to radiation. These effects can be raised in terms of doping, feature size, supply voltages, layout, shielding. The the operating point shift of the device forced to enter the logically-undefined region and cause upset and data errors under radiation exposure. In this letter, the threshold voltage shift of the floating gate transistor (FGT) is analyzed by a mathematical model.

Floating gate transistor is the basic building block of non-volatile flash memory, which is one of the most widely used memory gadgets in modern micro and nano electronic applications. Recently there has been a surge of interest to introduce a new generation of memory devices using graphene nanotechnology. In this paper we present a new floating gate transistor (FGT) design based on multilayer graphene nanoribbon (MLGNR) and carbon nanotube (CNT). In the proposed FGT a multilayer structure of graphene nanoribbon (GNR) would be used as the channel of the field effect transistor (FET) and a layer of CNTs would be used as the floating gate. We have performed an analysis of the charge accumulation mechanism in the floating gate and its dependence on the applied terminal voltages. Based on our analysis we have observed that proposed graphene based floating gate transistor could be operated at a reduced voltage compared to conventional silicon based floating gate devices. We have presented detail analysis of the operation and the programming and erasing processes of the proposed FGT, dependency of the programming and erasing current density on different parameters, impact of scaling the thicknesses of the control and tunneling oxides. These analysis are done based on the capacitance model of the device.

Flash memory based on floating gate transistor is the most widely used memory technology in modern microelectronic applications. We recently proposed a new concept of multilayer graphene nanoribbon (MLGNR) and carbon nanotube (CNT) based floating gate transistor design for future nanoscale flash memory technology. In this paper, we analyze the tunneling current mechanism in the proposed graphene-CNT floating gate transistor. We anticipate that the proposed floating gate transistor would adopt Fowler-Nordheim (FN) tunneling during its programming and erase operations. In this paper, we have investigated the mechanism of tunneling current and the factors that would influence this current and the behavior of the proposed floating gate transistor. The analysis reveals that FN tunneling is a strong function of the high field induced by the control gate, and the thicknesses of the control oxide and the tunnel oxide., Comment: in IEEE SOCC, Las Vegas, USA, 2014