As multimedia content often contains noise from intrinsic defects of digital devices, image denoising is an important step for high-level vision recognition tasks. Although several studies have developed the denoising field employing advanced Transformers, these networks are too momory-intensive for real-world applications. Additionally, there is a lack of research on lightweight denosing (LWDN) with Transformers. To handle this, we provide seven comparative baseline Transformers for LWDN, serving as a foundation for future research. We also demonstrate the parts of randomly cropped patches significantly affect the denoising performances during training. While previous studies have overlooked this aspect, we aim to train our baseline Transformers in a truly fair manner. Furthermore, we conduct empirical analyses of various components to determine the key considerations for constructing LWDN Transformers. Codes are available at https://github.com/rami0205/LWDN.

We propose a novel stock movement prediction model that learns Multi-Time Contexts and Randomness (MTC-R). We assume that the stock price movements are affected by (1) combination of time-based momentums and (2) tradings by noise traders which cause randomness in a stock market. MTC-R has three main procedures for modeling our hypothesis. First, the model learns time representations using time2vec and encodes the multi-time views using a GRU network. Second, MTC-R generates the multi-time contexts using a multi-head attention mechanism and inserts the randomness. Third, a loss function is designed to learn temporal differences between the results of the second and the current time-embedded vector. Our model improves the prediction results in terms of accuracy and the Matthews correlation coefficient on six benchmark datasets compared to baseline models. Furthermore, MCT-R shows the effectiveness of the prediction results using cumulative returns in portfolio trading simulations.

With the proliferation of the Internet and the huge amount of data it transfers, text summarization is becoming more important. We present an approach to the design of an automatic text summarizer that generates a summary by extracting sentence segments. First, sentences are broken into segments by special cue markers. Each segment is represented by a set of predefined features (e.g. location of the segment, average term frequencies of the words occurring in the segment, number of title words in the segment, and the like). Then a supervised learning algorithm is used to train the summarizer to extract important sentence segments, based on the feature vector. Results of experiments on U.S. patents indicate that the performance of the proposed approach compares very favorably with other approaches (including Microsoft Word summarizer) in terms of precision, recall, and classification accuracy.