Recently, various text-to-image generative models have been released, demonstrating their ability to generate high-quality synthesized images from text prompts. Despite these advancements, determining the appropriate text prompts to obtain desired images remains challenging. The quality of the synthesized images heavily depends on the user input, making it difficult to achieve consistent and satisfactory results. This limitation has sparked the need for an effective prompt optimization method to generate optimized text prompts automatically for text-to-image generative models. Thus, this study proposes a prompt optimization method that uses in-context few-shot learning in a pretrained language model. The proposed approach aims to generate optimized text prompts to guide the image synthesis process by leveraging the available contextual information in a few text examples. The results revealed that synthesized images using the proposed prompt optimization method achieved a higher performance, at 18% on average, based on an evaluation metric that measures the similarity between the generated images and prompts for generation. The significance of this research lies in its potential to provide a more efficient and automated approach to obtaining high-quality synthesized images. The findings indicate that prompt optimization may offer a promising pathway for text-to-image generative models.

The foundational technique of code similarity detection, which underpins plagiarism detection tools, has already reached a level of maturity where it can be effectively employed for practical applications, demonstrating commendable performance. However, although the understanding of code clones—referred to as similar codes—has evolved, there has been a noticeable decline in the emergence of novel proposals for code similarity detection techniques. The landscape of code similarity detection techniques is diverse and can be divided based on how codes are represented. Each method, designed to cater to different types of detectable code similarity instances, has distinct advantages and drawbacks. Therefore, the selection of an appropriate method is crucial and is contingent on the specific objectives of the analysis. This paper provides a comprehensive exploration of code similarity detection techniques and illuminates the prevailing trends in plagiarism detection research. It acquaints readers with a spectrum of distinct code similarity detection methods, accompanied by the requisite contextual background knowledge. Additionally, it presents a detailed overview of the trajectory of research trends in plagiarism detection.

Low-light image enhancement is a key technique to overcome the quality degradation of photos taken under challenging illumination conditions. Even though the significant progress has been made for enhancing the poor visibility, the intrinsic noise amplified in low-light areas still remains as an obstacle for further improvement in visual quality. In this paper, a novel and simple method for low-light image enhancement is proposed. Specifically, the subspace, which has an ability to separately reveal illumination and noise, is constructed from a group of similar image patches, so-called volume , at each pixel position. Based on the principal energy analysis onto this volume-based subspace, the illumination component is accurately inferred from a given image while the unnecessary noise is simultaneously suppressed. This leads to clearly unveiling the underlying structure in low-light areas without loss of details. Experimental results show the efficiency and robustness of the proposed method for low-light image enhancement compared to state-of-the-art methods.

Multi-task learning is a computationally efficient method to solve multiple tasks in one multi-task model, instead of multiple single-task models. MTL is expected to learn both diverse and shareable visual features from multiple datasets. However, MTL performances usually do not outperform single-task learning. Recent MTL methods tend to use heavy task-specific heads with large overheads to generate task-specific features. In this work, we (1) validate the efficacy of MTL in low-data conditions with early-exit architectures, and (2) propose a simple feature filtering module with minimal overheads to generate task-specific features. We assume that, in low-data conditions, the model cannot learn useful low-level features due to the limited amount of data. We empirically show that MTL can significantly improve performances in all tasks under low-data conditions. We further optimize the early-exit architecture by a sweep search on the optimal feature for each task. Furthermore, we propose a feature filtering module that selects features for each task. Using the optimized early-exit architecture with the feature filtering module, we improve the 15.937% in ImageNet and 4.847% in Places365 under the low-data condition where only 5% of the original datasets are available. Our method is empirically validated in various backbones and various MTL settings.

AutoPaperBench proposes a benchmark generation system to automatically evaluate the comprehensibility of papers in a Multimodal Large Language Model (MLLM). The proposed system efficiently structures the content of a paper through semantic parsing and automatically generates text-based QAs and visual-based VQAs. To ensure the quality of the generated QA, we introduce a reviewer system that evaluates six criteria such as logic and appropriateness. In our experiments on 60 research papers from the medical, natural, and engineering fields, the generated benchmarks demonstrate comparable performance rankings to those of previous benchmarks, and the performance improvements achieved through semantic parsing are validated. The system can run on a single GPU environment and provides a framework for efficiently evaluating LLM thesis comprehension.

Overloaded vehicles such as large cargo trucks tend to cause large traffic accidents. Such traffic accidents often bring high mortality rates, including injuries and deaths, and cause fatal damage to road structures such as roads and bridges. Therefore, there is a vicious circle in which a lot of the budgets is spent for accident restoration and road maintenance. It is important to control overloaded vehicles that are around roads in urban areas. However, it often takes a lot of manpower to track down on overloaded vehicles at appropriate interception points during a specific time. Moreover, the drivers tend to avoid interception by bypassing the interception point, while exchanging interception information with each other. In this work, the main bridges in a city are chosen as the interception point. Since installing vehicle-weighing devices on the road surface is expensive and the devices cause frequent faults after the installation, inexpensive general-purpose Internet of Things (IoT) sensors, such as acceleration and gyroscope sensors, are installed on the bridges. First, assuming that the sensing value of the overloaded vehicle is different from the nonoverloaded vehicle, we investigate the difference in the sensed values between the overloaded and nonoverloaded vehicles. Then, based on the hypothesis, we propose a new method to identify prime time zones with overloaded vehicles. Technically, the proposed method comprises two steps. In the first step, we propose a new bridge traffic classification model using Bidirectional Long Short-Term Memory (Bi–LSTM) that automatically classifies time series data to either high or low traffic condition. The Bi–LSTM model has higher accuracy than existing neural network models because it has a symmetric neural network structure, by which input information can be processed in forward and backward directions. In the second step, we propose a new method of automatically identifying top-k time zones with many overloaded vehicles under the high traffic condition. It first uses the k-Nearest Neighbor (NN) algorithm to find the sensing value, most similar to the actual sensing value of the overloaded vehicle, in the high traffic cluster. According to the experimental results, there is a high difference of the sensing values between the overloaded and the nonoverloaded vehicle, through statistical verification. Also, the accuracy of the proposed method in the first step is ~75%, and the top-k time zones in which overloaded vehicles are crowded are identified automatically.