Digital Library[ Search Result ]
Hallucination Detection and Explanation Model for Enhancing the Reliability of LLM Responses
Sujeong Lee, Hayoung Lee, Seongsoo Heo, Wonik Choi
http://doi.org/10.5626/JOK.2025.52.5.404
Recent advancements in large language models (LLMs) have achieved remarkable progress in natural language processing. However, reliability issues persist due to hallucination, which remains a significant challenge. Existing hallucination research primarily focuses on detection, lacking the capability to explain the causes and context of hallucinations. In response, this study proposes a hallucination-specialized model that goes beyond mere detection by providing explanations for identified hallucinations. The proposed model was designed to classify hallucinations while simultaneously generating explanations, allowing users to better trust and understand the model’s responses. Experimental results demonstrated that the proposed model surpassed large-scale models such as Llama3 70B and GPT-4 in hallucination detection accuracy while consistently generating high-quality explanations. Notably, the model maintained stable detection and explanation performance across diverse datasets, showcasing its adaptability. By integrating hallucination detection with explanation generation, this study introduces a novel approach to evaluating hallucinations in language models.
CLS Token Additional Embedding Method Using GASF and CNN for Transformer based Time Series Data Classification Tasks
Jaejin Seo, Sangwon Lee, Wonik Choi
http://doi.org/10.5626/JOK.2023.50.7.573
Time series data refer to a sequentially determined data set collected for a certain period of time. They are used for prediction, classification, and outlier detection. Although existing artificial intelligence models in the field of time series are mainly based on the Recurrent Neural Network, recent research trends are changing to transformer based models. Although these transformer based models show good performance for time series data prediction problem, they show relatively insufficient performance for classification tasks. In this paper, we propose an embedding method to add special classification tokens generated using Gramian Angular Summation Field and Convolution Neural Network to utilize time series data as input to transformers and found that we could leverage the pre-trained method to improve performance. To show the efficacy of our method, we conducted extensive experiments with 12 different models using the University of California, Riverside dataset. Experimental results show that our proposed model improved the average accuracy of 85 datasets from 1.4% to up to 21.1%.
Graph-based Wi-Fi Radio Map Construction and Update Method
http://doi.org/10.5626/JOK.2017.44.6.643
Among Wi-Fi based indoor positioning systems, fingerprinting localization is the most common technique with high precision. However, construction of the initial radio map and the update process require considerable labor and time effort. To address this problem, we propose an efficient method that constructs the initial radio map at each vertex based on a graph. In addition, we introduce a method to update the radio map automatically by mapping signal data acquired from users to the reference point created on each edge. Since the proposed method collects signal data manually only at the vertex of the graph to build the initial radio map and updates it automatically, our proposed method can dramatically reduce labor and time effort, which are the disadvantages of the conventional fingerprinting method. In our experimental study, we show validity of our radio map update method by comparing with the actual reference point data. We also show that our proposed method is able to construct the radio map with an accuracy of about 3.5m by automatically updating the radio map.
IFC-based Data Structure Design for Web Visualization
When using IFC data consisting of STEP schema based on the EXPRESS language, it is not easy for collaborating project stakeholders to share BIM modeling shape information. The IFC viewer application must be installed on the desktop PC to review the BIM modeling shape information defined within the IFC, because the IFC viewer application not only parse STEP structure information model but also process the 3D feature construction for a 3D visualization. Therefore, we propose a lightweight data structure design for web visualization by parsing IFC data and constructing 3D modeling data. Our experimental results show the weight reduction of IFC data is about 40% of original file size and the web visualization is able to see the same quality with all web browsers which support WebGL on PCs and smartphones. If applied research is conducted about the web visualization based on IFC data of the last construction phase, it could be utilized in various fields ranging from the facility maintenance to indoor location-based services.
An Energy-Aware Cooperative Communication Scheme for Wireless Multimedia Sensor Networks
Jeong-Oh Kim, Hyunduk Kim, Wonik Choi
Numerous clustering schemes have been proposed to increase energy efficiency in wireless sensor networks. Clustering schemes consist of a hierarchical structure in the sensor network to aggregate and transmit data. However, existing clustering schemes are not suitable for use in wireless multimedia sensor networks because they consume a large quantity of energy and have extremely short lifetime. To address this problem, we propose the Energy-Aware Cooperative Communication (EACC) method which is a novel cooperative clustering method that systematically adapts to various types of multimedia data including images and video. An evaluation of its performance shows that the proposed method is up to 2.5 times more energy-efficient than the existing clustering schemes.
Parallel Range Query Processing with R-tree on Multi-GPUs
Hongsu Ryu, Mincheol Kim, Wonik Choi
Ever since the R-tree was proposed to index multi-dimensional data, many efforts have been made to improve its query performances. One common trend to improve query performance is to parallelize query processing with the use of multi-core architectures. To this end, a GPU-base R-tree has been recently proposed. However, even though a GPU-based R-tree can exhibit an improvement in query performance, it is limited in its ability to handle large volumes of data because GPUs have limited physical memory. To address this problem, we propose MGR-tree (Multi-GPU R-tree), which can manage large volumes of data by dividing nodes into multiple GPUs. Our experiments show that MGR-tree is up to 9.1 times faster than a sequential search on a GPU and up to 1.6 times faster than a conventional GPU-based R-tree.
Fast Hilbert R-tree Bulk-loading Scheme using GPGPU
In spatial databases, R tree is one of the most widely used indexing structures and many variants have been proposed for its performance improvement. Among these variants, Hilbert R tree is a representative method using Hilbert curve to process large amounts of data without high cost split techniques to construct the R tree. This Hilbert R tree, however, is hardly applicable to large scale applications in practice mainly due to high pre processing costs and slow bulk load time. To overcome the limitations of Hilbert R tree, we propose a novel approach for parallelizing Hilbert mapping and thus accelerating bulk loading of Hilbert R tree on GPU memory. Hilbert R tree based on GPU improves bulk loading performance by applying the inversed cell method and exploiting parallelism for packing the R tree structure. Our experimental results show that the proposed scheme is up to 45 times faster compared to the traditional CPU based bulk loading schemes.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr