Author name cluster

Junghoon Kim

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

7 papers

2 author rows

TIST Journal 2026 Journal Article

A New DAWN for Fake News Detection: Exploiting Engagement Earliness for Temporality-aware Evaluation

Junghoon Kim
Junmo Lee
Yeonjun In
Kanghoon Yoon
Chanyoung Park

Social graph-based fake news detection aims to identify news articles containing false information by utilizing social contexts, e.g., user information, tweets and comments. However, conventional methods are evaluated under less realistic scenarios, where the model has access to future knowledge on article-related and context-related data during training. In this work, we newly formalize a more realistic evaluation scheme that mimics real-world scenarios, where the data is temporality-aware and the detection model can only be trained on data collected up to a certain point in time. We show that the discriminative capabilities of conventional methods decrease sharply under this new setting, and further propose DAWN, a method more applicable to such scenarios. Our empirical findings indicate that later engagements (e.g., consuming or reposting news) contribute more to noisy edges that link real news-fake news pairs in the social graph. Motivated by this, we utilize feature representations of engagement earliness to guide an edge weight estimator to suppress the weights of such noisy edges, thereby enhancing the detection performance of DAWN. Through extensive experiments, we demonstrate that DAWN outperforms existing fake news detection methods under real-world environments. The source code for DAWN is available at https://github.com/LeeJunmo/DAWN.

Details DOI

ICML Conference 2023 Conference Paper

Robust Non-Linear Feedback Coding via Power-Constrained Deep Learning

Junghoon Kim
Taejoon Kim
David J. Love
Christopher G. Brinton

The design of codes for feedback-enabled communications has been a long-standing open problem. Recent research on non-linear, deep learning-based coding schemes have demonstrated significant improvements in communication reliability over linear codes, but are still vulnerable to the presence of forward and feedback noise over the channel. In this paper, we develop a new family of non-linear feedback codes that greatly enhance robustness to channel noise. Our autoencoder-based architecture is designed to learn codes based on consecutive blocks of bits, which obtains de-noising advantages over bit-by-bit processing to help overcome the physical separation between the encoder and decoder over a noisy channel. Moreover, we develop a power control layer at the encoder to explicitly incorporate hardware constraints into the learning optimization, and prove that the resulting average power constraint is satisfied asymptotically. Numerical experiments demonstrate that our scheme outperforms state-of-the-art feedback codes by wide margins over practical forward and feedback noise regimes, and provide information-theoretic insights on the behavior of our non-linear codes. Moreover, we observe that, in a long blocklength regime, canonical error correction codes are still preferable to feedback codes when the feedback noise becomes high. Our code is available at https: //anonymous. 4open. science/r/RCode1.

Details

AAAI Conference 2022 Conference Paper

No Task Left Behind: Multi-Task Learning of Knowledge Tracing and Option Tracing for Better Student Assessment

Suyeong An
Junghoon Kim
Minsam Kim
JuneYoung Park

Student assessment is one of the most fundamental tasks in the field of AI Education (AIEd). One of the most common approach to student assessment is Knowledge Tracing (KT), which evaluates a student’s knowledge state by predicting whether the student will answer a given question correctly or not. However, in the context of multiple choice (polytomous) questions, conventional KT approaches are limited in that they only consider the binary (dichotomous) correctness label (i. e. , correct or incorrect), and disregard the specific option chosen by the student. Meanwhile, Option Tracing (OT) attempts to model a student by predicting which option they will choose for a given question, but overlooks the correctness information. In this paper, we propose Dichotomous- Polytomous Multi-Task Learning (DP-MTL), a multi-task learning framework that combines KT and OT for more precise student assessment. In particular, we show that the KT objective acts as a regularization term for OT in the DP-MTL framework, and propose an appropriate architecture for applying our method on top of existing deep learning-based KT models. We experimentally confirm that DP-MTL significantly improves both KT and OT performances, and also benefits downstream tasks such as Score Prediction (SP).

PDF Details

ICRA Conference 2018 Conference Paper

Disturbance Observer Based Linear Feedback Controller for Compliant Motion of Humanoid Robot

Mingon Kim
Junghoon Kim
Sanghyun Kim
Jaehoon Sim
Jaeheung Park

Actuator modules of humanoid robots have relatively higher joint elasticity than those of industrial robots. Such joint elasticity could lead to negative effects on both the tracking performance and stability for walking. Especially, unstable contact between the foot and ground caused by joint elasticity is a critical problem, as it decreases the stability of position-controlled humanoid robots. To address this problem, this paper introduces a novel control scheme for position-controlled humanoid robots by which we can obtain not only enhance compliance capability for unknown contact but also suppress the vibration caused by joint elasticity. To estimate the disturbance caused by external forces and modeling errors between the actual system and nominal system, a disturbance observer based estimator is designed at each joint. Furthermore, a linear feedback controller for the flexible joint model and a gravity compensator is considered to reduce vibration and deflection due to the joint elasticity. The proposed control scheme was implemented on our humanoid robot, DYROS-JET, and its performance was demonstrated by improved stability during dynamic walking and stepping on objects.

Details

IROS Conference 2016 Conference Paper

Novel apparatus for light touch threshold measurement

Junghoon Kim
Bum-Jae You
Youngjin Choi

All the people have their own individual force ranges to feel the sense of touch and its minimum value is called light touch threshold (in short LTT). For instance, if a touching force is very weak, people do not know whether the object is contacted to their skin, even though it is pushing the skin already. This paper presents a novel apparatus to measure the LTT felt by each subject, which is referred to as Active von Frey (AvF) in this paper. As far as the authors know, the LTT measurement device is for the first time proposed in this paper. It is possible for the AvF to provide the touching force ranges from 1[mg f ] to 400[mg f ]. In order to provide an accurate touching force for the subject, D'Arsonval movement is chosen as an actuator to rotate a touching AvF pin. Both an electric current applied to the D'Arsonval movement and a rotational angle of AvF pin are utilized to calculate the touching force by the electro-mechanical statics. In addition, the image processing technique is used to measure the rotational angle of AvF pin. Finally we show that the LTTs are very different from individual to individual through the experimental results of 10 participants. We hope that the personalized parameters is used to design or control haptic or tele-operation devices for various applications.

Details

YNIMG Journal 2008 Journal Article

Structural consequences of diffuse traumatic brain injury: A large deformation tensor-based morphometry study

Junghoon Kim
Brian Avants
Sunil Patel
John Whyte
Branch H. Coslett
John Pluta
John A. Detre
James C. Gee

Traumatic brain injury (TBI) is one of the most common causes of long-term disability. Despite the importance of identifying neuropathology in individuals with chronic TBI, methodological challenges posed at the stage of inter-subject image registration have hampered previous voxel-based MRI studies from providing a clear pattern of structural atrophy after TBI. We used a novel symmetric diffeomorphic image normalization method to conduct a tensor-based morphometry (TBM) study of TBI. The key advantage of this method is that it simultaneously estimates an optimal template brain and topology preserving deformations between this template and individual subject brains. Detailed patterns of atrophies are then revealed by statistically contrasting control and subject deformations to the template space. Participants were 29 survivors of TBI and 20 control subjects who were matched in terms of age, gender, education, and ethnicity. Localized volume losses were found most prominently in white matter regions and the subcortical nuclei including the thalamus, the midbrain, the corpus callosum, the mid- and posterior cingulate cortices, and the caudate. Significant voxel-wise volume loss clusters were also detected in the cerebellum and the frontal/temporal neocortices. Volume enlargements were identified largely in ventricular regions. A similar pattern of results was observed in a subgroup analysis where we restricted our analysis to the 17 TBI participants who had no macroscopic focal lesions (total lesion volume >1. 5 cm3). The current study confirms, extends, and partly challenges previous structural MRI studies in chronic TBI. By demonstrating that a large deformation image registration technique can be successfully combined with TBM to identify TBI-induced diffuse structural changes with greater precision, our approach is expected to increase the sensitivity of future studies examining brain–behavior relationships in the TBI population.

Details DOI

YNIMG Journal 2006 Journal Article

Continuous ASL perfusion fMRI investigation of higher cognition: Quantification of tonic CBF changes during sustained attention and working memory tasks

Junghoon Kim
John Whyte
Jiongjiong Wang
Hengyi Rao
Kathy Z. Tang
John A. Detre

Arterial spin labeling (ASL) perfusion fMRI is an emerging method in clinical neuroimaging. Its non-invasiveness, absence of low frequency noise, and ability to quantify the absolute level of cerebral blood flow (CBF) make the method ideal for longitudinal designs or low frequency paradigms. Despite the usefulness in the study of cognitive dysfunctions in clinical populations, perfusion activation studies to date have been conducted for simple sensorimotor paradigms or with single-slice acquisition, mainly due to technical challenges. Using our recently developed amplitude-modulated continuous ASL (CASL) perfusion fMRI protocol, we assessed the feasibility of a higher level cognitive activation study in twelve healthy subjects. Taking advantage of the ASL noise properties, we were able to study tonic CBF changes during uninterrupted 6-min continuous performance of working memory and sustained attention tasks. For the visual sustained attention task, regional CBF increases (6–12 ml/100 g/min) were detected in the right middle frontal gyrus, the bilateral occipital gyri, and the anterior cingulate/medial frontal gyri. During the 2-back working memory task, significantly increased activations (7–11 ml/100 g/min) were found in the left inferior frontal/precentral gyri, the left inferior parietal lobule, the anterior cingulate/medial frontal gyri, and the left occipital gyrus. Locations of activated and deactivated areas largely concur with previous PET and BOLD fMRI studies utilizing similar paradigms. These results demonstrate that CASL perfusion fMRI can be successfully utilized for the investigation of the tonic CBF changes associated with high level cognitive operations. Increased applications of the method to the investigation of cognitively impaired populations are expected to follow.

Details DOI