-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
413 lines (393 loc) · 21.9 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
<!DOCTYPE HTML>
<html lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta name="author" content="Jungwoo Kim">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Jungwoo Kim</title>
<link rel="shortcut icon" href="images/favicon.ico" type="image/x-icon">
<link rel="stylesheet" type="text/css" href="style.css">
</head>
<body>
<!-- Header Section (Name and Profile) -->
<table style="width:100%;max-width:800px;margin:auto;">
<tr>
<td style="padding:0px;">
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:2.5%;width:63%;vertical-align:middle">
<p class="name" style="text-align: center; font-weight: bold;">Jungwoo Kim</p>
<br>
<p>
My name is Jungwoo Kim.
I am an Integrated M.S./Ph.D student at the <a href="https://mcml.yonsei.ac.kr/">MCML Group</a>, under the supervision of Prof.
<a href="https://mcml.yonsei.ac.kr/professor/">Prof. Jong-Seok Lee</a>.
</p>
<p>
I have a broad interest in various research areas that utilize machine learning and computer vision.
Especially, I'm currently interested in <strong>image compression for machine (ICM)</strong> and <strong>evaluation of diffusion-based generative models</strong>.
</p>
<p style="text-align:center">
<a href="mailto:[email protected]">Email</a> /
<a href="data/Curriculum_Vitae_1011.pdf">CV</a> /
<a href="https://scholar.google.com/citations?user=gusyMHMAAAAJ&hl=en">Google Scholar</a> /
<a href="https://www.linkedin.com/in/jungwoo-kim-44a700288/">LinkedIn</a> /
<a href="https://github.com/kjungwoo03">Github</a>
</p>
<br>
<p style="text-align:center; font-size: 1.1em; font-weight: bold; margin-top: 10px;">
<em>Open to any collaboration opportunities across related research topics!</em>
</p>
</td>
<td style="padding:2.5%;width:40%;max-width:40%">
<a href="images/kjungwoo.png">
<img style="width:100%;max-width:100%;object-fit: cover; border-radius: 50%;" alt="profile photo" src="images/kjungwoo2.png">
</a>
</td>
</tr>
</table>
<!-- Education Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Education</h2>
<p>
<strong>Yonsei University</strong>, Seoul, Republic of Korea<br>
<em>M.S./Ph.D student in School of Integrated Technology</em><br>
<em>Mar. 2025 - Present</em><br>
<!-- <em style="display: block; text-align: right;">* One year early graduation.</em> -->
</p>
<p>
<strong>Yonsei University</strong>, Seoul, Republic of Korea<br>
<em>B.S. in School of Integrated Technology (GPA: 3.95 / 4.30)</em><br>
<em>Mar. 2022 – Feb. 2025 </em><br>
<em style="display: block; text-align: right;">* One year early graduation.</em>
</p>
</td>
</tr>
</table>
<!-- Research Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Research Interests</h2>
<p>
My primary research goal is <strong>to fully understand how deep learning models work</strong> and <strong>to fully utilize them in various research areas</strong>, especially in computer vision.
So My primary research interest just lies in various topics about machine learning and computer vision, covering my goal.<br><br>
Currently, I'm interested in <strong>Image Compression, Explainable AI, Generative Models</strong>, etc. In the past, I have also studied Graph Neural Network (GNN), Reinforcement Learning (RL) and Natural Language Processing (NLP), especially about evaluating the performance of large language models.
</p>
</td>
</tr>
</table>
<!-- Publication Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Publications</h2>
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/hae-rae.png" alt="HAE-RAE Bench" width="160" height="160">
</td>
<td width="75%" valign="middle">
<a href="https://aclanthology.org/2024.lrec-main.704">
<span class="papertitle">HAE-RAE Bench: Evaluation of Korean Knowledge in Language Models</span>
</a>
<br>
Guijin Son,
Hanwool Lee,
Suwan Kim, Huiseo Kim, Jaecheol Lee,
Je Won Yeom,
Jihyu Jung,
<strong><u>Jungwoo Kim</u></strong>, Songseong Kim
<br>
<em>Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)</em>
<p>
HAE-RAE Bench evaluates <strong>large language models' understanding of Korean culture</strong>, offering a challenge beyond traditional English-based benchmarks.
</p>
<p>
<em> Our Dataset is available at <a href="https://huggingface.co/datasets/HAERAE-HUB/HAE_RAE_BENCH_1.1">HAERAE-HUB</a>.</em>
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/k2-eval.png" alt="K2-EVAL" width="160" height="160">
</td>
<td width="75%" valign="middle">
<a href="https://openreview.net/forum?id=BAtM7tcMrm">
<span class="papertitle">K2-EVAL: Harnessing the Evaluation of Linguistic Fluency and Ethnolinguistic Knowledge in Korean</span>
</a>
<br>
Guijin Son, Hyunwoo Ko, Hoyoung Lee, Seunghyeok Hong, Yewon Kim, <strong><u>Jungwoo Kim</u></strong>
<br>
<em>ACL 2024 Workshop, The 2nd Workshop on Cross-Cultural Considerations in NLP (C3NLP 2024)</em>
<p>
K2-EVAL introduces a benchmark with 90 handwritten instructions that test knowledge of <strong>Korean language and culture</strong>, including 5400 human annotations to evaluate 32 LLMs. It also examines if GPT-4 can replace human evaluations.
</p>
</td>
</tr>
</table>
</td>
</tr>
</table>
<!-- Projects Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Projects</h2>
<table style="width:100%;margin:auto;">
<!-- Projects -->
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/petnow.png" alt="3D Length Estimation with Dog Data" width="160" height="160">
</td>
<td width="75%" valign="middle">
<!-- <a href="https://meteor-microwave-835.notion.site/24-1-DSL-x-WesomE-08a468c84ef94ed3855a464fff58fb32"> -->
<span class="papertitle">3D Length Estimation with Dog Data</span>
</a>
<p>
<em>This work is part of 24-2 Yonsei DSL Corporate Project with <a href="https://www.petnow.io/ko">PetNow</a>.</em>
</p>
<p>
Improving a dog body length estimation model. Based on pose estimation and depth estimation models, we improve <strong>2d-to-3d transformation</strong> with camera intrinsic parameters. Also, we utilize 3D Reconstruction-based method for length estimation.
</p>
<p>
<em>Due to security issues, the materials cannot be disclosed. Please contact me if needed.</em>
</p>
</td>
</tr>
<!-- Projects -->
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/choreography.png" alt="Audio Sentiment Classification" width="160" height="160">
</td>
<td width="75%" valign="middle">
<a href="https://github.com/kjungwoo03/24_2_DSL_Modeling_CV1">
<span class="papertitle">Group Choreography Data Crawling Framework</span>
</a>
<p>
<em>This work, <strong>NextLevel2: Neural Extraction and Tracking Framework for K-POP Group Choreography</strong> is part of 24-2 Yonsei DSL Modeling Project.</em>
</p>
<p>
We focus on the research gap that existing group choreography dataset collection methods require manual intervention, and propse a fully-automatic data collection framework.
This approach is expected to contribute to the scalability of group choreography dataset creation.
</p>
<p>
<em>Our presentation video is available <a href="https://www.youtube.com/watch?v=sxWe0EJ5Eck&t=1073s">here</a>. Code and Github will be fully released soon.</em>
</p>
</td>
</tr>
<!-- Projects -->
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/audio-sentiment.png" alt="Audio Sentiment Classification" width="160" height="160">
</td>
<td width="75%" valign="middle">
<a href="https://meteor-microwave-835.notion.site/24-1-DSL-x-WesomE-08a468c84ef94ed3855a464fff58fb32">
<span class="papertitle">Audio Sentiment Classification</span>
</a>
<p>
<em>This work is part of 24-1 Yonsei DSL Corporate Project with WesomE.</em>
</p>
<p>
Developed a lightweight model (under 500MB) for recognizing emotions from speaker's voice data, focusing on feature augmentation through analyzing <strong>high-frequency details</strong> of Fast Fourier Transform.
</p>
<p>
<em>Due to security issues, the materials cannot be disclosed. Please contact me if needed.</em>
</p>
</td>
</tr>
<!-- Projects -->
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/prompt.png" alt="Prompt Engineering" width="160" height="160">
</td>
<td width="75%" valign="middle">
<strong>Prompt Engineering in Image Generation Model</strong>
<p>
<em>This work, <strong>InferPrompt: Reframing Prompt Engineering with BLIP-2</strong>, is part of 24-1 Yonsei DSL Modeling Project.</em>
</p>
<p>
Prompt engineering in text-to-image generation model. We tried to improve the quality of generated image, by <strong>style transferring the inferred prompt</strong>. By adding re-questioning block, we could improve the detail quality of generated image.
<p>
<em>Our presentation video is available <a href="https://www.youtube.com/watch?v=XKE6IjfMUjg&t=4645s">here</a>. Code and Github will be released soon.</em>
</p>
</td>
</tr>
<!-- Projects -->
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/next-level.png" alt="Prompt Engineering" width="160" height="160">
</td>
<td width="75%" valign="middle">
<strong>NextLevel: A Choreography Generation Model with Lyric Understanding</strong>
<p>
<em>This work is part of 23-2 Software Project (IIT4204) with Jaeho Jin, Sangjoon Yeo, and advised by Prof. <a href="https://sites.google.com/site/jeonggilko">JeongGil Ko</a>.</em>
</p>
<p>
We tried to represent lyrics into group choreography generation model. By <strong>incorporating text embedding into previous model</strong>, we could generate lyric-aware dance motions.
</p>
<p>
<em>Our materials aren't disclosed now. Please contact me if needed.</em>
</p>
</td>
</tr>
<!-- Projects -->
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/red-cnn.png" alt="Medical Image Super Resolution" width="160" height="160">
</td>
<td width="75%" valign="middle">
<a href="https://github.com/kjungwoo03/SUPER_RESOLUTION_RED_CNN">
<span class="papertitle">Medical Image Super Resolution</span>
</a>
<p>
<em>Implementation of <a href="https://arxiv.org/abs/1702.00288">Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network</a>. This work is a part of the project in <a href="https://sites.google.com/view/yonsei-medisyslab/home">Medisys Lab</a>, advised by Prof. <a href="https://sites.google.com/view/yonsei-medisyslab/members/professor">Jongduk Baek</a>.</em>
</p>
<p>
The research focus on <strong>enhancing low-dose CT images by super-resolving</strong> them to match the quality of normal-dose images. Additionally, as a follow-up study, we replaced the original architecture with a U-Net model to further evaluate the effectiveness in preserving image details.
</p>
</td>
</tr>
<!-- Projects -->
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/coronary.png" alt="Acute Coronary Syndrome Detection" width="160" height="160">
</td>
<td width="75%" valign="middle">
<a href="https://github.com/kjungwoo03/2023-ECG_CHALLENGE">
<span class="papertitle">Acute Coronary Syndrome Detection on Wearable Device</span>
</a>
<p>
<em><strong>Won an Award for Excellence</strong> at the 2023 Yonsei Medical Convergence Challenge (with ChangeDae Lee, Hyelim Kim, YoungJun Na)</em>
</p>
<p>
Developed a ResNet-based deep learning model for <strong>diagnosing acute coronary syndrome using wearable devices</strong>, focusing on detecting the condition using ECG signals from a limited number of leads, optimized for real-world application on wearable platforms.
</p>
<p>
<em>Our presentation template is available <a href="data/ecg_final_template.pdf">here</a>.</em>
</p>
</td>
</tr>
</table>
</td>
</tr>
</table>
<!-- Honors and Awards Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Honors and Awards</h2>
<p>
<strong>Certification, 4th LG Aimers/Data Intelligence</strong><br>
<em>LG Aimers (Advancing AI for Young Talents): Online AI Education and Hackaton (2024.01.02. ~ 2024.02.26.) held by LG AI Research.</em>
</p>
<p>
<strong>Excellence Award, 2023 Yonsei Medical Convergence Challenge</strong><br>
<em>Yonsei Medical Convergence Challenge (2023.01.30. ~ 2023.01.31.) held by The Medical Scientist Training Program. (Undergraduate Course)</em>
</p>
<!-- <p> -->
<!-- <strong>Multiple Academic Scholarships</strong><br> -->
<!-- <em>Detailed information about specific scholarships can be found <a href='data/scholarship_1011.pdf'>here</a>.</em> -->
<!-- </p> -->
</td>
</tr>
</table>
<!-- Academic Service Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Academic Services</h2>
<p>
<em>TBD</em>
</p>
</td>
</tr>
</table>
<!-- Talk Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Talks</h2>
<p>
<strong>Normalizing Flow and Energy Based Model</strong><br>
<em>24-2 Regular Session Speech in Yonsei DSL (video available <a href="https://youtu.be/9PG_iqlH3J8?si=SVaR3PiKTenUfLo7">here</a>) - Sep 03, 2024</em><br>
</p>
<p>
<strong>Mamba Review: Linear-Time Sequence Modeling with Selective State Spaces</strong><br>
<em>24-2 Regular Session Speech in Yonsei DSL (video available <a href="https://youtu.be/Vm6oFFXoP74?si=9HtctY6BKEQqEVYt">here</a>) - Aug 22, 2024</em><br>
</p>
<p>
<strong>Life as a Undergraduate Student in Engineering Field</strong><br>
<em>(Invited) Yeungnam High School, Daegu, Republic of Korea - Mar 16, 2024</em><br>
</p>
</td>
</tr>
</table>
<!-- Teaching Experience Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Teaching Experiences</h2>
<p>
<strong>Computational Thinking and SW Programming (YCS1001)</strong><br>
<em>Winter 2024, Fall 2024, Summer 2024, Spring 2024, Spring 2023</em><br>
</p>
<p>
<strong>Mechatronics Project (IIT4312)*</strong><br>
<em>Spring 2024</em><br>
<em style="display: block; text-align: right;">* Via tutoring program hosted by Yonsei University.</em>
</p>
</td>
</tr>
</table>
<!-- Miscellanea Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<h2>Miscellanea</h2>
<p>
<strong>Yonsei Data ScienceLab</strong><br>
<em>11th Regular Member (Dec. 2023 - Present)</em><br>
<em>Head of Academic Team (Jun. 2024 - Dec. 2024)</em><br>
<a href="https://www.notion.so/Data-Science-Lab-dcc313da82774e7cb62fe76c2ad09d86">Yonsei Data Science Lab (DSL)</a> is a student community under the Department of Applied Statistics at Yonsei University, advised by Prof. <a href="https://dslab-with.github.io/web/#dslab-about">Taeyoung Park</a>.
Yonsei DSL focus on studying and applying various theories related to Data Science and Machine Learning, based on a statistical theory.
</p>
<p>
<strong>ElutherAI</strong><br>
<em>Project Member (Sep. 2023 - Nov. 2023)</em><br>
<a href="https://www.eleuther.ai/">Eluther AI</a> is a non-profit AI research lab founded in 2020. ElutherAI focuses on the interpretability and alignment of Large Language Models.
</p>
<p>
<strong>Yonsei Computer Club</strong><br>
<em>Regular Member (Sep. 2022 - Present)</em><br>
<em>Member of Friendship Team (Jul. 2024 - Dec. 2024)</em><br>
<a href="https://www.ycc.club/">Yonsei Computer Club (YCC)</a>, founded in 1970, is the only central computer club at Yonsei University. YCC brings together students with a shared interest in computers and supports a variety of academic activities.
</p>
<p>
<strong>Yonsei Engineering Student Council</strong><br>
<em>Executive Member (Apr. 2022 - Nov. 2023)</em><br>
<em>Freshman Vice Representative (Mar. 2022 - Feb. 2023)</em><br>
</p>
</td>
</td>
</tr>
</table>
<!-- Footer Section -->
<table style="width:100%;margin:auto;">
<tr>
<td style="padding:20px;">
<p style="text-align:center;font-size:small;">
Last updated on Jan 03, 2025.<br><br>
© 2025 Jungwoo Kim. All rights reserved. Design and source code adapted from
<a href="https://jonbarron.info" style="font-size:small;">Jon Barron's website</a>.
</p>
</td>
</tr>
</table>
</td>
</tr>
</table>
</body>
</html>