-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.html
306 lines (275 loc) · 17.6 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
<!DOCTYPE html>
<html lang="en" dir="ltr">
<head>
<link href="css/style.css" rel="stylesheet" type="text/css" media="all">
<link rel="script" href="js/script.js">
<link rel="icon" type="image/x-icon" href="img/favicon.ico">
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.min.css">
<title>Taishi Nakamura portfolio</title>
</head>
<body>
<nav>
<ul>
<li><a href="#about">About</a></li>
<li><a href="#education">Education</a></li>
<li><a href="#research">Research experience</a></li>
<li><a href="#industrial">Industrial experience</a></li>
<li><a href="#publications">Publications</a></li>
<li><a href="#contact">Contact</a></li>
</ul>
</nav>
<div class="centering-block">
<div class="centering-block-inner">
<div class="profile-header">
<img src="./img/profile.jpg" alt="Taishi Nakamura" class="profile-image">
<div class="profile-info">
<h1 id="about">Taishi Nakamura</h1>
<div class="text-container">
<p>
I love training large-scale models and am passionate about scalable neural network architectures. I am fascinated by the computing power that enables this, particularly distributed computing systems. My interests extend to building multimodal systems, developing advanced reasoning capabilities, and creating agents that continually evolve.
</p>
</div>
</div>
</div>
<h2 id="education"> Education </h2>
<h3>Oct 2024 - Present Institute of Science Tokyo</h3>
<div class="logo-text">
<img src="./img/science-tokyo.jpg" alt="Institute of Science Tokyo logo">
<div class="text-container">
<p>Master of Science in Computer Science</p>
<p>Supervisor: Professor <a href="https://www.rio.gsic.titech.ac.jp/en/member/yokota.html">Rio Yokota</a></p>
<p><i>Note: Formerly Tokyo Institute of Technology, renamed after merging with Tokyo Medical and Dental University in October 2024</i></p>
</div>
</div>
<h3>Apr 2021 - Sep 2024 Tokyo Institute of Technology</h3>
<div class="logo-text">
<img src="./img/tokyotech.webp" alt="Tokyo Tech logo">
<div class="text-container">
<p>Master of Science in Computer Science (Apr 2024 - Sep 2024, continued at Institute of Science Tokyo)</p>
<p>Bachelor of Science in Computer Science (Apr 2021 - Mar 2024)</p>
<p>Supervisor: Professor <a href="https://www.rio.gsic.titech.ac.jp/en/member/yokota.html">Rio Yokota</a></p>
<p>Graduated one year early from the bachelor's program due to outstanding academic performance</p>
</div>
</div>
<h2 id="research">Research experience</h2>
<h3>Jun 2024 - Present RIKEN</h3>
<div class="logo-text">
<img src="./img/riken.jpeg" alt="RIKEN logo">
<div class="text-container">
<p>
Research Assistant
</p>
</div>
</div>
<h3>May 2024 - Present NII</h3>
<div class="logo-text">
<img src="./img/nii.jpeg" alt="Sakana AI logo">
<div class="text-container">
<p>
Research Assistant at National Institute of Informatics Large-Scale Language Model Research and Development Center
</p>
<p>
Mentors: Professor <a href="https://www.fai.cds.tohoku.ac.jp/members/js/">Jun Suzuki</a> and Professor <a href="https://www.rio.gsic.titech.ac.jp/en/member/yokota.html">Rio Yokota</a></p>
</p>
</div>
</div>
<h3>Feb 2024 - Present Sakana AI</h3>
<div class="logo-text">
<img src="./img/sakana-ai.jpeg" alt="Sakana AI logo">
<div class="text-container">
<p>
Research Internship
</p>
<p>
Mentor: Dr. <a href="https://takiba.net/">Takuya Akiba</a> (Research Scientist)</p>
</p>
</div>
</div>
<h3>Oct 2023 - Apr 2024 LLM-JP</h3>
<div class="logo-text">
<img src="./img/llm-jp.png" alt="llm-jp logo">
<div class="text-container">
<p>
Research Internship
</p>
<p>
Active as a member of the Model Building WG
</p>
<p>
Mentor: Professor <a href="https://www.fai.cds.tohoku.ac.jp/members/js/">Jun Suzuki</a></p>
</p>
</div>
</div>
<h3>Apr 2022 - Aug 2023 A*Quantum</h3>
<div class="logo-text">
<img src="./img/aq.jpeg" alt="A*Quantum logo">
<div class="text-container">
<p>
Research Internship
</p>
</div>
</div>
<h2 id="industrial">Industrial Experience</h2>
<h3>Jun 2023 - Feb 2024 MITOU TARGET</h3>
<div class="logo-text">
<img src="./img/mitou-tg.png" alt="MITOU TARGET logo">
<div class="text-container">
<p>Selected for the <a href="https://www.ipa.go.jp/en/it-talents/mitou/target-quantum-computing-2023.html">MITOU TARGET program for Quantum Computing</a>.</p>
<p>Developed an educational platform for quantum computing, now available at <a href="https://qualsimu.com/textbook/">Qualsimu Textbook</a>.</p>
<p>
Mentor: Dr. <a href="https://www.rd.ntt/e/organization/researcher/special/s_023.html">Yuuki Tokunaga</a></p>
</p>
</div>
</div>
<h3>Feb 2022 - Dec 2022 Crystal Method</h3>
<div class="logo-text">
<img src="./img/crystal.jpeg" alt="crystal method logo">
<div class="text-container">
<p>
Engineering Internship
</p>
</div>
</div>
<h2>Open Source Projects </h2>
<h3>Oct 2023 - Present Swallow LLM</h3>
<div class="logo-text">
<img src="./img/tokyotech-llm.png" alt="TokyoTech-LLM logo">
<div class="text-container">
<p>
This project contributes to Japan's AI sovereignty efforts and involves collaboration with esteemed researchers including
Professor <a href="https://www.chokkan.org/">Naoaki Okazaki</a>,
Professor <a href="https://www.rio.gsic.titech.ac.jp/en/member/yokota.html">Rio Yokota</a>,
and Dr. <a href="https://sites.google.com/view/hjtakamura">Hiroya Takamura</a>.
</p>
<p>
We invite you to explore our progress:
<br>
• <a href="https://huggingface.co/tokyotech-llm">View our published models on Hugging Face</a>
<br>
• <a href="https://swallow-llm.github.io/index.en.html">Learn more about the project on our website</a>
</p>
</div>
</div>
<h3>Jun 2023 - Present Ontocord.AI</h3>
<div class="logo-text">
<img src="./img/ontocord.png" alt="Ontocord.AI logo">
<div class="text-container">
<p>
We've developed Aurora-M, a multilingual model designed to address the challenges of low-resource languages. Our latest release, aurora-m-v0.1-biden-harris-redteamed, is the first open-source multilingual model red-teamed against the Biden-Harris Executive Order on AI.
</p>
<p>
Through these projects, we're actively contributing to the open science community, promoting research that aims to reduce illegal and biased AI outputs while enhancing the utility of AI applications across multiple domains and languages.
</p>
<p>
Explore our work: <a href="https://huggingface.co/aurora-m/aurora-m-v0.1">Aurora-M v0.1 on Hugging Face</a>
</p>
</div>
</div>
<h2 id="publications">Publications</h2>
<h3>COLING '2025 (Industry Track) Aurora-M: The First Open Source Multilingual Language Model Red-teamed according to the U.S. Executive Order</h3>
<div class="paper-info">
<p>
<b>Taishi Nakamura*</b>, Mayank Mishra*, Simone Tedeschi*, ..., Matthew Blumberg, Victor May, Huu Nguyen, Sampo Pyysalo (49 authors)
<a href="https://arxiv.org/abs/2404.00399" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
</div>
<h3>COLM '2024 Continual Pre-Training for Cross-Lingual LLM Adaptation: Enhancing Japanese Language Capabilities</h3>
<div class="paper-info">
<p>
Kazuki Fujii∗, <b>Taishi Nakamura∗</b>, Mengsay Loem, Hiroki Iida, Masanari Ohi, Kakeru Hattori, Hirai Shota, Sakae Mizuki, Rio Yokota, Naoaki Okazaki
<a href="https://arxiv.org/abs/2404.17790" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
<p>
Acceptance Rate: 28.8%
</p>
</div>
<h3>COLM '2024 Building a Large Japanese Web Corpus for Large Language Models</h3>
<div class="paper-info">
<p>
Naoaki Okazaki, Kakeru Hattori, Hirai Shota, Hiroki Iida, Masanari Ohi, Kazuki Fujii, <b>Taishi Nakamura</b>, Mengsay Loem, Rio Yokota, Sakae Mizuki
<a href="https://arxiv.org/abs/2404.17733" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
<p>
Acceptance Rate: 28.8%
</p>
</div>
<h2>Preprints</h2>
<h3>2024 Agent Skill Acquisition for Large Language Models via CycleQD</h3>
<div class="paper-info">
<p>
So Kuroki, <b>Taishi Nakamura</b>, Takuya Akiba, Yujin Tang
</p>
<p>
<a href="https://arxiv.org/pdf/2410.14735" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
</div>
<h3>2024 LLM-jp: A Cross-organizational Project for the Research and Development of Fully Open Japanese LLMs</h3>
<div class="paper-info">
<p>
LLM-jp: Akiko Aizawa, Eiji Aramaki, ..., <b>Taishi Nakamura</b>, ..., Koichiro Yoshino (79 authors)
</p>
<p>
<a href="https://arxiv.org/abs/2407.03963" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
<p>
Note: Authors are listed in alphabetical order.
</p>
</div>
<h2>Domestic Conference in Japan</h2>
<h3>IPSJ 2024 LLMに日本語テキストを学習させる意義</h3>
<div class="paper-info">
<p>
齋藤 幸史郎, 水木 栄, 大井 聖也, <b>中村 泰士</b>, 塩谷 泰平, 前田 航希, Ma Youmi, 服部 翔, 藤井 一喜, 岡本 拓己, 石田 茂樹, 高村 大也, 横田 理央, 岡崎 直観
<a href="https://sites.google.com/sig-nl.ipsj.or.jp/sig-nl/%E7%A0%94%E7%A9%B6%E7%99%BA%E8%A1%A8%E4%BC%9A/NL261" target="_blank"><i class="fa fa-external-link" aria-hidden="true"></i></a>
</p>
<span>🎉 Best Paper Award (6.7%).</span>
</div>
<h3>IPSJ 2024 継続学習を用いた効率の良いマルチリンガル・マルチエキスパートモデルの開発</h3>
<div class="paper-info">
<p>
<b>中村泰士</b>, 横田理央
<a href="https://www.ipsj.or.jp/event/taikai/86/WEB/data/pdf/5J-01.html" target="_blank"><i class="fa fa-external-link" aria-hidden="true"></i></a>
</p>
</div>
<h3>NLP 2024 継続事前学習による日本語に強い大規模言語モデルの構築</h3>
<div class="paper-info">
<p>
藤井一喜∗, <b>中村泰士∗</b>, Mengsay Loem, 飯田大貴, 大井聖也, 服部翔, 平井翔太, 水木栄, 横田理央, 岡崎直観
<a href="https://www.anlp.jp/proceedings/annual_meeting/2024/pdf_dir/A8-5.pdf" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
<p>
<a href="https://www.anlp.jp/nlp2024/award.html#outstanding" target="_blank">
<span>🎉 Outstanding Papers (2.0%).</span>
</a>
</p>
</div>
<h3>NLP 2024 大規模言語モデルの日本語能力の効率的な強化: 継続事前学習における語彙拡張と対訳コーパスの活用</h3>
<div class="paper-info">
<p>
水木栄*, 飯田大貴*, 藤井一喜, <b>中村泰士</b>, Mengsay Loem, 大井聖也, 服部翔, 平井翔太, 横田理央, 岡崎直観
<a href="https://www.anlp.jp/proceedings/annual_meeting/2024/pdf_dir/A6-4.pdf" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
</div>
<h3>NLP 2024 Swallowコーパス: 日本語大規模ウェブコーパス</h3>
<div class="paper-info">
<p>
岡崎直観, 服部翔, 平井翔太, 飯田大貴, 大井聖也, 藤井一喜, <b>中村泰士</b>, Mengsay Loem, 横田理央, 水木栄
<a href="https://www.anlp.jp/proceedings/annual_meeting/2024/pdf_dir/A6-1.pdf" target="_blank"><i class="fa fa-file-pdf-o" aria-hidden="true"></i></a>
</p>
<a href="https://www.anlp.jp/nlp2024/award.html#outstanding" target="_blank">
<span>🎉 Outstanding Papers (2.0%).</span>
</a>
</div>
<h2 id="contact"> Contact </h2>
<a href="https://twitter.com/Setuna7777_2" target="_blank" rel="noreferrer" class="sns"><img src="https://raw.githubusercontent.com/danielcranney/readme-generator/main/public/icons/socials/twitter.svg" width="32" height="32" /></a>
<a href="https://www.github.com/Taishi-N324" target="_blank" rel="noreferrer"><img src="https://raw.githubusercontent.com/danielcranney/readme-generator/main/public/icons/socials/github.svg" width="32" height="32" /></a>
<a href="https://www.linkedin.com/in/taishi-nakamura" target="_blank" rel="noreferrer"><img src="https://raw.githubusercontent.com/danielcranney/readme-generator/main/public/icons/socials/linkedin.svg" width="32" height="32" /></a>
<a href="https://scholar.google.com/citations?hl=en&user=nbPQwgUAAAAJ" target="_blank" rel="noreferrer">
<img src="./img/google-scholar.svg" width="32" height="32" alt="Google Scholar" />
</a>
<P></p>
<br>Last updated: October 2024
</div>
</div>
</body>
</html>