| CARVIEW |
Select Language
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Wed, 15 Oct 2025 13:11:08 GMT
access-control-allow-origin: *
strict-transport-security: max-age=31556952
etag: W/"68ef9d6c-4569"
expires: Mon, 29 Dec 2025 20:42:56 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: 0A58:2D64E0:93F71F:A610EA:6952E576
accept-ranges: bytes
age: 0
date: Mon, 29 Dec 2025 20:32:56 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210070-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1767040376.399387,VS0,VE221
vary: Accept-Encoding
x-fastly-request-id: bfb09b3c18a677d82a22fbdb6fac6e580b5c0612
content-length: 3255
Junyeob Baek
Junyeob Baek
I am a first-year Ph.D. student at KAIST working with Prof. Sungjin Ahn. Previously, I worked at Mila-Quebec AI Institute as a visiting student, collaborating with Yoshua Bengioโs group and LawZero team members. My research goal lies in building lifelong developmental agents, but safety-garuanteed. To achieve this, I am particularly interested in world modeling, interpretability, open-ended learning, and the alignment problem. Prior to my Ph.D., I received my MS from KAIST under the supervision of Prof. Sungjin Ahn. Before entering academia, I worked for 4.5 years in industry, gaining practical experience in robotics at Syscon Robotics and AI at Makinarocks.
News
- May 2025: Iโll be joining Yoshua Bengioโs group at Mila as a Visiting Student this summer in Montreal! ๐จ๐ฆ
- Jan 2025: One paper has been accepted to ICLR 2025! ๐
- Nov 2024: Successfully defended my Masterโs thesis and embarking on my PhD journey at KAIST!
- May 2024: One paper has been accepted to ICML 2024! ๐
Publications
Discrete JEPA: Learning Discrete Token Representations without Reconstruction
ICML 2025 Tokenization Workshop (TokShop)
Services
- Workshop Organizer: 2nd KAIST-Mila Prefrontal AI Workshop (6 invited speakers incl. Y. Bengio)
- Reviewer: ICLR 2025