no code implementations • 31 Oct 2024 • Eugene Jang, Kimin Lee, Jin-Woo Chung, Keuntae Park, Seungwon Shin
In this work, we investigate incomplete tokens, i. e., undecodable tokens with stray bytes resulting from byte-level byte-pair encoding (BPE) tokenization.
no code implementations • 18 Oct 2024 • Hanna Kim, Minkyoo Song, Seung Ho Na, Seungwon Shin, Kimin Lee
Recent advancements in Large Language Models (LLMs) have established them as agentic systems capable of planning and interacting with various tools.
1 code implementation • 25 Sep 2024 • Minkyoo Song, Hanna Kim, Jaehan Kim, Youngjin Jin, Seungwon Shin
Recent advances in natural language processing and the increased use of large language models have exposed new security vulnerabilities, such as backdoor attacks.
no code implementations • 21 Sep 2024 • Jaehan Kim, Minkyoo Song, Seung Ho Na, Seungwon Shin
Parameter-efficient fine-tuning (PEFT) has become a key training strategy for large language models.
no code implementations • 15 Mar 2024 • Eugene Jang, Jian Cui, Dayeon Yim, Youngjin Jin, Jin-Woo Chung, Seungwon Shin, YongJae lee
We use our domain-customized methodology to train CyBERTuned, a cybersecurity domain language model that outperforms other cybersecurity PLMs on most tasks.
no code implementations • 15 May 2023 • Youngjin Jin, Eugene Jang, Jian Cui, Jin-Woo Chung, YongJae lee, Seungwon Shin
Recent research has suggested that there are clear differences in the language used in the Dark Web compared to that of the Surface Web.
no code implementations • 2 Feb 2023 • Na Hyeon Park, Hanna Kim, Chanhee Lee, Changhoon Yoon, Seunghyeon Lee, Youngjin Jin, Seungwon Shin
NFT (Non-fungible Token) has drastically increased in its size, accounting for over \$16. 9B of total market capitalization.
no code implementations • NAACL 2022 • Youngjin Jin, Eugene Jang, YongJae lee, Seungwon Shin, Jin-Woo Chung
By leveraging CoDA, we conduct a thorough linguistic analysis of the Dark Web and examine the textual differences between the Dark Web and the Surface Web.
no code implementations • 13 Sep 2021 • Jian Cui, Kwanwoo Kim, Seung Ho Na, Seungwon Shin
We then propose Meta-Path instance encoding and aggregation methods to capture the temporal information of user engagement and produce news representation end-to-end.
no code implementations • 3 Dec 2020 • Cristian R. Constante-Amores, Lyes Kahouadji, Assen Batchvarov, Seungwon Shin, Jalel Chergui, Damir Juric, Omar K. Matar
The thinning of the lobes induces the creation of holes which expand to form liquid threads that undergo capillary breakup to form droplets.
Fluid Dynamics