On January 16, CCADD was selected as the lead research institution for the 2025 Contract Research and Development Project by the Ministry of Food and Drug Safety (MFDS). The project aims to develop a large-scale Korean-English corpus and translation model specialized in the pharmaceutical field. The total research funding amounts to KRW 1 billion (approximately USD 692,900). For this project, CCADD will collaborate with the Structure & Knowledge Injection into Machine Learning lab (SKIML) at the Graduate School of Data Science, Seoul National University (PI: Prof. Jaeyoon Lee) and the Korea Regulatory Science Center (PI: Dr. Sungkoo Son).

As part of this research, the team will develop a high-quality, domain-specific Korean-English corpus, ensuring consistent translation quality. The corpus will be rigorously validated and evaluated by experts to maintain translation accuracy and reliability. Furthermore, the team will develop a -level Korean-English neural machine translation (NMT) model and a large language model (LLM) in the pharmaceutical and medical fields based on this corpus.

CCADD and its research team officially launched the project with an initiation meeting on February 13 with MFDS. During the meeting, key stakeholders established research objectives and outlined a strategic roadmap for the next two years.

The team members, David Seung U Lee, Yujin Kim, and Prof. Howard Lee anticipate this research will establish Korea’s first high-quality pharmaceutical corpus and develop an advanced translation model processing pharmaceutical information accurately. With expertise in regulatory science, drug development, and AI, CCADD aims to successfully complete the project and contribute to advancements in pharmaceutical and medical fields.

Yoomin Jeon and Siun Kim were awarded the Best Oral Presentation Award at the 2021 KSCPT Conference.The Best Oral Presentation Award aims to recognize outstanding scientific papers delivered through oral presentations at the conference. The awards were given to oral presentations of the paper titled "A Randomized, Double-blind, Placebo- and Active-controlled, Escalating Single-dose Study to Evaluate Safety, Tolerability, Pharmacokinetic, and Pharmacodynamic Profiles of Subcutaneous Eflapegrastim in Healthy Japanese and Caucasians" and "An Annotated Corpus and Natural Language Processing Model for Drug Safety Information Extraction from Free-texts Reporting Adverse Events" for Yoomin and Siun, respectively. 

Yoomin presented the results of a phase 1 clinical trial evaluating the pharmacokinetics and pharmacodynamics of eflapegrastim following subcutaneous administration to healthy subjects. Eflapegrastim (Rolontis®) is a novel, long-acting Granulocyte-Colony Stimulating Factor (G-CSF) being studied as a treatment for neutropenia that utilizes Hanmi's Labscovery, a proprietary platform technology. 

Siun presented an annotated corpus for extracted Korean adverse events using free-texts from the KAERS (Korean Adverse Event Reporting System) database. To this end, Siun Kim developed deep learning models that automatically extract medication and adverse event information, cooperating with MILAB in Seoul National University.

Prof. Howard Lee and Yoomin Jeon attended the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP 2018) held at the Square Meeting Center in Brussels from October 31^st through November 4^th, 2018. EMNLP is one of the top-tier conferences in the area of natural language processing organized by SIGDAT, the Association for Computational Linguistics (ACL) special interest group on linguistic data and corpus-based approaches to NLP. With the current artificial intelligence and machine learning boom from a variety of disciplines, this year's conference had a record-breaking number of submissions (2,231) and 2,500+ attendees, a 48% increase compared with EMNLP 2017. Of all the long and short papers submitted, only 549 papers were accepted, which can be found in the EMNLP proceedings. The conference included two days of workshops, followed by three days of main conference sessions including long and short presentations, tutorials, and demos. Researchers and practitioners from around the world gathered to share their latest research findings and emerging trends in machine learning for NLP, machine translation, language models, text mining, information extraction, and many more. 

In the 'Writing Code for NLP Research' tutorial, researchers from Allen AI Institute shared their best practices learned from designing the development of the Allen NLP toolkit, a PyTorch-based open-source library for deep learning NLP research. It was an excellent tutorial to learn practical advice on writing research code and to recognize the importance of having good quality codes. In the 'Deep Latent Variable Models of Natural Language' tutorial, deep latent variable models were covered, which are emerging as a useful tool in NLP applications. Mainly, three types of latent variable models (discrete, continuous, and structured) and inference strategies (exact gradient, sampling, and conjugacy) over the latent variables were examined. 

There were a number of presentations/papers in the clinical domain. One of the interesting papers, titled 'emrQA: A Large Corpus for Question Answering on Electronic Medical Records', proposed a novel semi-automated framework to generate domain-specific large-scale question answering (QA) corpus by leveraging expert annotations on clinical notes for various NLP tasks from the i2b2 challenge datasets. For medical QA on EMR, they released 400K+ question-answer evidence pairs and 1M question-logical forms, a representation useful for generating corpus. Generating patient-specific QA from an EMR is an important NLP task to process a natural language question and provide a human level accuracy answer. Among many datasets that were presented at the conference, a clinical domain dataset MedNLI was included. MedNLI is a Natural Language Inference (NLI) dataset on the medical history of patients annotated by physicians, which we can later refer in our research.