Hello, I'm chanmuzi
I'm a super passionate AI researcher/engineer from Korea 🇰🇷
- 🧑🏻💻 [Upstage] AI Research Engineer (2024.05 ~ Present)
- 🔎 [LILAB] NLP Research Intern, Chung Ang University (2024.01 ~ 2024.04)
- 🤖 [DIYA] 5th member of NLP team of AI Society (2023.01 ~ Present)
- 💼 [Dacon] Data Science Team manager (2023.05 ~ 2023.12)
- 🏕️ [Google] Machine Learning Bootcamp 4th (2023.09 ~ 2023.11)
- 🌏 [OSSCA] Open Source Contribution Academy, PyTorch and PyTorch Lightning mentee (2023.07 ~ 2023.10)
- 🧑🏻💻 [Boostcamp AI Tech] 4th NLP track completion (2022.09 ~ 2023.02)
- 🎓 [University of Seoul] Bachelor of Arts in English Language and Literature (2022.08)
- ❤️ I love NLP, especially (s)LLM.
- 📫 How to reach me: chanmuzi@naver.com
open/close
- RAG-based Law Specialized Chatbot Model / Chung Ang University - Mar. 2024 ~ April. 2024
- Establishing baselines for instruction tuning of LLMs
- Researching evaluation methodologies for LLMs
- Development of Educational Chatbot Services Using LLM / Dacon - Oct. 2023 ~ Dec. 2023
- Leading the integration of chatbots in machine learning educational content
- Creating AI services for quantitative assessments adn qualitative feedback, incorporating API-based solutions
- AI vs. Human Text Discrmination Competition Design & Management / Dacon - Sep. 2023 ~ Nov. 2023
- Initiated and managed a competition on AI-generated content detection using the OpenAI API
- Crafted a novel dataset by modifying colloquial Twitter texts for diverse epressions
- Established a baseline and explored various solution strategies
- Visual Question Answering (VQA) / Dacon - Jun. 2023 ~ Aug. 2023
- Orchestrated a VQA competition, handling dataset preprocessing and competition structuring
- multi-modal task, accuracy metric
- Reviewed and verified competition methodologies and outcomes, ensuring fairness and data integrity
- Document Visual Question Answering (DocVQA) - Dec. 2022 ~ Feb. 2023
- Collaborated in a corporate-linked project with UpStage
- Addressed the challenge of aligning visual (bbox) and linguistic information (text) in DocVQA tasks
- Developed attention maps to visually represent model focus areas, addressing weakness in diagram and figure interpretation
In addition to the above list, I have also undertaken the following NLP tasks:
- ChatGPT Prompt Engineering Competition - Apr. 2023
- Open Domain Question Answering (ODQA) - Dec. 2022
- building a Korean Law Dataset for Relation Extraction - Nov. 2022
- Relation Extraction (RE) - Oct. 2022
- Semantic Text Similarity (STS) - Oct. 2022
These projects have significantly enhanced my foundational skills and knowledge in Natural Language Processing (NLP).
