Weerayut Buaphet (วีรยุทธ บัวเพชร)

Ph.D. Student | NLP Researcher | VISTEC, Thailand

Summary

I'm a five-year Ph.D. scholarship student in the Natural Language Processing and Representation Learning Lab (NRL) at VISTEC, Thailand, supervised by Associate Professor Prof. Dr. Sarana.

PhD Thesis: Resource-Constrained Named Entity Recognition — Contributed a Thai-language dataset for fine-grained nested NER and a bilingual financial NER dataset for the stock market; analyzed the generalization of encoder-based and LLM-based NER models to unseen entity types and new domains; and addressed multilingual text normalization challenges in informal language.

Currently, I am focused on developing and evaluating an LLM-based retrieval-augmented generation (RAG) system for the medical domain, using supervised fine-tuning (SFT) and reinforcement learning methods (e.g., DPO, GRPO) to train models for multi-turn medical question answering.

Education

• Ph.D. in Information Science and Technology (5-year program): GPA: 4.00/4.00
Relevant coursework: Natural Language Processing, Computational Machine Intelligence and Applications.

• B.Eng. in Computer Engineering: GPA: 3.62/4.00 (Top 1)
Relevant coursework: Data Structures and Algorithms, Operating Systems, Software Engineering.

Internship

• Research Assistant: VISTEC, Rayong, Thailand (Nov 2019 – Aug 2020)
Developed a Thai Nested Named Entity Recognition (N-NER) model, ensuring accuracy through testing and analysis.

• Visiting Ph.D.: IT University of Copenhagen, Denmark (Sep 2024 – July 2025)
Conducted research on cross-lingual NER and multilingual representation learning under the supervision of Prof. Rob van der Goot.

Services

• Co-organizer:
- W-NUT 2025 (collocated with NAACL2025) and MultiLexNorm2

• Reviewers:
- ARR-EMNLP 2024, 2025