Da-Chen Lian

連大成

About

I am a Ph.D. candidate in Linguistics at National Taiwan University. My interests include language-model interpretability and LLM research more broadly. I have passed my oral defense and expect to complete the degree in 2027.

I serve as Director of Research Computing at NTU’s Co-Intelligence Humanities AI Future Lab (CHAI). I am also a member and system administrator at LOPE.

Doctoral study
Ph.D. Candidate in Linguistics
Dissertation defended · Degree expected 2027
Roles
Director of Research Computing · CHAI
Member & System Administrator · LOPE
Profiles

Research

My current interests center on large language models: understanding how they work and finding useful ways to apply them. My work approaches these questions through model evaluation, language resources, and applied systems.

Interpretability and evaluation

I am interested in understanding and evaluating language-model behavior. My work includes benchmarks and analyses of reasoning, bias, semantic relations, and word sense, with interpretability as a developing direction.

Language resources and representation

My work includes corpora, datasets, and computational representations for Chinese, English, and Formosan languages.

Applied language models

I am interested in practical systems for corpus research, education, domain adaptation, and other language-related workflows.

Publications

  1. 2026 Balancing accuracy and efficiency: Evaluating encoder- and decoder-based models for word sense disambiguation and regular polysemy detection Pin-Er Chen, Da-Chen Lian, Shu-Kai Hsieh Natural Language Processing · First View Journal
  2. 2026 CorPilot: An Agentic Framework for Corpus Linguistics Da-Chen Lian, Mao-Chang Ku, Shu-Kai Hsieh Intelligent Computing · Computing Conference Chapter
  3. 2026 When Structure Matters: Cross-Lingual Hyperbolic Embeddings for Chinese and English Wordnets Mao-Chang Ku, Da-Chen Lian, Pin-Er Chen, Po-Ya Angela Wang, Wei-Ling Chen, Shu-Kai Hsieh Language Resources and Evaluation Conference · LREC Paper
  4. 2025 LOBSTER: Linguistics Olympiad Benchmark for Structured Evaluation on Reasoning Da-Chen Lian, Ri-Sheng Huang, Pin-Er Chen, Chunki Lim, You-Kuan Lin, Guan-Yu Tseng, Zhen-Yu Lin, Pin-Cheng Chen, Shu-Kai Hsieh ROCLING 2025 Paper
  5. 2025 Empowering Elementary Learning: Utilizing Large Language Models to Craft Tailored Textbooks with Expert Insight Da-Chen Lian, Mao-Chang Ku, Po-Ya Angela Wang, Wei-Ling Chen, Shu-Kai Hsieh Journal of Library and Information Studies Journal
  6. 2025 Continual Pre-Training is (not) What You Need in Domain Adaption Pin-Er Chen, Da-Chen Lian, Shu-Kai Hsieh, et al. arXiv:2504.13603 Preprint
  7. 2024 The Semantic Relations in LLMs: An Information-theoretic Compression Approach Yu-Hsiang Tseng, Pin-Er Chen, Da-Chen Lian, Shu-Kai Hsieh NeusymBridge · LREC-COLING 2024 Paper
  8. 2024 Self-supervised learning for Formosan speech representation and linguistic phylogeny Shu-Kai Hsieh, Yu-Hsiang Tseng, Da-Chen Lian, Chi-Wei Wang Frontiers in Language Sciences Journal
View the ORCID record

Profile

Ph.D. Candidate in Linguistics
Director of Research Computing

Graduate Institute of Linguistics
National Taiwan University

  1. Ph.D. Candidate in Linguistics
    Dissertation defended · Degree expected 2027
    National Taiwan University

  2. Ph.D. Candidate in Networking and Multimedia
    National Taiwan University

  3. M.A. in Linguistics
    National Taiwan University

  4. B.A. in English
    National Taipei University of Technology