Li-Chun Lu

I am an incoming PhD student at National Taiwan University (NTU) Robot Learning Lab, in collaboration with Professor Shao-Hua Sun. I will receive my B.S.E. in Electrical Engineering from NTU in July 2025. My research interests lie in the creativity of Large Language Models (LLMs) and human-AI collaboration. I am honored to have worked with Professor Shao-Hua Sun and Professor Hung-yi Lee, where I designed a multi-agent discussion framework incorporating role-play techniques to enhance LLM creativity.

I also conducted research as a visiting researcher at the University of California, Los Angeles (UCLA), in Professor Nanyun Peng’s PLUS Lab (Nov 2024 – Apr 2025), where I led a project analyzing and comparing creativity metrics across domains and identifying limitations that underscore the need for improved evaluation methods.

News

Jun 2025	I will begin my Ph.D. program at NTU’s Robot Learning Lab, working with Professor Shao-Hua Sun, starting in September 2025!
Nov 2024	I have started working as a visiting scholar at UCLA in the Computer Science Department, under Professor Nanyun Peng’s PLUS Lab, for a six-month period.
Jul 2024	Our paper LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play is accepted by COLM 2024. Join us during the poster session on the afternoon of October 7th in Philadelphia! Click to check out our work!
Jun 2024	I’ve started working as a Research Assistant at National Taiwan University’s Robot Learning Lab.
May 2024	Yu Lun Hsu and I presented our work, AnimalSense: Understanding Beyond-human Sensory Capabilities of Animals via VR Games, at CHI 2024 Student Games Competition and got a Runner Up Award.

Selected Publications

COLM ’24
LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play

Li-Chun Lu^*, Shou-Jen Chen^*, Tsung-Min Pai, and 3 more authors

2024

Abs arXiv Bib Code

Large language models (LLMs) have shown exceptional proficiency in natural language processing but often fall short of generating creative and original responses to open-ended questions. To enhance LLM creativity, our key insight is to emulate the human process of inducing collective creativity through engaging discussions with participants from diverse backgrounds and perspectives. To this end, we propose LLM Discussion, a three-phase discussion framework that facilitates vigorous and diverging idea exchanges and ensures convergence to creative answers. Moreover, we adopt a role-playing technique by assigning distinct roles to LLMs to combat the homogeneity of LLMs. We evaluate the efficacy of the proposed framework with the Alternative Uses Test, Similarities Test, Instances Test, and Scientific Creativity Test through both LLM evaluation and human study. The results show that our proposed framework outperforms single-LLM approaches and existing multi-LLM frameworks across various creativity metrics.
@misc{lu2024llmdiscussionenhancingcreativity, title = {LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play}, author = {Lu, Li-Chun and Chen, Shou-Jen and Pai, Tsung-Min and Yu, Chan-Hung and Lee, Hung-yi and Sun, Shao-Hua}, year = {2024}, eprint = {2405.06373}, archiveprefix = {arXiv}, primaryclass = {cs.CL}, url = {https://arxiv.org/abs/2405.06373}, }