Chung-En Sun

Chung-En Sun

Ph.D. Student, Computer Science, UC San Diego

Google Scholar Scholar GitHub GitHub CV CV

About Me

I am a Ph.D. student in Computer Science at the University of California, San Diego, advised by Prof. Tsui-Wei (Lily) Weng. My research focuses on large language models, reasoning, and AI agents, with the goal of making these systems more capable, reliable, and efficient. I study how models reason, use tools, and make decisions, and develop methods to understand and improve their behavior through training, representation engineering, and targeted model editing. My recent work spans tool-use efficiency, reasoning quality, interpretability, and AI alignment and safety. Feel free to reach out about research collaborations, opportunities, or shared interests!

Education

News

Publications

First-Author Conference Papers

LLM Agents Already Know When to Call Tools -- Even Without Reasoning [code] [project website] [dataset]
Chung-En Sun, Linbo Liu, Ge Yan, Zimo Wang, Tsui-Wei Weng.
NeurIPS 2026.
Steer2Edit: From Activation Steering to Component-Level Editing [code]
Chung-En Sun, Ge Yan, Zimo Wang, Tsui-Wei Weng.
NeurIPS 2026.
The Cold-Start Safety Gap in LLM Agents [code] [project website] [dataset] [judge model]
Chung-En Sun, Linbo Liu, Tsui-Wei Weng.
EMNLP 2026 Main.
From Detection to Refusal: Safer LLMs via Circuit-Guided Weight Scaling
Kuan-Lin Chu*, Chung-En Sun*, Tsui-Wei Weng.
EMNLP 2026 Findings. * Equal contribution.
ReFIne: A Framework for Trustworthy Large Reasoning Models with Reliability, Faithfulness, and Interpretability [code]
Chung-En Sun, Ge Yan, Akshay Kulkarni, Tsui-Wei Weng.
COLM 2026.
ThinkEdit: Interpretable Weight Editing to Mitigate Overly Short Thinking in Reasoning Models [code]
Chung-En Sun, Ge Yan, Tsui-Wei Weng.
EMNLP 2025 Main.
Concept Bottleneck Large Language Models [code]
Chung-En Sun, Tuomas Oikarinen, Berk Ustun, Tsui-Wei Weng.
ICLR 2025.
Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities [code] [featured @ Microsoft Research Blog]
Chung-En Sun, Xiaodong Liu, Weiwei Yang, Tsui-Wei Weng, Hao Cheng, Aidan San, Michel Galley, Jianfeng Gao.
NAACL 2025 Main Oral.
Breaking the Barrier: Enhanced Utility and Robustness in Smoothed DRL Agents [code]
Chung-En Sun, Sicun Gao, Tsui-Wei Weng.
ICML 2024.
Melody harmonization using orderless NADE, chord balancing, and blocked Gibbs sampling
Chung-En Sun, Yi-Wei Chen, Hung-Shin Lee, Yen-Hsing Chen, Hsin-Min Wang.
ICASSP 2021.

Conference Papers

Distance Marching for Generative Modeling
Zimo Wang, Ishit Mehta, Haolin Lu, Chung-En Sun, Ge Yan, Tsui-Wei Weng, Tzu-Mao Li.
NeurIPS 2026.
ReflCtrl: Controlling LLM Reflection via Representation Engineering
Ge Yan, Chung-En Sun, Tsui-Wei Weng.
COLM 2026.
Interpretable Generative Models through Post-hoc Concept Bottlenecks
Akshay Kulkarni, Ge Yan, Chung-En Sun, Tuomas Oikarinen, Tsui-Wei Weng.
CVPR 2025.
Effective Skill Unlearning through Intervention and Abstention
Yongce Li, Chung-En Sun, Tsui-Wei Weng.
NAACL 2025 Main.

Workshop Papers

FoCus: Improving Faithfulness in Chain-of-Thoughts by Training on Structured Reasoning Data
Guan-Yi Lin, Chung-En Sun, Tsui-Wei Weng.
NeurIPS MATH-AI Workshop 2025.
Crafting Large Language Models for Enhanced Interpretability
Chung-En Sun, Tuomas Oikarinen, Tsui-Wei Weng.
ICML MI Workshop 2024.
Fooling GPT with Adversarial In-Context Examples for Text Classification
Sudhanshu Ranjan, Chung-En Sun, Linbo Liu, Tsui-Wei Weng.
NeurIPS R0-FoMo Workshop 2023.
NTIRE 2020 Challenge on NonHomogeneous Dehazing
Codruta O. Ancuti, Cosmin Ancuti, ..., Chung-En Sun, ..., Murari Mandal.
CVPR Workshop 2020.

Experience

Work Experience

Academic Service