Jiabao Ji
Ph.D. Student at UCSB
jiabaoji [AT] ucsb.edu
Bio
Hi! I'm Jiabao Ji, a third-year Ph.D. student at UCSB advised by Prof. Shiyu Chang. I also work closely with Dr. Yang Zhang.
My research focuses primarily on generative AI models. Recently, I am mainly focused on following areas:
- LLM trustworthiness, such as defense methods for jailbreak attack, and efficient method for LLM unlearning.
- LLM reasoning capability, such as LLM-based multi-robot control, and LLM coding.
- Enhanced controllability for genrative models, such as the text-to-image diffusion model.
Before joining UCSB, I obtained B.Eng. in Computer Science at ACM class, Shanghai Jiao Tong University. During my undergrad, I interned at X-lance Lab at SJTU working with Prof. Kai Yu and Prof. Lu Chen. I also had a remote summer intern at SLS group working with Prof. Tianxing He and Prof. James Glass.
Publications
Most recent publications on Google
Scholar.
‡ indicates equal contribution.
HardTests: Synthesizing High-Quality Test Cases for LLM Coding
Zhongmou He‡, Yee Man Choi‡, Kexun Zhang‡, Jiabao Ji, Junting Zhou, Dejia Xu, Ivan Bercovich, Aidan Zhang, Lei Li
Arxiv'25
Collision-and Reachability-Aware Multi-Robot Control with Grounded LLM Planners
Jiabao Ji, Yongchao Chen, Yang Zhang, Ramana Rao Kompella, Chuchu Fan, Gaowen Liu, Shiyu Chang
Arxiv'25
Thinkprune: Pruning long chain-of-thought of llms via reinforcement learning
Bairu Hou, Yang Zhang, Jiabao Ji, Yujian Liu, Kaizhi Qian, Jacob Andreas, Shiyu Chang
Arxiv'25
Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference
Jiabao Ji, Yujian Liu, Yang Zhang, Gaowen Liu, Ramana Rao Kompella, Sijia Liu, Shiyu Chang
NeurIPS'24
Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing
Jiabao Ji‡, Bairu Hou‡, Alexander Robey‡, George J Pappas, Hamed Hassani, Yang Zhang, Eric Wong, Shiyu Chang
Arxiv'24
Advancing the Robustness of Large Language Models through Self-Denoised Smoothing
Jiabao Ji‡, Bairu Hou‡, Zhen Zhang‡, Guanhua Zhang‡, Wenqi Fan, Qing Li, Yang Zhang, Gaowen Liu, Sijia Liu, Shiyu Chang
NAACL'24
Improving Diffusion Models for Scene Text Editing with Dual Encoders
Jiabao Ji‡, Guanhua Zhang‡, Zhaowen Wang, Bairu Hou, Zhifei Zhang, Brian Price, Shiyu Chang
TMLR'23
Towards Coherent Image Inpainting Using Denoising Diffusion Implicit Models
Guanhua Zhang‡, Jiabao Ji‡, Yang Zhang, Mo Yu, Tommi Jaakkola, Shiyu Chang
ICML'23
Controlling the Focus of Pretrained Language Generation Models
Jiabao Ji, Yoon Kim, James Glass, Tianxing He
Findings of ACL'22
HardTests: Synthesizing High-Quality Test Cases for LLM Coding
Zhongmou He‡, Yee Man Choi‡, Kexun Zhang‡, Jiabao Ji, Junting Zhou, Dejia Xu, Ivan Bercovich, Aidan Zhang, Lei Li
Arxiv'25
Collision-and Reachability-Aware Multi-Robot Control with Grounded LLM Planners
Jiabao Ji, Yongchao Chen, Yang Zhang, Ramana Rao Kompella, Chuchu Fan, Gaowen Liu, Shiyu Chang
Arxiv'25
Thinkprune: Pruning long chain-of-thought of llms via reinforcement learning
Bairu Hou, Yang Zhang, Jiabao Ji, Yujian Liu, Kaizhi Qian, Jacob Andreas, Shiyu Chang
Arxiv'25
Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference
Jiabao Ji, Yujian Liu, Yang Zhang, Gaowen Liu, Ramana Rao Kompella, Sijia Liu, Shiyu Chang
NeurIPS'24
Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing
Jiabao Ji‡, Bairu Hou‡, Alexander Robey‡, George J Pappas, Hamed Hassani, Yang Zhang, Eric Wong, Shiyu Chang
Arxiv'24
Advancing the Robustness of Large Language Models through Self-Denoised Smoothing
Jiabao Ji‡, Bairu Hou‡, Zhen Zhang‡, Guanhua Zhang‡, Wenqi Fan, Qing Li, Yang Zhang, Gaowen Liu, Sijia Liu, Shiyu Chang
NAACL'24
Improving Diffusion Models for Scene Text Editing with Dual Encoders
Jiabao Ji‡, Guanhua Zhang‡, Zhaowen Wang, Bairu Hou, Zhifei Zhang, Brian Price, Shiyu Chang
TMLR'23
Towards Coherent Image Inpainting Using Denoising Diffusion Implicit Models
Guanhua Zhang‡, Jiabao Ji‡, Yang Zhang, Mo Yu, Tommi Jaakkola, Shiyu Chang
ICML'23
DFM: Dialogue Foundation Model for Universal Large-Scale Dialogue-Oriented Task Learning
Zhi Chen, Jiabao Ji, Lu Chen, Yuncong Liu, Da Ma, Bei Chen, Mengyue Wu, Su Zhu, Jian-Guang Lou, Kai Yu
Arxiv'22
Controlling the Focus of Pretrained Language Generation Models
Jiabao Ji, Yoon Kim, James Glass, Tianxing He
Findings of ACL'22
WebSRC: A Dataset for Web-Based Structural Reading Comprehension
Xingyu Chen, Zihan Zhao, Lu Chen, Jiabao Ji, Danyang Zhang, Ao Luo, Yuxuan Xiong, Kai Yu
EMNLP'21
Vitæ
Full Resume in PDF.
-
Apple AI/ML 2025 SummerResearch Intern
-
Meta GenAI 2024 SummerResearch Intern at GenAI Llama team
-
Adobe research 2023 SummerResearch Intern
-
UCSB 2022 - nowPh.D. Student
-
X-Lance Lab 2019 - 2022Research Intern
-
SJTU 2018 - 2022B.S. Student
Services
Conference reviewer: ICML'23, 24, 25; Neurips'23, 24, 25; ICLR'24, 25; CVPR'24, 25; ICCV' 25, ACL'24; COLM'24, 25; ARR 24 Dec
Workshop reviewer: Neurips'24 AdvML-Frontiers; RSS'25 FM4RoboPlan
TA: CSW8-W23; CS190I-W23,W24; CS165B-S22,S24
Misc
• I was a KDE+archlinux user with good animation and enormous amount of apps brought by pacman. But I changed to MacOS for more stable working environment. I also bought a Framework 16 for a better Linux experience.• I started to use Vim and tiling window manager like i3/dwm/amethyst recently. Now I find aerospace to be a better TLM for macos.
• I watch many sports. I'm a fan of Leo Messi, Derrick Rose, Tadej Pogačar.