👋 Welcome to my personal academic website! I am Renjie Gu, from Jiangyin City, Jiangsu Province, currently an undergraduate student majoring in Computer Science and Technology at Central South University, passionate about the application of computer technology! What’s more, I’m interested in trustworthy AI,T2I and LLM/MLLM.I am currently working as a research assistant at the Nanyang Technological University, Singapore, collaborating with Prof. Yiming Li and Boheng Li. 🔥

Education & Experience 🎓

Central South University, Changsha, China
Undergraduate student in Computer Science and Technology
GPA: 90.87/100 (3.92/4.0)
09/2022 - Present

Nanyang Technological University, Singapore Research Assistant,12/2024 - Present My time here is spent studing building resilient unlearning for diffusion models.I’m also interested in other AI safety topics and general mllm and diffusion model for image generation.

SYNC (Beijing, China)(AI Startup)
LLM and iOS Development Intern,07/2024 - 09/2024
I focus on optimizing specific chat scenarios using LLM agents. My work also included enhancing the humor and empathy of LLMs by fine-tuning models.

Publications & Research 🧠

• A secret now becaused it’s on submiited to Nips 2025, Second author, collabrated with Boheng Li I extensively studied methods for defending LLMs and diffusion models against harmful fine-tuning and methods for efficient Bilevel Optimization. I finished the experiment codes.
• MHALO: Evaluating MLLMs as Fine-grained Hallucination Detectors, Equal first contribution, co-authored with Yishuo Cai ACL 2025 Findings We developed a fine-grained hallucination benchmark for a multimodal large language model as a judge.

• Course-Correction: Safety Alignment Using Synthetic Preferences, co-authored with Rongwu xu,Yishuo Cai EMNLP 2024
  This research provides in-depth insights into AI safety and model alignment, highlighting the concept of course-correction and proposing the C-Syn benchmark and methods for course-correction LLM fine-tuning. I was mainly responsible for the experimental part of the team, including dataset synthesis and model performance testing.

• During my internship, I conducted in-depth research on the performance of AI agents and LLMs in practical applications, especially on how to enhance the empathy and humor of LLMs and optimize model performance through synthetic data generation.

Awards & Honors 🏆

• **First Prize in the Central South Region of the China Computer Design Competition🥇 **
• Third Prize in the National Service Innovation Competition🥉
• Second Prize in the National College English Competition🥈
• Yangguang Excellent Student Scholarship (awarded to top 1% of students)

Selected Projects 💡

• Job Recommendation and Skill Assessment System Based on Knowledge Graph
  Responsible for researching and developing recommendation algorithms, implemented resume entity recognition and knowledge graph-based content recommendation algorithms using Python, PyTorch, and Flask, and optimized suggestions by deploying ChatGLM.

• iOS Online Chat Application
  Developed a chat application using SwiftUI and Google Firebase, focusing on user-friendliness and efficient data processing.

Skills 🔧

• Programming Languages and Tools: Python, PyTorch, C++, Swift, Git, SQL, Conda, HTML/CSS
• Language Proficiency: IELTS 7.0; made significant progress in reading English academic papers and effective communication over the past year. Achieved a national second prize in the National College English Competition.
• Team Collaboration: Extensive experience in cross-functional team collaboration, working closely with team members in academic and internship projects.

✨ This website uses the Academic Pages template and is hosted on GitHub Pages. Feel free to explore my research achievements, projects, and contact information!