CV

Teaching Assistant

University of California, Davis

• ECS 153 Computer Security.
• ECS 036A Programming and Problem Solving.
• ECS 036C Data Structures, Algorithms, and Programming.
• ECS 140A Programming Languages.

Software and Operations Engineering Intern

Volkswagen Group China

• Implemented data masking and auto-populated JIRA logs in Python.
• Analyzed service and financial data with Tableau to support CIO-level strategy decisions.

Quality Assurance Intern

Didi Global

• Developed distributed Java invoice services to support communication between servers and end-user devices.
• Automated unit testing with JUnit and Redis-backed test cases.

Ph.D.

Purdue University

Computer Science; Advisor: Prof. Tianyi Zhang

M.Sc.

University of California, Davis

Computer Science

B.Eng. (Honours)

Australian National University

Computer Science

B.Eng.

Shandong University

Computer Science

LLAMAFUZZ: Large Language Model Enhanced Greybox Fuzzing

AST 2026

LLM-enhanced greybox fuzzing that learns structured input formats and mutation strategies.

FLARE: Fine-Grained Diagnostic Feedback for LLM Code Refinement

Under review

Efficient code refinement with a lightweight line-level fault localizer and test-time scaling.

Enhancing Multi-Agent Communication through Attention Steering with Context Relevance

Under review

Training-free context management for multi-agent systems with attention steering and temporal/spatial decay.

Active Layer-Contrastive Decoding Reduces Hallucination in Large Language Model Generation

EMNLP 2025 Main

RL-guided layer-contrastive decoding for reducing hallucinations in large language model generation.

Attention-Aligned Reasoning for Large Language Models

Under review

Reasoning-time attention alignment and prompt engineering for keeping LLMs focused on prompts and intermediate goals.

SteerDiff: Steering towards Safe Text-to-Image Diffusion Models

arXiv preprint

Safe generative AI through lightweight prompt steering for text-to-image diffusion models.