cv

Graduate Researcher - Temporal Knowledge Leakage Detection

Northwestern University

Developing methods to detect temporal knowledge leakage in LLMs.

• Formalized temporal knowledge leakage problem and introduced a four-phase detection pipeline
• Computed Shapley values and developed Shapley-weighted Decision-Critical Leakage Rate metric
• Proposed Temporal LLM Agent architecture with iterative claim verification to prevent leakage
• Demonstrated significant leakage reduction across legal, salary, and stock ranking tasks

Graduate Researcher - Factual Memorization in LLMs

Northwestern University

Analysis of LLM memorization behaviors under SFT and DPO. Qualifying examination paper.

• Analyzed LLM memorization behaviors under SFT and DPO fine-tuning paradigms
• Reproduced and extended findings from Stanford's Finetunebench
• Proposed formal distinction between passive and positive memorization
• Designed temporal generalization experiments with system prompts to mitigate overfitting

Graduate Researcher - LLM Interpretability (LAMP)

Northwestern University

Developed LAMP for interpreting black-box LLMs. Accepted as Spotlight at AISTATS 2026.

• Developed LAMP framework for interpreting black-box LLMs by fitting locally linear surrogate models
• Designed perturbation-based method to probe model sensitivity without internal access
• Enabled practical interpretability for proprietary LLMs through self-reported explanations
• Conducted experiments across sentiment, controversial-topic detection, and harmful auditing

Graduate Researcher - Unified Off-Policy Learning to Rank

Northwestern University

Unified framework for off-policy Learning to Rank. Published at NeurIPS 2023.

• Formulated unified framework modeling various click models in LTR as MDP
• Proposed Click model-agnostic Unified Off-Policy Learning to Rank (CUOLR) algorithm
• Demonstrated effective application of offline RL methods (DQN, SAC, BCQ, CQL) to LTR
• Achieved state-of-the-art performance across heterogeneous click models

Doctor of Philosophy

Northwestern University

Statistics and Data Science

Certificate (Minor)

University of Science and Technology of China (USTC)

Artificial Intelligence

• Talent Program in Artificial Intelligence
• Outstanding Undergraduate Honorary Rank (top 5%)

Bachelor of Engineering

University of Science and Technology of China (USTC)

Electronic Information Engineering

• Wang Xiaomo Talent Program in Cyber Science and Technology
• Talent Program in Information Science and Technology
• Rank: 5/213 in School of Information Science and Technology

China National Scholarship

Ministry of Education of the P.R.China

Top 1% of students. Academic year 2020-2021.

Outstanding Student Scholarship (Class A)

University of Science and Technology of China

Top 1% of students.

The National Undergraduate Electronic Design Contest

Ministry of Education of China

2nd Prize Nationally, 1st Prize in Anhui Province.

Scholarship for Talent Program in Basic Disciplines (Class A)

University of Science and Technology of China

Top 3% of students.

Signal Distortion Measurement Device Design

High-precision signal distortion measurement device achieving 0.5% error.

• Applied Hanning window functions to reduce spectrum leakage effectively
• Designed algorithm to detect center spectrum by aggregating energy from nearby lines
• Developed LCD interface to visualize data and input analog signals