cv

Ph.D.

EPFL (Swiss Federal Institute of Technology of Lausanne)

Machine Learning

Master

EPFL (Swiss Federal Institute of Technology of Lausanne)

Data Science

Master

University of California, San Diego

Computer Science

Bachelor

University of Electronic Science and Technology of China

Microelectronic Science and Engineering

Research Intern

Oracle Labs

Developed and evaluated a machine learning model that detects anomalous Linux sessions in the cloud servers.

Algorithm Engineering Intern

Didi Chuxing (China's largest taxi-hailing platform)

Developed an algorithm for learning road segment weights from historical ride data, improving route planning quality for ride-hailing services.

Efficient Training Data Attribution for Large Language Models

Training Data Attribution faces severe computational challenges for modern-sized LLMs. We improved conventional influence functions by exploiting the low-rank property of the gradient to make them more efficient and scalable.

Improving Waste Detection and Sorting on Conveyor Belt

Supervising two projects: (1) YOLO uncertainty estimation and confidence recalibration and (2) real‑time object brand recognition on conveyor belts.

Learning to Weight Parameters for Training Data Attribution

Identified the heterogeneity of attribution strengths across parameters; proposed a method that learns to re-weight layers to amplify the true signal, boosting accuracy and enabling fine-grained (e.g., subject vs. style) attribution.

Enhancing Compositional Text-to-Image Generation with Reliable Random Seeds

Identified the significant role of initial noise in text-to-image inconsistencies; proposed a method to mine reliable random seeds to improve text-to-image generation.

Controlling the Fidelity and Diversity of Deep Generative Models via Pseudo Density

Proposed an approach to bias deep generative models, such as GANs and diffusion models, towards generating data with either enhanced fidelity or increased diversity.

Interlock-Free Multi-Aspect Rationalization for Text Classification

Proposed a multi-stage training method incorporating an additional self-supervised contrastive loss that helps to alleviate the interlocking issue and extract more semantically diverse rationales.

Head Teaching Assistant

CS-233: Introduction to Machine Learning

Teaching Assistant

CS-401: Applied Data Analysis

Teaching Assistant

CS-233: Introduction to Machine Learning

Teaching Assistant

COM-407: TCP/IP Networking

Teaching Assistant

COM-112: Object-Oriented Programming (in C++)

Student Teaching Assistant

CS-456: Deep Reinforcement Learning

National Scholarship

China Collegiate Programming Contest - Gold Medal

China Collegiate Computing Contest - First Prize

First-class People's Scholarship

ACM International Collegiate Programming Contest (Asia Regional) - Bronze Medal

China Collegiate Computing Contest (Group Programming Ladder Tournament) - First Prize

First-class People's Scholarship

Learning to Weight Parameters for Data Attribution

arXiv (submitted to ICLR 2026)

Enhancing Compositional Text-to-Image Generation with Reliable Random Seeds

International Conference on Learning Representations

Controlling the Fidelity and Diversity of Deep Generative Models via Pseudo Density

Transactions on Machine Learning Research

Interlock-Free Multi-Aspect Rationalization for Text Classification

arXiv