cv

Research Assistant

Center for AI & Robotics, HKISI, CAS

Researching MLLMs for , and continual learning.

Graduate Researcher

Interplay Lab (ROC-HCI Lab), University of Rochester

Design agentic LLM framework to engage K-12 children in collaborative reasoning.

Research Assistant

URseismo, University of Rochester

Develop deep learning models to solve seismic problems & computational imaging problems.

Master's

University of Rochester

Computer Science

• Large Language Models
• End-to-End Deep Learning
• Machine Vision
• AR/VR Interactive Design
• Parallel & Distributed Systems
• Machine Learning
• Data Mining
• Deep Learning

Bachelor of Science

Hong Kong Polytechnic University

Computing

VTCBench: Can Vision-Language Models Understand Long Context with Vision-Text Compression?

Arxiv Preprint

The first comprehensive benchmark specifically designed to evaluate long-context understanding capabilities of Vision-Language Models (VLMs) within the Vision-Text Compression (VTC) paradigm.

MLLM-CL: Continual Learning for Multimodal Large Language Models

Arxiv Preprint

A benchmark for continual learning in multimodal large language models with IID and non-IID data.

Deep Algorithm Unrolling for Seismic Migration

2025 IEEE Conference on Computational Imaging Using Synthetic Apertures

We propose the inverse Radon transform Network (iRADNet), a deep learning-based framework that unrolls the Iterative Shrinkage Thresholding Algorithm (ISTA) into a trainable RNN for solving the inverse Radon transform.