I am a PhD student in the Department of Electrical and Computer Engineering at Rutgers, the State University of New Jersey, working with Prof. Anand Sarwate. I earned my Master’s degree in Electronic Engineering and Information Science in 2020, under the supervision of Prof. Bin Liu and my Bachelor’s degree in Information Security in 2017, both from the University of Science and Technology of China(USTC).

I work on LLM agents and multi-LLM collaboration — model routing and cascades, adaptive deferral, and reliable inference under resource constraints. I am currently a Research Scientist Intern at AnalogyAI, building agentic data-curation systems for LLM training data. Earlier in my PhD I worked on learning with a reject option and WiFi-based human activity recognition.

Publications

Accepted

Pre-print

Internship Experience

Agentic Data Curation for LLM Training Data

Position: Research Scientist Intern
Location: AnalogyAI
Duration: Summer 2026

• Building the curation agent of an autonomous, multi-agent data pipeline for LLM training data: tool-calling agent loops, budget-aware source rationing, and validated LLM decision-making with deterministic fallbacks.

On Accuracy of Delay and Extrapolate Method in Online Misalignment Estimation

Position: Research Intern
Location: Bell Lab, USA
Duration: June 2022 - August 2022
Advisor: Carl Nuzman, Shirin Jalali

• We rigorously analyze the accuracy and effectiveness of the classic method “delay and extrapolate” for estimating the misalignment error when applied to the least mean squares (LMS) algorithm.
• Our analysis for mean squared error of filter coefficients made corrections to similar, well-known results in the previous LMS literature. report

Tessutivo: Contextual Interactions on Interactive Fabrics with Inductive Sensing

Position: Research Intern
Location: Dartmouth College, USA
Duration: December 2018 - April 2019
Advisor: Xing-Dong Yang

• Proposed Tessutivo, a contact-based inductive sensing technique for contextual interactions on interactive fabrics.
• Built a prototype containing coils made of conductive thread with maximized sensitivity conducted by a new inductance approximation formula.

Teaching Experience

Role: Teaching Assistant

Institution: Rutgers

Linear Signal and System (Fall 2023, Fall 2024, Fall 2025, Spring 2026)

Discrete Mathematics (Spring 2022)

Principles of Electrical Engineering (Fall 2021, Fall 2020)

Probability and Random Processes (Spring 2021)

Institution: USTC

Mathematical Logic and Graph Theory (Fall 2018)