I’m a Data Science undergraduate at CUHK-Shenzhen, currently on exchange in the Department of Computer Science and Engineering at UC San Diego.
I study language-model agents in changing markets. In my first-author IEEE PES APPEEC 2026 study, we combine cost-based reasoning and market feedback to adapt bids without fine-tuning.
More about my researchFirst author · IEEE PES APPEEC 2026
A generation company has to account for electricity and carbon costs. We separate cost calculation from strategic markup, then use rule checks and recent market feedback to guide a language-model bidder.
On the IEEE 14-bus system, it earned higher average profit than a fixed-markup baseline while clearing in every round, without fine-tuning.
Project lead · Original research with MCM Team 2627761
What can public scores and eliminations reveal when audience votes stay private? I lead a project combining Bayesian support models, voting-rule comparisons and historical elimination forecasts, building on our MCM team study.
Under a declared deterministic voting rule, many vote allocations can explain the same exit. Comparing alternative rules therefore requires explicit assumptions about the hidden votes.
In our Renewable Energy paper, LLM-CECM, I built the simulation pipeline for an IEEE 118-bus market study. The public code companion connects the paper to its implementation. My pricing-agent study now includes public stationary-market and demand-cycle experiments.
Beyond markets, I lead a forest-type mapping project combining seasonal satellite imagery, terrain and climate, with a completed three-seed comparison on ForTy. I also lead the current driving-scene review workflow, linking sparse features, saved predictions and a browser explorer.
B.S. in Data Science and Big Data Technology
In progress
Department of Computer Science and Engineering
Exchange student · Current
