Biography

I am a Master's student in Economics & Management at the University of Electronic Science and Technology of China (UESTC), where I ranked 1st in the national entrance examination as a cross-disciplinary candidate. My research interest is data governance, and I co-authored FedLing, accepted to EMNLP 2026 Findings.

Previously, I was a Product Operations Intern at Trip.com Group (Shanghai HQ, Overseas Business Unit), where I worked on Google-channel travel products — translating platform requirements into shipped features, governing cross-platform data quality, and lifting content exposure by 11% and channel conversion by 3%. Before that, I interned in marketing at Chengdu China International Travel Service, growing a user community by 14k members in three months.

I enjoy turning ambiguous business problems into clear requirements, metrics and shipped outcomes.

News

Experience

Trip.com Group · Shanghai HQ — Product Operations Intern, Overseas Business Unit2024.05 — 2024.09
Owned the Google channel for overseas attraction products. Translated Google platform requirements into feature scopes, data definitions and acceptance criteria; shipped 2 features on schedule (multi-day pass support, Google Discover content integration), lifting content exposure by 11%. Ran funnel analysis on Google Trends / Search Console data, surfacing 3+ growth opportunities monthly and raising target-page conversion by ~6%. Led data governance across 20+ fields and first-time onboarding of 55 high-potential attractions (Tokyo Disneyland, Colosseum); delivered 183% of monthly listing targets with zero major data errors; closed 60+ cross-team issues within 12 hours each.
Chengdu China International Travel Service — Marketing Intern2023.09 — 2024.02
Designed WeChat campaigns for new-product launch and community growth, reviewed through a reach–join–consult funnel: 15,000+ reach, 200+ inquiries, 235 day-one signups (117% of target); community grew by nearly 14k members in 3 months. Independently closed 16 customized travel orders end-to-end.

Publication

FedLing: Linguistic-Structure-Aware Federated Learning for Fair Multilingual LLM AdaptationEMNLP 2026 Findings
A typology-aware federated learning framework that routes LoRA update subspaces across 40 languages, improving low-resource multilingual adaptation while reducing upload cost. I contributed experimental data analysis, robustness studies and communication-cost accounting.

Honors & Awards

Leadership

Skills