Skip to content
View LucasXinYu's full-sized avatar
🏠
Working from home
🏠
Working from home
11 followers · 4 following

Achievements

Achievement: Pull SharkAchievement: Starstruck

Achievements

Achievement: Pull SharkAchievement: Starstruck

Block or report LucasXinYu

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
LucasXinYu/README.md

Xin (Lucas) Yu

Ph.D. Candidate in Statistics at Penn State · Student Researcher at Google.

Working on agentic RL, tool use, and distillation.

Homepage · Google Scholar · CV · LinkedIn · Email

Selected work

  • MLToolBench — Tool-augmented agents for ML development.
  • PBSD — Preference-based self-distillation.
  • AltLoRA — Efficient low-rank adaptation · NeurIPS 2025.
  • FedCLUP — Personalized federated learning · ICML 2025.

Previously at Meta, TikTok, ByteDance Seed, and Microsoft Research Asia.

Pinned Loading

  1. PBSD PBSD Public

    Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization. [arXiv 2026]

    Python 26

  2. AltLoRA AltLoRA Public

    [NeurIPS 2025] Better gradient approximation in low-rank adaptation with alternating projections.

    Jupyter Notebook 7

  3. FedCLUP FedCLUP Public

    [ICML 2025] Statistical accuracy–communication trade-offs in personalized federated learning with minimax guarantees.

    Python 8

  4. -MiniOneRec -MiniOneRec Public

    Generative recommendation experiments based on MiniOneRec: semantic IDs, supervised fine-tuning, and GRPO.

    Python 5