Experience

  1. Technical Staff (PostTrain)

    Moonshot AI (月之暗面)
    Cooking K2.5 series: Native Multimodal RL (K2.5 Report Chap2.2, ZeroVision ColdStart -> Vision-Centric RL), Vision Reasoning & Knowledge & TIR RL, Chart Understanding and Chart-to-Code.

  2. Research Intern

    01.ai (零一万物) & Rhymes.ai
    Core Contributor of Aria (an Open Multimodal Native MoE)

  3. Research Intern

    IDEA (International Digital Economy Academy)
    Co-first Author of ICLR2025(Oral) - ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding

  4. Research Intern

    Huawei Noah's Ark Lab
    Preliminary Research about MLLM on Document Recognition, Parsing and Understannding.

Education

  1. Master of Science

    Peking University (PKU)
    Vision-Language Models, MLLM Reasoning, AI-Generated Image/Video Quality Assesment

  2. Bachelor of Engineering

    Huazhong University of Science and Technology (HUST)
    Grade: 3.9/4 Honours Bachelor Degree (Summa Cum Laude, Top 2%)
Awards
ICLR2025 Oral (Top 1.8%) - ChartMoE
ICLR 2025 ∙ February 2025

ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding

Selected for Oral Presentation at the 2025 International Conference on Learning Representations (Top 1.8%).

🐙 See github repo
🥉3rd winner of NTIRE 2024 Quality Assessment for AI-Generated Content - Track 2 Video
CVPR & NTIRE ∙ April 2024

Exploring AIGC Video Quality: A Focus on 1️⃣Visual Harmony, 2️⃣Video-Text Consistency and 3️⃣Domain Distribution Gap.

Paper: https://arxiv.org/abs/2404.13573

See certificate
🥈Kaggle Silver Medal：Stable Diffusion - Image to Prompts
Kaggle ∙ May 2023
As a captain, win a silver medal🥈.(42/1231, Top3%)
See certificate
🏅Honorous Bachelor Degree
Huazhong University of Science and Technology (HUST) ∙ June 2022
Honours Bachelor Degree (Summa Cum Laude, Top 2%), due to GPA, scientific research achievements, and student work awards.