Welcome to Tieyuan Chen’s Homepage

👋 About Me

Hello! I am Tieyuan Chen, a third-year Ph.D. student (2023–present) at Shanghai Jiao Tong University, School of Electronic Information and Electrical Engineering (SEIEE), advised by Prof. Weiyao Lin. To date, during my Ph.D. studies, I have published 4 first-author papers (7 papers in total), including top-tier venues such as T-PAMI * 2, ICLR * 2, NeurIPS * 2, and T-CSVT.

Previously, I received my B.Eng. degree from Sichuan University, College of Electronics and Information Engineering (CEIE) (2019–2023), ranking 1 / 29.

I was selected for the Joint PhD Program at Beijing Zhongguancun Academy (Sep. 2024 – June 2028).

Currently, I am a Research Intern at AGI Center, Ant Research Institute (Mar. 2025 – Present), working under the supervision of Jianguo Li, Tao Lin, Haoxing Chen, and Huabin Liu.

🔬 Research Interests

My research focuses on:

• 🎥 Video Understanding & Video Reasoning
• 🧠 Large Language Models (LLMs) & Multimodal LLMs (MLLMs)
• 🔗 Causal Reasoning and Event-level Modeling

📫 Feel free to reach out via email:
tieyuanchen@sjtu.edu.cn

🥇 Honors and Awards

• China National Scholarship (2021) — Top 1%
• China National Scholarship (2022) — Top 1%
• Sichuan University Comprehensive Special Scholarship (2022) — Top 0.1%
• Sichuan University Hundred Excellent Student (2022) — Top 0.2%
• Sichuan Province Outstanding Graduate (2023) — Top 3%

📝 First-Author Publications

	MECD: Unlocking Multi-Event Causal Discovery in Video Reasoning Tieyuan Chen, Huabin Liu, Tianyao He, Yihang Chen, Chaofan Gan, Xiao Ma, Cheng Zhong, Yang Zhang, Yingxue Wang, Hui Lin, Weiyao Lin Conference on Neural Information Processing Systems (NeurIPS), 2024 (Spotlight, Top 2.4%)
	DND: Boosting Large Language Models with Dynamic Nested Depth Tieyuan Chen, Xiaodong Chen, Haoxing Chen, Zhenzhong Lan, Weiyao Lin, Jianguo Li International Conference on Learning Representations (ICLR), 2026
	MECD+: Unlocking Event-Level Causal Graph Discovery for Video Reasoning Tieyuan Chen, Huabin Liu, Yi Wang, Yihang Chen, Tianyao He, Chaofan Gan, Huanyu He, Weiyao Lin IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
	CSTA: Spatial-Temporal Causal Adaptive Learning for Exemplar-Free Video Class-Incremental Learning Tieyuan Chen, Huabin Liu, Chern Hong Lim, John See, Xing Gao, Junhui Hou, Weiyao Lin IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

📑 Technical Reports

Since March 2025, I have been actively participating in many AR-Based LLM, Diffusion-Based LLM, Diffusion-Based VLM researches at inclusion AI. Below are the technical reports and open-source models I have contributed to during this period:

• LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Model
  LLaDA2.0-Uni is a unified discrete diffusion language model, achieving performance comparable to specialized vision-language models while enabling efficient inference and high-fidelity image generation.
  Role: Contributor (Mask Token Reweighting Loss & Data Processing) | Apr. 2026
  

• LLaDA-MoE: A Sparse MoE Diffusion Language Model
  The first open-source Mixture-of-Experts (MoE) diffusion large language model, pre-trained from scratch on approximately 20 trillion tokens.
  Role: Contributor (Megatron AI infra) | Oct. 2025
  

• DND: Boosting Large Language Models with Dynamic Nested Depth
  Improves LLM inference efficiency and reasoning capabilities by dynamically adjusting compute depth via a novel nested architecture.
  Role: Independent First Author | Sep. 2025
  

• Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts
  Proposes a novel MoE architecture utilizing adjugate experts to achieve better parameter efficiency and overall model performance.
  Role: Core Contributor | Aug. 2025
  

📊 Academic Service

Reviewer for Top Conferences

• NeurIPS (2025), ICLR (2025, 2026), ICML (2026), CVPR (2025, 2026), ICCV (2025), ECCV (2026), AAAI (2025), BMVC (2026)