Skip to content
View TiezMind's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report TiezMind

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
TiezMind/README.md

Hi there, I'm Pengyu Li (李鹏宇) 👋

Typing SVG

Homepage Google Scholar Email GitHub


🙋 About Me

I am a second-year M.S. student in Computer Science & Technology at Xi'an Jiaotong University. My research centers on large language models and multimodal / omni-modal foundation models, with a focus on the full post-training pipeline.

I have interned at ByteDance Seed, TikTok, working on omni-modal (speech) foundation models, large-scale multimodal content understanding.

  • 🔭 Currently working on the omni-modal foundation model @ ByteDance Seed
  • 🌱 Interested in unifying perception, reasoning, and generation across modalities — efficiently
  • 💬 Happy to chat about multimodal LLMs, omni-modal training, and RL post-training
  • 📫 Feel free to email me for any form of academic cooperation!

🔬 Research Interests

  • Multimodal & Omni-modal LLMs — unifying text, vision, and speech into a single foundation model
  • LLM Continued Pre-training · Mid-training · Post-training — data recipes, task composition & interference
  • Efficient & Unified Multimodal Reasoning — token pruning, distillation, RL (GRPO) for reasoning

📄 Selected Publications

⭐ = first author  ·  full list on Google Scholar

  • ⭐ AGTAO: Robust and Stabilized LLM Unlearning via Adversarial Gating Training with Adaptive Orthogonality
    ACL   Paper Code

  • ⭐ Visual-OPSD: Cross-Modal On-Policy Self-Distillation for Efficient Unified Multimodal Reasoning
    Review   Paper Code

  • ⭐ Look Closer, Look Back, or Wait for Agentic Streaming Video Understanding
    Review   Paper Code

  • ⭐ Do Thinking VLMs Need to Keep Looking at Long Videos? Write It Down, Then Look Away
    Review   Paper Code

  • ⭐ EMDFNet: Efficient Multi-scale and Diverse Feature Network for Traffic Sign Detection
    ICANN   Code

  • PhysPRM: A Generative Process Reward Model with Fine-grained Diagnosis for Physics Problem Solving
    ACL   Code

  • AERO: Autonomous Evolutionary Reasoning Optimization via Endogenous Dual-Loop Feedback
    COLM   Code

  • LogicGraph: Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification
    EMNLP   Code

💼 Experience

Role Organization Period Focus
Intern ByteDance Seed 2026.02 – Present Omni-modal foundation models (speech)
Intern TikTok, ByteDance 2025.02 – 2025.09 Multimodal content understanding

🛠️ Tech Stack


Profile Views
"Long-term correctness over short-term convenience."

Pinned Loading

  1. AGT-unlearning AGT-unlearning Public

    [ACL2026] AGTAO : Robust and Stabilized LLM Unlearning via Adversarial Gating Training with Adaptive Orthogonality

    Python 6 1

  2. EMDFNet EMDFNet Public

    [ICANN2024] EMDFNet: Efficient Multi-scale and Diverse Feature Network for Traffic Sign Detection

    Python 1

  3. Visual-OPSD Visual-OPSD Public

    [arxiv 2606] Visual-OPSD: Cross-Modal On-Policy Self-Distillation for Efficient Unified Multimodal Reasoning

    Python 52 1