Skip to content
View zqinglin's full-sized avatar
4 followers · 5 following

Highlights

  • Pro

Block or report zqinglin

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.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. 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
zqinglin/README.md

Hi, I'm Qinglin Zeng! 👋

I am currently a sophomore undergraduate researcher at Sun Yat-sen University, advised by Prof. Keze Wang. My research aims to build intelligent, interactive, and reasoning AI systems, with a strong focus on the foundations of Large Language Models (LLMs), Vision Language Models (VLMs), and Representation Learning.

🔥 I am actively seeking Research Internship opportunities! Please feel free to reach out if our interests align.

🔭 Research Interests

My current research trajectory spans across:

  • LLMs & Reasoning: Enhancing reasoning capabilities, model efficiency, and real-time hallucination correction.
  • Multimodal & Vision Language Models: Coherent video generation, collaborative planning, and bridging continuous signals.
  • Agentic Systems: Building collaborative, planning-capable agents for complex tasks.
  • On-Policy Distillation (OPD): Exploring efficient distillation strategies for model alignment and optimization.

📝 Selected Publications

  • [ICASSP 2026 Oral] Reflective Confidence: Correcting Reasoning Flaws via Online Self-Correction
    Qinglin Zeng, J Yang, K Wang.
  • [ACL 2026 Findings] RACC: Regret-Aware Confidence Calibration for Consistent Masked Discrete Diffusion Decoding
    Qinglin Zeng, J Zhang, J Yang, N Liu, K Wang.
  • [Preprint] CoAgent: Collaborative Planning and Consistency Agent for Coherent Video Generation
    Qinglin Zeng, K Cai, R Chen, Q Lv, K Wang.
  • [AAAI 2026] Llm-cas: Dynamic neuron perturbation for real-time hallucination correction
    J Zhang, N Liu, Y Fan, Z Huang, Qinglin Zeng, K Cai, J Wang, K Wang.

🛠️ Tech Stack & Workflow

  • Deep Learning & LLMs: PyTorch, Hugging Face Ecosystem (Transformers, Diffusers), vLLM / DeepSpeed, sglang / verl.
  • Embodied AI: Gym/Gymnasium, MuJoCo, Isaac Gym.
  • System & Infrastructure: Linux, Docker, SLURM (Cluster Computing).

📫 Let's Connect

Popular repositories Loading

  1. SYSU-CV SYSU-CV Public

    A CV template for SYSU, more suitable for undergraduate students majoring in CS.

    TeX 6 1

  2. ReflectConf ReflectConf Public

    Reflective Confidence: a research release of our confidence-aware reasoning pipeline with a patched vLLM runtime for online self-correction.

    Python 1

  3. milk_loong milk_loong Public

    A Alexnet base on about 200 pictures of milk loong and other cartoon character ,and this just provide the dataset and train code,didn't save the trained model

  4. PaperNet PaperNet Public

    Construct a wonderful net of the paper you have read!

  5. cemera-vla cemera-vla Public

    Python

  6. Wan2.1Server Wan2.1Server Public

    Python