Here is Weijue Bu. You can call me Weijue. I am a researcher passionate about Vision-Language Models (VLM) and Embodied AI. I am currently working on mitigating hallucinations in large models and bridging the gap between vision and robotic actions.
- 🔭 I’m currently working on Hallucination Mitigation in VLMs (Author of "Conscious Gaze")
- 🌱 I’m currently learning Serverless Computing Optimization and Compilation Theory
- 👯 I would like to collaborate on Vision-Language-Action (VLA) models and Robotics
- ⚡ Fun fact: I am also building SachetAI, a generative AI project for traditional Chinese patterns.
- 📫 The way to reach me:
These are the skills that I know and have mastered:
- Current: China University of Mining and Technology
Mainly conduct research on Multimodal Learning, aiming to align visual perception with linguistic reasoning in complex environments.
- Key Project: Conscious Gaze: Adaptive Attention Mechanisms for Hallucination Mitigation in Vision-Language Models.
I am actively exploring the application of Reinforcement Learning (PPO) in Serverless Computing and the principles of Compilation Technology.
The number of views of this page is as follows.