Chao Yang
杨超, Post-doctoral Researcher, Shanghai AI Lab.
I am Post-doctoral Researcher at Shanghai AI Lab (上海人工智能实验室), leading a fundamental Large Model Safety & Decision Intelligence research group. I am also co-working with Yu Qiao, Jing Shao and Yu Liu from Shanghai AI Lab closely.
Previously, I received my Ph.D. in Tsinghua University in 2022, advised by Prof. Fuchun Sun. I was fortunate enough to get the guidance of Prof. Huaping Liu and Prof. Wenbing Huang when pursuing my doctoral degree. I graduated from the Department of Computer Science and Technology, Tsinghua University with a master’s degree and from Sichuan University with a bachelor’s degree.
My research interest includes Large Language Model Safety, Multi-modal Large Model, and Robotic Embodied Intelligence. Some of my current research keywords can be found below:
- Large Language Model: Alignment and Finetuning, LLM Attack and Defense.
- Multimodal LLM: Vision and language Fusion, Video-QA, VQA.
- Embodied Robotics: Robotic Manipulation, Reinforcement and Imitation Learning.
Please feel free to email me at yangchao9264 [at] 126 [dot] cn or my GMAIL: yangchaoemigmo [at] gmail [dot] com.
news
Apr 20, 2024 | One Paper accepted by IJCAI 2024 Survey Track. |
---|---|
Mar 13, 2024 | One LLM safety survey paper accepted by NAACL 2024. |
Feb 27, 2024 | Two Papers accepted by CVPR 2024. |
Dec 9, 2023 | One Offline RL Paper accepted by AAAI 2024. |
selected publications
*(Equal contribution), +(Corresponding author).2024
- VideoDistill: Language-aware Vision Distillation for Video Question AnsweringarXiv preprint arXiv:2404.00973, 2024
- LLaMA-Excitor: General Instruction Tuning via Indirect Feature InteractionarXiv preprint arXiv:2404.00913, 2024
- Emulated Disalignment: Safety Alignment for Large Language Models May Backfire!arXiv preprint arXiv:2402.12343, 2024
- Attacks, defenses and evaluations for llm conversation safety: A surveyarXiv preprint arXiv:2402.09283, 2024
- Safety of Multimodal Large Language Models on Images and TextarXiv preprint arXiv:2402.00357, 2024