I got my PhD in Computer Science from University of Illinois at Urbana–Champaign, advised by Prof. Svetlana Lazebnik. 

The ultimate goal of my research interest is to build multimodal interactive AI systems that not only understand and recreate the visual world but also communicate like human beings with natural language.

Recently, my main research themes include:

• joint modeling of images and natural language,
• multimodal learning and multimodal interactive system,
• controllable language generation and conversational AI,
• ​machine learning methods for Computer Vision / NLP,
• ... 

Email :
lwang97@illinois.edu

News: 

• New work of Improving Weakly Supervised Grounding by Contrastive Knowledge Distillation is on arxiv.
• Two ECCV 2020 papers on vision+language are accepted! Congrats to my 2019 interns at Bellevue!
• To generate coherent video paragraphs? Take a look at our ACL 2020 long paper-MART.
• I am co-organizing the second Learning from Imperfect Data workshop in CVPR 2020.
• One ICCV 2019 Oral for real time image-text grounding.
• One CVPR 2019 Oral for diverse and controllable language generation.