PrimeNet: Human-inspired Framework for Commonsense Knowledge Representation and Reasoning

A human-inspired framework for commonsense knowledge representation and reasoning.

Behavioural Understanding and Narrative Descriptions from Videos

An effectively and flexibly multimodal system to analyze and narrate real world events captured in video.

Neural Sequence Labeling

An unified and accurate deep learning based approach for various sequence labeling tasks such as POS, Chunking, NER, SRL, Punctuation Restoration and etc.

Removing Backscatter to Enhance the Visibility of Underwater Object

A novel and effective method based on underwater optical model for underwater object visibility enhancement.

Image Enhancement based on Retinex Theory and Dual-tree Complex Wavelet Transform

An improved and effective method for image enhancement based on retinex theory and dual-tree complex wavelet transform techniques.

Recent Posts

More Posts

Explanation of variational autoencoders.


An introduction of TensorFlow dataset API.


Tutorial about how to update multiple tensors using tensorflow scan function.


Some useful commands and tips to raise efficiency under Mac OS and Linux, keep updating!



. Dual Adversarial Neural Transfer for Low-resource Named Entity Recognition. In ACL, 2019.

. RoSeq: Robust Sequence Labeling. In IEEE TNNLS, 2019.

PDF Code

. Learning with Annotation of Various Degrees. In IEEE TNNLS, 2019.

PDF Code

. Towards Verifying Semantic Roles Co-occurrence. In ArXiv, 2018.


. Action Hierarchy Extraction and its Application. In LREC, 2018.

PDF Code


* Accomplished the Scientific Writing Skills Workshop Courses
       Jan. 2019     |       Agency of Science, Technology and Research, Singapore     |       Certificate

1st Runner Up in Artificial Intelligence (AI) Hackathon of A*STAR
       Sep. 2018     |       Agency of Science, Technology and Research, Singapore     |       Certificate


  • 1 Fusionopolis Way, #08-10 Connexis North Tower, Artificial Intelligence Initiative, 138632, Singapore
  • Tuesday 13:00 to 17:30 or email for appointment