Xing Yi (Peter) Liu

I received my M.S. in Computer Science from Columbia University. My research interest is in multimodal learning, efficient machine learning, and applications of deep learning.

liu.peter@columbia.edu  |  CV  |  Google Scholar

Human pharmaceutical experts draw insight from both structured knowledge in knowledge bases and unstructured knowledge in biomedical literature. Our proposal, KEDD, extracts knowledge similarly and solves a variety of drug discovery tasks in a unified framework.

EfficientPunct outperforms the previous best punctuation restoration model by 1.0 F1 points, using less than a tenth of its parameters to process embeddings. We create an ensemble streamlined from a speech recognizer to extract audio embeddings.

We introduce MolFM, a multimodal molecular foundation model designed to facilitate joint representation learning from molecular structures, biomedical texts, and knowledge graphs. We propose cross-modal attention to facilitate comprehension between these modalities.

Our submission to the NTCIR-12 task treats short text conversation as a community question-answering problem. We achieved performances of mean nDCG@1 0.2767, mean P+ 0.4284 and mean nERR@10 0.4095.

Thank you to Jon Barron for this website's template.