Jialin Wu

Jialin Wu

Ph.D. student

The University of Texas at Austin

Biography

I am a fifth year Ph.D. student at UTCS advised by Raymond J. Mooney. Before coming to UT Austin, I received my BEng. degree from the Department of Automation supervised by Prof. Xiangyang Ji at Tsinghua University in 2017.

My research focuses on commonsense reasoning and knowledge-based model in vision-language tasks, such as open-knowledge retrieval, question answering and explanation generation. In particular, I explore using external resources for answering various types of visual questions, including commonsense questions and knowledge-based questions. I also work on explaining the model’s decision both visually and textually and multi-modal knowledge retrieval projects.

I am also the organizer of UT XAI reading group.

I am actively looking for full-time opportunity starting summer/fall 2022 and my CV is here.

Interests
  • Language and Vision
  • Explainable AI
Education
  • PhD in Artificial Intelligence, 2017 - present

    UT Austin

  • BEng in Automation, 2013 - 2017

    Tsinghua University

Publications

(2021). Multi-Modal Answer Validation for Knowledge-Based VQA. AAAI 2022.

PDF

(2021). Improving VQA and its Explanations by Comparing Competing Explanations. AAAI 2021 Explainable Agency in Artificial Intelligence Workshop.

PDF

(2019). Self-Critical Reasoning for Robust Visual Question Answering. NeurIPS 2019.

PDF

(2019). Hidden State Guidance: Improving Image Captioning using An Image Conditioned Autoencoder. NeurIPS 2019 Vigil Workshop.

PDF

(2019). Generating Question Relevant Captions to Aid Visual Question Answering. ACL 2019.

PDF

(2019). Faithful Multimodal Explanation for Visual Question Answering. ACL 2019 BlackboxNLP Workshop.

PDF

(2018). Dynamic Filtering with Large Sampling Field for Convnets. ECCV 2018.

PDF

Experience

 
 
 
 
 
Research Intern
May 2020 – Aug 2020 Seattle
 
 
 
 
 
Research Intern
Google Inc.
May 2019 – Aug 2019 New York City

Contact