About

Dr Jie (Jack) Yang is a lecturer (assistant professor) from School of Computing and Information Technology at University of Wollongong. He is working on Natural Language Processing (NLP), a technique to enable computers understanding and using human language effectively. He focuses on theory development, novel techniques and effective solutions of NLP in broad domains, such as Education, Law, Healthcare and Finance management.

  • Interest: Large Language Model (LLM), Multimodal NLP, Machine Reading Comprehension, Knowledge Graph, Information Retrieval

Research

  • Masking the Unknown: Leveraging Masked Samples for Enhanced Data Augmentation, UAI (conference, core: A), 2024
  • Asynchronous Joint-based Temporal Pooling for Skeleton-based Action Recognition, IEEE Transactions on Circuits and Systems for Video Technology (journal), 2024
  • ConClue: Conditional Clue Extraction for Multiple Choice Question Answering, ICDAR (conference, core: A), 2024
  • Extractive Question Answering with Contrastive Puzzles and Reweighted Clues, ICDAR (conference, core: A), 2024
  • COTER: Conditional Optimal Transport meets Table Retrieval, WSDM (conference, core: A), 2024
  • SCAD: Subspace Clustering based Adversarial Detector, WSDM (conference, core: A), 2024
  • Conditional Prototypical Optimal Transport for Enhanced Clue Identification in Multiple Choice Question Answering, AJCAI (conference, core: Australia-B), 2024
  • Denoising Implicit Feedback for Extractive Question Answering, CASA (conference, core: -), 2024
  • An attention-based CNN for automatic whole-body postural assessment, Expert Systems With Applications (Journal) , 2023
  • [MASK] Insertion for anti-adversarial attacks, EACL (conference, core: A), 2023
  • Improving Machine Reading Comprehension through A Simple Masked-Training Scheme, IJCNLP-AACL (conference, core: B), 2023
  • CREAM: Named Entity Recognition with Concise query and Region-Aware Minimization, WISE (conference, core: A), 2023
  • Towards robust token embeddings for Extractive Question Answering, WISE (conference, core: A), 2023
  • MIRS: [MASK] Insertion based Retrieval Stabilizer for Query Variations, DEXA (conference, core: B), 2023
  • Improving Adversarially Robust Sequential Recommendation through Generalizable Perturbations, IEEE-Big data (conference, core: B), 2023
  • Killing Many Birds With One Stone: Single-Source to Multiple-Target Domain Adaptation for Extractive Question Answering, UIC (conference, core: B), 2023
  • Context-Guided Triple Matching for Multiple Choice Question Answering, UIC (conference, core: B), 2023
  • MDA: Masked Language Modeling Meets Data Augmentation for Multiple Choice Question Answering, UIC (conference, core: B), 2023
  • Knowledge Graph Question Answering Based on Contrastive Learning and Feature Transformation, ICQS (conference, core: B), 2022
  • Joint Extraction of Biomedical Entities and Relations based on Decomposition and Recombination Strategy, BIBM (conference, core: B), 2022
  • Seeing the wood for the trees: A contrastive regularization method for the low-resource Knowledge Base Question Answering, NAACL (conference, core: A), 2022
  • Multi-view learning for context-aware extractive summarization, SSCI (conference, core: B), 2020
  • Characteristics of chinese online movie reviews and opinion leadership identification, International Journal of Human–Computer Interaction (Journal), 2020
  • Mining Chinese social media UGC: a big-data framework for analyzing Douban movie reviews, Journal of Big Data (Journal), 2016

Projects

Screenshot of  web app
AMKD.AI

An automatic and modularised toolbox for Knowledge Discovery.

Accomplishments
  • Document corpus analysis of searching and displaying the underlying knowledge from large-scale corpus of documents.
  • Deliver an open-source toolkit to support automatic Knowledge Graph construction given sample documents.
  • Funded by Artificial Intelligence for Decision Making Initiative, NSW, 2020
Screenshot of  web app
iSee

An NLG-based model converts data associated with a photo and a few prompts into coherent sentences.

Accomplishments
  • Social Media Integration.
  • Object Detection.
  • credited to CSIT321 student project
Screenshot of  web app
Sentence annotation tool

Annotation tool for masking evidence sentences for multiple choice.

Accomplishments
  • One can use to tag key sentences in the article that supporting correct answer(s).
  • More detailed tutorial is here
Screenshot of  web app
Next-Gen AIOT

AIOT-powered interactive kiosk.

Accomplishments
  • Enhance visitor experiences at information centers with an AIOT-powered interactive kiosk.
  • Advanced Speech Recognition and Natural Language Processing to understand user inquiries and a robust chatbot to assist in crafting ideal journeys.
  • Telstra-UOW seed funding, 2023

Contact

jiey@uow.edu.au

University of Wollongong, NSW, 2522