Thesis Defense: Self-Training for Natural Language Processing
, CSAIL MIT
Date: Friday, April 29, 2022
Time: 11:00 AM to 12:00 PM Note: all times are in the Eastern Time Zone
Event Type: Thesis Defense
Host: James Glass, CSAIL MIT
Contact: Hongyin Luo, email@example.com
Relevant URL: https://mit.zoom.us/j/98047395332
Speaker URL: https://www.csail.mit.edu/person/hongyin-luo
TALK: Thesis Defense: Self-Training for Natural Language Processing
For remote access to this event: https://mit.zoom.us/j/98047395332
Thesis advisor: James Glass
Thesis committee: Peter Szolovits, Yoon Kim
Data annotation is critical for machine learning based natural language processing models. Although many large-scale corpora and standard benchmarks have been annotated and published, they cannot cover all possible applications. As a result, it is difficult to transfer models trained with public corpora to tasks that require domain-specific knowledge, different inference skills, unseen text styles, and explainability. In this thesis, we explore self-training methods for mitigating the data distribution gaps between training and evaluation domains and tasks. In contrast to traditional self-training methods that study the best practice of training models with real data and pseudo labels, we also explore the possibility of automatically generating synthetic data for better explainability, robustness, and domain adaptation performance. We show the performance improvement achieved by our methods on different natural language understanding and generation tasks, including question answering, question generation, and dialog response selection.
AI & Machine Learning
Created by Hongyin Luo at Monday, April 11, 2022 at 4:24 PM.