README.md

CapstoneDesing(CS4102) 2020 Fall Semester

"Question Answering using Queestion Generation"

Team: 'Information Retrievers'

1. Joochan Kim(Idea, Train model, Test, Presentation)
2. Sanghyun Lee(Idea, Code, Test, Presentaion)
3. Junyoung Rhee(Idea, Train model, Test, Presentation)

Video PPT

---

Idea originated from Assist. Hojae Han @ Data Intelligence Lab (Currently moved to SNU)

Advisor Seungwon Hwang @ Data Intelligence Lab (Currently moved to SNU)

Question Answering task is vulnerable to Adversarial Attack and QA model has a possibility to learn unnecessary dependency between Question and Answer. To solve this problem, we propose a new approach to imporve Question Answering model using Question Generation.

---

This Repository doesn't contains full sources. Only contains what I've done.

---

Result

---

References (Has dependency to those source file)

• BERT
  • https://github.com/huggingface/transformers/tree/master/examples/pytorch/question-answering
• GPT-2
  • https://github.com/huggingface/transformers/tree/master/examples/pytorch/language-modeling
  • https://github.com/huggingface/transformers/tree/master/examples/pytorch/text-generation