Summary of Constructing Cloze Questions Generatively, by Yicheng Sun (1) and Jie Wang (2)
Constructing Cloze Questions Generatively
by Yicheng Sun, Jie Wang
First submitted to arxiv on: 5 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents a novel generative method called CQG, which constructs cloze questions from articles using neural networks and WordNet. The approach emphasizes generating multigram distractors that are semantically similar to the correct answer. CQG employs sense disambiguation, text-to-text transformation, and lexical labels from WordNet’s synset taxonomies to select an answer key for a given sentence. It then segments the sentence into instances, generates instance-level distractor candidates using transformers, and ranks them based on contextual embedding similarities and synset-relatedness. The top-ranked distractors are selected as legitimate phrases by replacing instances with corresponding IDCs. Experimental results show that CQG outperforms state-of-the-art (SOTA) methods, and human judges confirm the high quality of generated distractors. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to make questions from articles using computers and a big dictionary called WordNet. The method is called CQG and it helps create tricky options that are similar in meaning to the correct answer. CQG uses special algorithms to choose the right words for each sentence, then mixes them up to create different answers. This makes it hard for students to cheat on tests! The paper shows that this new way works better than previous methods. |
Keywords
» Artificial intelligence » Embedding
