Summary of Abex: Data Augmentation For Low-resource Nlu Via Expanding Abstract Descriptions, by Sreyan Ghosh and Utkarsh Tyagi and Sonal Kumar and C. K. Evuru and S Ramaneswaran and S Sakshi and Dinesh Manocha
ABEX: Data Augmentation for Low-Resource NLU via Expanding Abstract Descriptions
by Sreyan Ghosh, Utkarsh Tyagi, Sonal Kumar, C. K. Evuru, S Ramaneswaran, S Sakshi, Dinesh Manocha
First submitted to arxiv on: 6 Jun 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 The proposed ABEX methodology offers a novel approach to generative data augmentation for low-resource Natural Language Understanding (NLU) tasks. By converting documents into concise abstract descriptions and then generating new documents based on expanding these abstractions, ABEX preserves the original semantic properties of the documents while facilitating diverse generations. The method consists of two stages: training BART on a synthetic dataset with abstract-document pairs to learn the task of expanding abstract descriptions, and proposing a simple, controllable, and training-free method for generating abstract descriptions based on editing AMR graphs. ABEX outperforms baselines qualitatively, achieving improvements of 0.04% – 38.8%, and exhibits context and length diversity in its generated outputs. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary ABEX is a new way to generate text that helps machines understand language better. It starts by making a summary of an original text, then uses this summary to create new texts that are similar but not the same. This approach keeps the important information from the original text and makes it easier for machines to learn about language. The results show that ABEX does a great job on several tasks and can generate diverse texts. |
Keywords
» Artificial intelligence » Data augmentation » Language understanding
