Summary of Explainable News Summarization — Analysis and Mitigation Of Disagreement Problem, by Seema Aswani et al.
Explainable News Summarization – Analysis and mitigation of Disagreement Problem
by Seema Aswani, Sujala D. Shetty
First submitted to arxiv on: 24 Oct 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
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 addresses a major challenge in Explainable AI (XAI) for text summarization: the disagreement problem. Different XAI methods generate contradictory explanations for the same input article, hindering reliability and interpretability. The authors propose a novel approach using sentence transformers and k-means clustering to segment the input article and generate regional explanations. This method reduces observed disagreement between XAI methods when evaluated on two news summarization datasets: XSum and CNN-DailyMail. Experimental results validate this hypothesis, showing a significant decrease in disagreement among different XAI methods. Additionally, a JavaScript visualization tool is developed to interactively explore attribution scores for each sentence. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps solve a problem with AI that explains how it summarizes text. Right now, different ways of explaining how AI does this disagree on what the summary means. To fix this, the authors came up with a new idea. They split the original article into smaller parts and explain each part separately. This makes different AI methods agree more often when they summarize the same text. |
Keywords
» Artificial intelligence » Clustering » Cnn » K means » Summarization
