Summary of Explaining Mixtures Of Sources in News Articles, by Alexander Spangher et al.
Explaining Mixtures of Sources in News Articles
by Alexander Spangher, James Youn, Matt DeButts, Nanyun Peng, Emilio Ferrara, Jonathan May
First submitted to arxiv on: 7 Nov 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 abstract explores how large language models (LLMs) can be used in longer-form article generation by understanding the planning steps humans make before writing. The authors examine one type of planning, source-selection in news, as a case-study to evaluate plans in long-form generation. They propose a generative process for story writing where a source-selection schema is first selected and then sources are chosen based on categories in that schema. To develop this concept, the authors adapt five existing schemata and introduce three new ones to describe journalistic plans for including sources in documents. Using Bayesian latent-variable modeling, they create metrics to select the most likely plan or schema underlying a story. The results show that two schemata (stance and social affiliation) best explain source plans in most documents, while other schemata like textual entailment explain source plans in factually rich topics like “Science”. Additionally, the authors demonstrate that they can predict the most suitable schema given just an article’s headline with reasonable accuracy. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research paper explores how artificial intelligence (AI) can help write longer articles. The authors studied how journalists plan their work and chose which sources to use in news stories. They developed a new way of thinking about this process by creating “schemas” that describe different planning strategies. These schemas helped the authors understand why certain types of sources are used for specific kinds of stories. The study found that some schemas were better at explaining how journalists choose sources than others. This research could help AI systems write more realistic and informative articles. |
