Explaining Mixtures of Sources in News Articles

Alexander Spangher, James Youn, Matt DeButts, Nanyun Peng, Emilio Ferrara, Jonathan May


Abstract
Human writers plan, _then_ write. For large language models (LLMs) to play a role in longer-form article generation, we must understand the planning steps humans make before writing. We explore one kind of planning, source-selection in news, as a case-study for evaluating plans in long-form generation. We ask: why do _specific_ stories call for _specific_ kinds of sources? We imagine a generative process for story writing where a source-selection schema is first selected by a journalist, and then sources are chosen based on categories in that schema. Learning the article’s _plan_ means predicting the schema initially chosen by the journalist. Working with professional journalists, we adapt five existing schemata and introduce three new ones to describe journalistic plans for the inclusion of sources in documents. Then, inspired by Bayesian latent-variable modeling, we develop metrics to select the most likely plan, or schema, underlying a story, which we use to compare schemata. We find that two schemata: _stance_ and _social affiliation_ best explain source plans in most documents. However, other schemata like _textual entailment_ explain source plans in factually rich topics like “Science”. Finally, we find we can predict the most suitable schema given just the article’s headline with reasonable accuracy. We see this as an important case-study for human planning, and provides a framework and approach for evaluating other kinds of plans, like discourse or plot-oriented plans. We release a corpora, _NewsSources_, with annotations for 4M articles, for further study.
Anthology ID:
2024.findings-emnlp.930
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2024
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
15837–15859
Language:
URL:
https://aclanthology.org/2024.findings-emnlp.930
DOI:
10.18653/v1/2024.findings-emnlp.930
Bibkey:
Cite (ACL):
Alexander Spangher, James Youn, Matt DeButts, Nanyun Peng, Emilio Ferrara, and Jonathan May. 2024. Explaining Mixtures of Sources in News Articles. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 15837–15859, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Explaining Mixtures of Sources in News Articles (Spangher et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-emnlp.930.pdf