English

Picking Apart Story Salads

Computation and Language 2018-11-01 v1

Abstract

During natural disasters and conflicts, information about what happened is often confusing, messy, and distributed across many sources. We would like to be able to automatically identify relevant information and assemble it into coherent narratives of what happened. To make this task accessible to neural models, we introduce Story Salads, mixtures of multiple documents that can be generated at scale. By exploiting the Wikipedia hierarchy, we can generate salads that exhibit challenging inference problems. Story salads give rise to a novel, challenging clustering task, where the objective is to group sentences from the same narratives. We demonstrate that simple bag-of-words similarity clustering falls short on this task and that it is necessary to take into account global context and coherence.

Keywords

Cite

@article{arxiv.1810.13391,
  title  = {Picking Apart Story Salads},
  author = {Su Wang and Eric Holgate and Greg Durrett and Katrin Erk},
  journal= {arXiv preprint arXiv:1810.13391},
  year   = {2018}
}

Comments

Accepted at EMNLP 2018 (long paper)

R2 v1 2026-06-23T04:59:22.179Z