Datasets:

Modalities:
Text
Formats:
json
Languages:
English
Size:
< 1K
ArXiv:
Libraries:
Datasets
pandas
License:
WildSeek / README.md
YuchengJiang's picture
Update README.md
f890ba1 verified
metadata
license: cc-by-sa-4.0
task_categories:
  - text-generation
language:
  - en
size_categories:
  - n<1K

WildSeek Dataset

To study users’ interests in complex information seeking tasks in the wild, we utilized data collected from the open-source STORM web application, which generates comprehensive long-form reports based on users’ specified topics of interest and goals for using the site. Each data point is a pair comprising a topic and the user’s goal. For more details, please refer to Section 2.2 and Appendix A of our paper.

This dataset could be valuable for researchers working on tasks like report generation, knowledge curation, information retrieval, etc. The text data in this dataset is licensed under the Creative Commons Attribution-ShareAlike (CC BY-SA) license.

Please cite our paper if you found the dataset useful for your research.

@misc{jiang2024unknownunknowns,
      title={Into the Unknown Unknowns: Engaged Human Learning through Participation in Language Model Agent Conversations}, 
      author={Yucheng Jiang and Yijia Shao and Dekun Ma and Sina J. Semnani and Monica S. Lam},
      year={2024},
      eprint={2408.15232},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2408.15232}, 
}