SynthBio: A Case Study in Human-AI Collaborative Curation of Text Datasets
By Ann Yuan, Daphne Ippolito, Vitaly Nikolaev, Chris Callison-Burch, Andy Coenen, Sebastian Gehrmann. 2021
25 citations recorded by the source corpus.
NLP researchers need more, higher-quality text datasets. Human-labeled datasets are expensive to collect, while datasets collected via automatic retrieval from the web such as WikiBio are noisy and can include undesired biases. Moreover, data sourced from the web is often included in datasets used to pretrain models, leading to inadvertent cross-contamination of training and test sets. In this work we introduce a novel method for efficient dataset curation: we use a large language model to provide seed generations to human raters, thereby changing dataset authoring from a writing task to an editing task. We use our method to curate SynthBio - a new evaluation set for WikiBio - composed of structured attribute lists describing fictional individuals, mapped to natural language biographies. We show that our dataset of fictional biographies is less noisy than WikiBio, and also more balanced with respect to gender and nationality.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Luke S. Zettlemoyer
- Quoc V. Le
- Mohit Bansal
- Noah A. Smith
- Yejin Choi
- Percy Liang
- Jianfeng Gao
- Graham Neubig
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Quoc V. Le
- Trevor Darrell
- Mohit Bansal
- Xiaogang Wang
- Jitendra Malik
- Ming-Hsuan Yang
In Natural Language Processing
- Yejin Choi
- Mohit Bansal
- Noah A. Smith
- Graham Neubig
- Luke S. Zettlemoyer
- Hinrich Schütze
- Hannaneh Hajishirzi
- Yoav Goldberg
Citations in the corpus
- Scaling Instruction-Finetuned Language Models (2022, methodology): “The SynthBio evaluation set is mined from a subset of (Yuan et al., 2021), which includes synthetically generated English biography passages.”
- WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creation (2022, background): “(Yuan et al., 2021).”
- TaTa: A Multilingual Table-to-Text Dataset for African Languages (2022, methodology): “SynthBio (Yuan et al., 2021) followed a similar strategy but started with text generated by a large language model (Thop-pilan et al., 2022).”