HellaSwag: Can a Machine Really Finish Your Sentence?
University of Washington · Allen Institute for Artificial Intelligence
Abstract
Introduced a new task of commonsense natural language inference: given an event description such as "A woman sits at a piano," a machine must select the most likely followup: "She sets her fingers on the keys." With the introduction of BERT (Devlin et al., 2018), near human-level performance was reached. Does this mean that machines can perform human level commonsense inference?
Citation impact
- FWCI
- 25.68
- Percentile
- 100%
- References
- 20
Authors
5- RZRowan ZellersCorresponding
University of Washington, Allen Institute for Artificial Intelligence
- AHAri Holtzman
Allen Institute for Artificial Intelligence, University of Washington
- YBYonatan Bisk
Allen Institute for Artificial Intelligence, University of Washington
- AFAli Farhadi
University of Washington, Allen Institute for Artificial Intelligence
- YCYejin Choi
Allen Institute for Artificial Intelligence, University of Washington
Topics & keywords
- Computer science
- Artificial intelligence
- Adversarial system
- Inference
- Language model
- Commonsense reasoning
- Set (abstract data type)
- Machine learning
- Reduced inequalities