The rapid improvement of language models has raised the specter of abuse of text generation systems. This progress motivates the development of simple methods for detecting generated text that can be used by and explained to non-experts. We develop GLTR, a tool to support humans in detecting whether a text was generated by a model. GLTR applies a suite of baseline statistical methods that can detect generation artifacts across common sampling schemes. In a human-subjects study, we show that the annotation scheme provided by GLTR improves the human detection-rate of fake text from 54% to 72% without any prior training. GLTR is open-source and publicly deployed, and has already been widely used to detect generated outputs
GLTR: Statistical Detection and Visualization of Generated Text
GLTR is a statistical tool that enhances human detection of generated text from models through an annotation scheme, improving detection rates without prior training.
- Year
- 2019
- Venue
- gltr-statistical-detection-and-visualization-1
- Authors
- 3
- Hosting
- Abstract onlyARXIV-DEFAULT
Cite
Notes
Only stored in your browser.
Attribution
- Abstract & full text
- arxiv.org/abs/1906.04043ARXIV-DEFAULT
- TL;DR
- Semantic Scholar