Paper Image

Detecting AI-Generated Text

Published on:

6 September 2023

Primary Category:

Computation and Language

Paper Authors:

Zhenpeng Su,

Xing Wu,

Wei Zhou,

Guangyuan Ma,

Songlin Hu


Key Details

Current detectors fail on semantic-invariant tasks like summarization

Proposed a new dataset with more hard semantic-invariant samples

Also created a better detector using instruction tuning

Their detector outperforms previous state-of-the-art methods

AI generated summary

Detecting AI-Generated Text

This paper introduces a new dataset and method for detecting text generated by AI models like ChatGPT. The authors found existing detectors struggled on semantic-invariant tasks like summarization and translation. Their dataset has more of these challenging samples. They also propose a detector using instruction tuning that outperforms previous methods.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up