Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

EAGLE: A Domain Generalization Framework for AI-generated Text Detection

About

With the advancement in capabilities of Large Language Models (LLMs), one major step in the responsible and safe use of such LLMs is to be able to detect text generated by these models. While supervised AI-generated text detectors perform well on text generated by older LLMs, with the frequent release of new LLMs, building supervised detectors for identifying text from such new models would require new labeled training data, which is infeasible in practice. In this work, we tackle this problem and propose a domain generalization framework for the detection of AI-generated text from unseen target generators. Our proposed framework, EAGLE, leverages the labeled data that is available so far from older language models and learns features invariant across these generators, in order to detect text generated by an unknown target generator. EAGLE learns such domain-invariant features by combining the representational power of self-supervised contrastive learning with domain adversarial training. Through our experiments we demonstrate how EAGLE effectively achieves impressive performance in detecting text generated by unseen target generators, including recent state-of-the-art ones such as GPT-4 and Claude, reaching detection scores of within 4.7% of a fully supervised detector.

Amrita Bhattacharjee, Raha Moraffah, Joshua Garland, Huan Liu• 2024

Related benchmarks

TaskDatasetResultRank
Machine-generated text detectionSQuAD--
30
MGT detectionYelp
Accuracy91.36
16
MGT detectionHSWAG
Accuracy95.8
16
MGT detectionXsum
Accuracy88.08
16
MGT detectionELI5
Accuracy82.31
16
MGT detectionWP
Accuracy82.84
16
MGT detectionSCI
Accuracy90.47
16
MGT detectionCMV
Accuracy78.54
16
MGT detectionTLDR
Accuracy65.85
16
MGT detectionROCT
Accuracy57.36
16
Showing 10 of 20 rows

Other info

Follow for update