Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces MD-ProTector, a novel approach for detecting LLM-generated text by utilizing multiple trainable reference vectors, or prototypes, in the encoder embedding space to better capture the diversity within text classes. By employing a Prototype Positioning loss, the method effectively separates class-level structure from within-class variations, allowing for distinct decision boundaries for different text groups. Evaluations across multiple benchmarks demonstrate that MD-ProTector outperforms existing encoder-based methods in key metrics, including AvgRec, AUROC, and FPR95.
MD-ProTector achieves unprecedented accuracy in distinguishing LLM-generated text by leveraging multiple prototypes to capture nuanced variations within text classes.
As LLM-generated content becomes more sophisticated, detection systems for distinguishing those texts from human-written text must operate at scale while handling diverse writing styles, domains, languages, and generator models. Input-only encoder detectors are suitable for practical deployment setting, but standard binary classification supplies only the class label and does not explicitly organize the substantial variation within either class. We propose MD-ProTector, which represents each class with multiple trainable reference vectors in the encoder embedding space, referred to as prototypes. These prototypes provide separate decision boundaries for different groups of texts within the same class. However, adding multiple prototypes alone does not determine which variation each prototype should represent. MD-ProTector addresses this problem with Prototype Positioning loss, which separates class-level structure from the within-class variation that differentiates individual prototypes. Evaluated across five settings from three large-scale benchmarks covering domain, generator, language, and adversarial variation, MD-ProTector achieves the highest AvgRec on MAGE CDCM and RAID and the highest AUROC and lowest FPR95 on RAID among the compared encoder-based methods.