# MD-ProTector: Positioning Multiple Data-Driven Prototypes for LLM-Generated Text Detection

DevFeed: [MD-ProTector: Positioning Multiple Data-Driven Prototypes for LLM-Generated Text Detection](<https://devfeed.tech/articles/md-protector-positioning-multiple-data-driven-prototypes-for-llm-generated-text-detection-118255.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2608.10459>)

Author: Jinmo Han, Jimin Hong, Chanyeong Moon, Ju Yeon Kang, Seonuk Kim, Nam Soo Kim

Published: 2026-11-05T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [adversarial](<https://devfeed.tech/tags/adversarial.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [class](<https://devfeed.tech/tags/class.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [detection](<https://devfeed.tech/tags/detection.md>), [domain](<https://devfeed.tech/tags/domain.md>), [domains](<https://devfeed.tech/tags/domains.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [generated-content](<https://devfeed.tech/tags/generated-content.md>), [generator](<https://devfeed.tech/tags/generator.md>), [language](<https://devfeed.tech/tags/language.md>), [languages](<https://devfeed.tech/tags/languages.md>), [llm](<https://devfeed.tech/tags/llm.md>), [models](<https://devfeed.tech/tags/models.md>), [multiple](<https://devfeed.tech/tags/multiple.md>), [positioning](<https://devfeed.tech/tags/positioning.md>), [prototypes](<https://devfeed.tech/tags/prototypes.md>), [text](<https://devfeed.tech/tags/text.md>)

## AI overview

MD-ProTector detects LLM-generated text by assigning each class multiple trainable prototypes in an encoder's embedding space, giving different groups of texts within a class separate decision boundaries. Its Prototype Positioning loss structures class-level separation while differentiating the within-class variation represented by each prototype. Across five settings in three large-scale benchmarks covering domain, generator, language, and adversarial variation, it achieved the highest AvgRec on MAGE CDCM and RAID, and the highest AUROC and lowest FPR95 on RAID among the compared encoder-based methods.

## Source excerpt

arXiv:2608.10459v2 Announce Type: replace-cross Abstract: 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.