Spotlights and Blindspots: Evaluation Machine-Generated Text Detection

arXiv cs.CL / 4/21/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper argues that machine-generated text detection is hard to benchmark because available models use inconsistent datasets, metrics, and evaluation strategies.
It evaluates 15 detection models from six different systems plus seven trained models across multiple English test sets, including creative human-written datasets.
The study finds no single detection approach dominates across all tasks, while most models perform well only in specific scenarios.
It shows that reported model performance and ranking can vary greatly depending on dataset and metric choices, and that results tend to degrade on novel human-written texts in high-risk domains.
The authors conclude that “spotlights and blindspots” in evaluation methodology—often overlooked assumptions—are crucial for accurately reflecting true model effectiveness.

Abstract

With the rise of generative language models, machine-generated text detection has become a critical challenge. A wide variety of models is available, but inconsistent datasets, evaluation metrics, and assessment strategies obscure comparisons of model effectiveness. To address this, we evaluate 15 different detection models from six distinct systems, as well as seven trained models, across seven English-language textual test sets and three creative human-written datasets. We provide an empirical analysis of model performance, the influence of training and evaluation data, and the impact of key metrics. We find that no single system excels in all areas and nearly all are effective for certain tasks, and the representation of model performance is critically linked to dataset and metric choices. We find high variance in model ranks based on datasets and metrics, and overall poor performance on novel human-written texts in high-risk domains. Across datasets and metrics, we find that methodological choices that are often assumed or overlooked are essential for clearly and accurately reflecting model performance.