RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns

• URL: http://arxiv.org/abs/2508.13152v1
• Date: Mon, 18 Aug 2025 17:59:15 GMT
• Title: RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns
• Authors: Xin Chen, Junchao Wu, Shu Yang, Runzhe Zhan, Zeyu Wu, Ziyang Luo, Di Wang, Min Yang, Lidia S. Chao, Derek F. Wong,
• Abstract summary: Large language models (LLMs) are crucial for preventing misuse and building trustworthy AI systems.<n>We propose RepreGuard, an efficient statistics-based detection method.<n> Experimental results show that RepreGuard outperforms all baselines with average 94.92% AUROC on both in-distribution (ID) and OOD scenarios.
• Score: 50.401907401444404
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Detecting content generated by large language models (LLMs) is crucial for preventing misuse and building trustworthy AI systems. Although existing detection methods perform well, their robustness in out-of-distribution (OOD) scenarios is still lacking. In this paper, we hypothesize that, compared to features used by existing detection methods, the internal representations of LLMs contain more comprehensive and raw features that can more effectively capture and distinguish the statistical pattern differences between LLM-generated texts (LGT) and human-written texts (HWT). We validated this hypothesis across different LLMs and observed significant differences in neural activation patterns when processing these two types of texts. Based on this, we propose RepreGuard, an efficient statistics-based detection method. Specifically, we first employ a surrogate model to collect representation of LGT and HWT, and extract the distinct activation feature that can better identify LGT. We can classify the text by calculating the projection score of the text representations along this feature direction and comparing with a precomputed threshold. Experimental results show that RepreGuard outperforms all baselines with average 94.92% AUROC on both in-distribution (ID) and OOD scenarios, while also demonstrating robust resilience to various text sizes and mainstream attacks. Data and code are publicly available at: https://github.com/NLP2CT/RepreGuard

Related papers

• Variation is the Key: A Variation-Based Framework for LLM-Generated Text Detection [14.828776526024617]
  VaryBalance is a simple but effective method for detecting text generated by large language models (LLMs)<n>The core of VaryBalance is that, compared to LLM-generated texts, there is a greater difference between human texts and their version rewritten via LLMs.<n> Comprehensive experiments demonstrated that VaryBalance outperforms the state-of-the-art detectors, i.e., Binoculars, by up to 34.3% in terms of AUROC.
  arXiv  Detail & Related papers  (2026-01-27T05:48:39Z)
• Human Texts Are Outliers: Detecting LLM-generated Texts via Out-of-distribution Detection [71.59834293521074]
  We develop a framework to distinguish between human-authored and machine-generated text.<n>Our method achieves 98.3% AUROC and AUPR with only 8.9% FPR95 on DeepFake dataset.<n>Code, pretrained weights, and demo will be released.
  arXiv  Detail & Related papers  (2025-10-07T08:14:45Z)
• You've Changed: Detecting Modification of Black-Box Large Language Models [4.7541096609711]
  Large Language Models (LLMs) are often provided as a service via an API, making it challenging for developers to detect changes in their behavior.<n>We present an approach to monitor LLMs for changes by comparing the distributions of linguistic and psycholinguistic features of generated text.
  arXiv  Detail & Related papers  (2025-04-14T04:16:43Z)
• Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders [20.557610461777344]
  We use Sparse Autoencoders (SAE) to extract features from Gemma-2-2b residual stream.<n>We identify both interpretable and efficient features, analyzing their semantics and relevance.<n>Our methods offer valuable insights into how texts from various models differ from human-written content.
  arXiv  Detail & Related papers  (2025-03-05T15:33:52Z)
• Idiosyncrasies in Large Language Models [54.26923012617675]
  We unveil and study idiosyncrasies in Large Language Models (LLMs)<n>We find that fine-tuning text embedding models on LLM-generated texts yields excellent classification accuracy.<n>We leverage LLM as judges to generate detailed, open-ended descriptions of each model's idiosyncrasies.
  arXiv  Detail & Related papers  (2025-02-17T18:59:02Z)
• SPOT: Text Source Prediction from Originality Score Thresholding [6.790905400046194]
  countermeasures aim at detecting misinformation, usually involve domain specific models trained to recognize the relevance of any information. Instead of evaluating the validity of the information, we propose to investigate LLM generated text from the perspective of trust.
  arXiv  Detail & Related papers  (2024-05-30T21:51:01Z)
• Who Wrote This? The Key to Zero-Shot LLM-Generated Text Detection Is GECScore [51.65730053591696]
  We propose a simple yet effective black-box zero-shot detection approach based on the observation that human-written texts typically contain more grammatical errors than LLM-generated texts.<n> Experimental results show that our method outperforms current state-of-the-art (SOTA) zero-shot and supervised methods.
  arXiv  Detail & Related papers  (2024-05-07T12:57:01Z)
• DPIC: Decoupling Prompt and Intrinsic Characteristics for LLM Generated Text Detection [56.513637720967566]
  Large language models (LLMs) can generate texts that pose risks of misuse, such as plagiarism, planting fake reviews on e-commerce platforms, or creating inflammatory false tweets. Existing high-quality detection methods usually require access to the interior of the model to extract the intrinsic characteristics. We propose to extract deep intrinsic characteristics of the black-box model generated texts.
  arXiv  Detail & Related papers  (2023-05-21T17:26:16Z)
• DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature [143.5381108333212]
  We show that text sampled from an large language model tends to occupy negative curvature regions of the model's log probability function. We then define a new curvature-based criterion for judging if a passage is generated from a given LLM. We find DetectGPT is more discriminative than existing zero-shot methods for model sample detection.
  arXiv  Detail & Related papers  (2023-01-26T18:44:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.