ISACL: Internal State Analyzer for Copyrighted Training Data Leakage

• URL: http://arxiv.org/abs/2508.17767v2
• Date: Sat, 13 Sep 2025 01:49:58 GMT
• Title: ISACL: Internal State Analyzer for Copyrighted Training Data Leakage
• Authors: Guangwei Zhang, Qisheng Su, Jiateng Liu, Cheng Qian, Yanzhou Pan, Yanjie Fu, Denghui Zhang,
• Abstract summary: Large Language Models (LLMs) pose risks of inadvertently exposing copyrighted or proprietary data.<n>This study introduces a proactive approach: examining LLMs' internal states before text generation to detect potential leaks.<n> integrated with a Retrieval-Augmented Generation (RAG) system, this framework ensures adherence to copyright and licensing requirements.
• Score: 28.435965753598875
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) have revolutionized Natural Language Processing (NLP) but pose risks of inadvertently exposing copyrighted or proprietary data, especially when such data is used for training but not intended for distribution. Traditional methods address these leaks only after content is generated, which can lead to the exposure of sensitive information. This study introduces a proactive approach: examining LLMs' internal states before text generation to detect potential leaks. By using a curated dataset of copyrighted materials, we trained a neural network classifier to identify risks, allowing for early intervention by stopping the generation process or altering outputs to prevent disclosure. Integrated with a Retrieval-Augmented Generation (RAG) system, this framework ensures adherence to copyright and licensing requirements while enhancing data privacy and ethical standards. Our results show that analyzing internal states effectively mitigates the risk of copyrighted data leakage, offering a scalable solution that fits smoothly into AI workflows, ensuring compliance with copyright regulations while maintaining high-quality text generation. The implementation is available on GitHub.\footnote{https://github.com/changhu73/Internal_states_leakage}

Related papers

• Privacy-Aware In-Context Learning for Large Language Models [12.605629953620495]
  Large language models (LLMs) raise privacy concerns due to potential exposure of sensitive information.<n>We introduce a novel private prediction framework for generating high-quality synthetic text with strong privacy guarantees.
  arXiv  Detail & Related papers  (2025-09-17T01:50:32Z)
• Dataset Protection via Watermarked Canaries in Retrieval-Augmented LLMs [67.0310240737424]
  We introduce a novel approach to safeguard the ownership of text datasets and effectively detect unauthorized use by the RA-LLMs.<n>Our approach preserves the original data completely unchanged while protecting it by inserting specifically designed canary documents into the IP dataset.<n>During the detection process, unauthorized usage is identified by querying the canary documents and analyzing the responses of RA-LLMs.
  arXiv  Detail & Related papers  (2025-02-15T04:56:45Z)
• Investigating the Feasibility of Mitigating Potential Copyright Infringement via Large Language Model Unlearning [0.0]
  Pre-trained Large Language Models (LLMs) have demonstrated remarkable capabilities but also pose risks by learning and generating copyrighted material.<n>We propose Stable Sequential Unlearning (SSU), a novel framework designed to unlearn copyrighted content from LLMs over multiple time steps.<n>SSU sometimes achieves an effective trade-off between unlearning efficacy and general-purpose language abilities, outperforming existing baselines, but it's not a cure-all for unlearning copyrighted material.
  arXiv  Detail & Related papers  (2024-12-16T20:01:06Z)
• Inner-Probe: Discovering Copyright-related Data Generation in LLM Architecture [39.425944445393945]
  InnerProbe is a framework designed to evaluate the influence of copyrighted sub-datasets on generated texts.<n>It uses a lightweight LSTM-based network trained on MHA results in a supervised manner.<n>It achieves 3x improved efficiency compared to semantic model training in sub-dataset contribution analysis on Books3, achieves 15.04%-58.7% higher accuracy over baselines on the Pile, and delivers a 0.104 increase in AUC for non-copyrighted data filtering.
  arXiv  Detail & Related papers  (2024-10-06T11:41:39Z)
• Can Watermarking Large Language Models Prevent Copyrighted Text Generation and Hide Training Data? [62.72729485995075]
  We investigate the effectiveness of watermarking as a deterrent against the generation of copyrighted texts.<n>We find that watermarking adversely affects the success rate of Membership Inference Attacks (MIAs)<n>We propose an adaptive technique to improve the success rate of a recent MIA under watermarking.
  arXiv  Detail & Related papers  (2024-07-24T16:53:09Z)
• Targeted Unlearning with Single Layer Unlearning Gradient [15.374381635334897]
  Machine unlearning methods aim to remove sensitive or unwanted content from trained models.<n>We propose Single Layer Unlearning Gradient computation (SLUG) as an efficient method to unlearn targeted information.
  arXiv  Detail & Related papers  (2024-07-16T15:52:36Z)
• The Good and The Bad: Exploring Privacy Issues in Retrieval-Augmented Generation (RAG) [56.67603627046346]
  Retrieval-augmented generation (RAG) is a powerful technique to facilitate language model with proprietary and private data. In this work, we conduct empirical studies with novel attack methods, which demonstrate the vulnerability of RAG systems on leaking the private retrieval database.
  arXiv  Detail & Related papers  (2024-02-23T18:35:15Z)
• A Dataset and Benchmark for Copyright Infringement Unlearning from Text-to-Image Diffusion Models [52.49582606341111]
  Copyright law confers creators the exclusive rights to reproduce, distribute, and monetize their creative works. Recent progress in text-to-image generation has introduced formidable challenges to copyright enforcement. We introduce a novel pipeline that harmonizes CLIP, ChatGPT, and diffusion models to curate a dataset.
  arXiv  Detail & Related papers  (2024-01-04T11:14:01Z)
• Domain Watermark: Effective and Harmless Dataset Copyright Protection is Closed at Hand [96.26251471253823]
  backdoor-based dataset ownership verification (DOV) is currently the only feasible approach to protect the copyright of open-source datasets. We make watermarked models (trained on the protected dataset) correctly classify some hard' samples that will be misclassified by the benign model.
  arXiv  Detail & Related papers  (2023-10-09T11:23:05Z)
• SILO Language Models: Isolating Legal Risk In a Nonparametric Datastore [159.21914121143885]
  We present SILO, a new language model that manages this risk-performance tradeoff during inference. SILO is built by (1) training a parametric LM on Open License Corpus (OLC), a new corpus we curate with 228B tokens of public domain and permissively licensed text. Access to the datastore greatly improves out of domain performance, closing 90% of the performance gap with an LM trained on the Pile.
  arXiv  Detail & Related papers  (2023-08-08T17:58:15Z)
• Can Copyright be Reduced to Privacy? [23.639303165101385]
  We argue that while algorithmic stability may be perceived as a practical tool to detect copying, such copying does not necessarily constitute copyright infringement. If adopted as a standard for detecting an establishing copyright infringement, algorithmic stability may undermine the intended objectives of copyright law.
  arXiv  Detail & Related papers  (2023-05-24T07:22:41Z)

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.