Inner-Probe: Discovering Copyright-related Data Generation in LLM Architecture

• URL: http://arxiv.org/abs/2410.04454v2
• Date: Thu, 23 Jan 2025 09:11:30 GMT
• Title: Inner-Probe: Discovering Copyright-related Data Generation in LLM Architecture
• Authors: Qichao Ma, Rui-Jie Zhu, Peiye Liu, Renye Yan, Fahong Zhang, Ling Liang, Meng Li, Zhaofei Yu, Zongwei Wang, Yimao Cai, Tiejun Huang,
• Abstract summary: 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.
• Score: 39.425944445393945
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models (LLMs) utilize extensive knowledge databases and show powerful text generation ability. However, their reliance on high-quality copyrighted datasets raises concerns about copyright infringements in generated texts. Current research often employs prompt engineering or semantic classifiers to identify copyrighted content, but these approaches have two significant limitations: (1) Challenging to identify which specific sub-dataset (e.g., works from particular authors) influences an LLM's output. (2) Treating the entire training database as copyrighted, hence overlooking the inclusion of non-copyrighted training data. We propose InnerProbe, a lightweight framework designed to evaluate the influence of copyrighted sub-datasets on LLM-generated texts. Unlike traditional methods relying solely on text, we discover that the results of multi-head attention (MHA) during LLM output generation provide more effective information. Thus, InnerProbe performs sub-dataset contribution analysis using a lightweight LSTM-based network trained on MHA results in a supervised manner. Harnessing such a prior, InnerProbe enables non-copyrighted text detection through a concatenated global projector trained with unsupervised contrastive learning. InnerProbe demonstrates 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.

Related papers

• Information-Guided Identification of Training Data Imprint in (Proprietary) Large Language Models [52.439289085318634]
  We show how to identify training data known to proprietary large language models (LLMs) by using information-guided probes. Our work builds on a key observation: text passages with high surprisal are good search material for memorization probes.
  arXiv  Detail & Related papers  (2025-03-15T10:19:15Z)
• Robust Detection of LLM-Generated Text: A Comparative Analysis [0.276240219662896]
  Large language models can be widely integrated into many aspects of life, and their output can quickly fill all network resources. It becomes increasingly important to develop powerful detectors for the generated text. This detector is essential to prevent the potential misuse of these technologies and to protect areas such as social media from the negative effects.
  arXiv  Detail & Related papers  (2024-11-09T18:27:15Z)
• A Bayesian Approach to Harnessing the Power of LLMs in Authorship Attribution [57.309390098903]
  Authorship attribution aims to identify the origin or author of a document. Large Language Models (LLMs) with their deep reasoning capabilities and ability to maintain long-range textual associations offer a promising alternative. Our results on the IMDb and blog datasets show an impressive 85% accuracy in one-shot authorship classification across ten authors.
  arXiv  Detail & Related papers  (2024-10-29T04:14:23Z)
• Measuring Copyright Risks of Large Language Model via Partial Information Probing [14.067687792633372]
  We explore the data sources used to train Large Language Models (LLMs) We input a portion of a copyrighted text into LLMs, prompt them to complete it, and then analyze the overlap between the generated content and the original copyrighted material. Our findings demonstrate that LLMs can indeed generate content highly overlapping with copyrighted materials based on these partial inputs.
  arXiv  Detail & Related papers  (2024-09-20T18:16:05Z)
• 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. We find that watermarking adversely affects the success rate of Membership Inference Attacks (MIAs) 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)
• Entropy Law: The Story Behind Data Compression and LLM Performance [115.70395740286422]
  We find that model performance is negatively correlated to the compression ratio of training data, which usually yields a lower training loss. Based on the findings of the entropy law, we propose a quite efficient and universal data selection method. We also present an interesting application of entropy law that can detect potential performance risks at the beginning of model training.
  arXiv  Detail & Related papers  (2024-07-09T08:14:29Z)
• Evaluating Copyright Takedown Methods for Language Models [100.38129820325497]
  Language models (LMs) derive their capabilities from extensive training on diverse data, including potentially copyrighted material. This paper introduces the first evaluation of the feasibility and side effects of copyright takedowns for LMs. We examine several strategies, including adding system prompts, decoding-time filtering interventions, and unlearning approaches.
  arXiv  Detail & Related papers  (2024-06-26T18:09:46Z)
• Peering into the Mind of Language Models: An Approach for Attribution in Contextual Question Answering [9.86691461253151]
  We introduce a novel method for attribution in contextual question answering, leveraging the hidden state representations of large language models (LLMs) Our approach bypasses the need for extensive model retraining and retrieval model overhead, offering granular attributions and preserving the quality of generated answers. We present Verifiability-granular, an attribution dataset which has token level annotations for LLM generations in the contextual question answering setup.
  arXiv  Detail & Related papers  (2024-05-28T09:12:44Z)
• Data Augmentation for Text-based Person Retrieval Using Large Language Models [16.120524750964016]
  Text-based Person Retrieval (TPR) aims to retrieve person images that match the description given a text query. It is difficult to construct a large-scale, high-quality TPR dataset due to expensive annotation and privacy protection. This paper proposes an LLM-based Data Augmentation (LLM-DA) method for TPR.
  arXiv  Detail & Related papers  (2024-05-20T11:57:50Z)
• 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. 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)
• Unsupervised Information Refinement Training of Large Language Models for Retrieval-Augmented Generation [128.01050030936028]
  We propose an information refinement training method named InFO-RAG. InFO-RAG is low-cost and general across various tasks. It improves the performance of LLaMA2 by an average of 9.39% relative points.
  arXiv  Detail & Related papers  (2024-02-28T08:24:38Z)
• Digger: Detecting Copyright Content Mis-usage in Large Language Model Training [23.99093718956372]
  We introduce a framework designed to detect and assess the presence of content from potentially copyrighted books within the training datasets of Large Language Models (LLMs) This framework also provides a confidence estimation for the likelihood of each content sample's inclusion.
  arXiv  Detail & Related papers  (2024-01-01T06:04:52Z)
• Source Attribution for Large Language Model-Generated Data [57.85840382230037]
  It is imperative to be able to perform source attribution by identifying the data provider who contributed to the generation of a synthetic text. We show that this problem can be tackled by watermarking. We propose a source attribution framework that satisfies these key properties due to our algorithmic designs.
  arXiv  Detail & Related papers  (2023-10-01T12:02:57Z)

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.