Membership Inference Attack against Long-Context Large Language Models

• URL: http://arxiv.org/abs/2411.11424v1
• Date: Mon, 18 Nov 2024 09:50:54 GMT
• Title: Membership Inference Attack against Long-Context Large Language Models
• Authors: Zixiong Wang, Gaoyang Liu, Yang Yang, Chen Wang,
• Abstract summary: We argue that integrating all information into the long context makes it a repository of sensitive information. We propose six membership inference attack strategies tailored for LCLMs. We examine the underlying reasons why LCLMs are susceptible to revealing such membership information.
• Score: 8.788010048413188
• License:
• Abstract: Recent advances in Large Language Models (LLMs) have enabled them to overcome their context window limitations, and demonstrate exceptional retrieval and reasoning capacities on longer context. Quesion-answering systems augmented with Long-Context Language Models (LCLMs) can automatically search massive external data and incorporate it into their contexts, enabling faithful predictions and reducing issues such as hallucinations and knowledge staleness. Existing studies targeting LCLMs mainly concentrate on addressing the so-called lost-in-the-middle problem or improving the inference effiencicy, leaving their privacy risks largely unexplored. In this paper, we aim to bridge this gap and argue that integrating all information into the long context makes it a repository of sensitive information, which often contains private data such as medical records or personal identities. We further investigate the membership privacy within LCLMs external context, with the aim of determining whether a given document or sequence is included in the LCLMs context. Our basic idea is that if a document lies in the context, it will exhibit a low generation loss or a high degree of semantic similarity to the contents generated by LCLMs. We for the first time propose six membership inference attack (MIA) strategies tailored for LCLMs and conduct extensive experiments on various popular models. Empirical results demonstrate that our attacks can accurately infer membership status in most cases, e.g., 90.66% attack F1-score on Multi-document QA datasets with LongChat-7b-v1.5-32k, highlighting significant risks of membership leakage within LCLMs input contexts. Furthermore, we examine the underlying reasons why LCLMs are susceptible to revealing such membership information.

Related papers

• NewsInterview: a Dataset and a Playground to Evaluate LLMs' Ground Gap via Informational Interviews [65.35458530702442]
  We focus on journalistic interviews, a domain rich in grounding communication and abundant in data. We curate a dataset of 40,000 two-person informational interviews from NPR and CNN. LLMs are significantly less likely than human interviewers to use acknowledgements and to pivot to higher-level questions.
  arXiv  Detail & Related papers  (2024-11-21T01:37:38Z)
• Information Anxiety in Large Language Models [21.574677910096735]
  Large Language Models (LLMs) have demonstrated strong performance as knowledge repositories. We take the investigation further by conducting a comprehensive analysis of the internal reasoning and retrieval mechanisms of LLMs. Our work focuses on three critical dimensions - the impact of entity popularity, the models' sensitivity to lexical variations in query formulation, and the progression of hidden state representations.
  arXiv  Detail & Related papers  (2024-11-16T14:28:33Z)
• LLM-PBE: Assessing Data Privacy in Large Language Models [111.58198436835036]
  Large Language Models (LLMs) have become integral to numerous domains, significantly advancing applications in data management, mining, and analysis. Despite the critical nature of this issue, there has been no existing literature to offer a comprehensive assessment of data privacy risks in LLMs. Our paper introduces LLM-PBE, a toolkit crafted specifically for the systematic evaluation of data privacy risks in LLMs.
  arXiv  Detail & Related papers  (2024-08-23T01:37:29Z)
• ReCaLL: Membership Inference via Relative Conditional Log-Likelihoods [56.073335779595475]
  We propose ReCaLL (Relative Conditional Log-Likelihood), a novel membership inference attack (MIA) ReCaLL examines the relative change in conditional log-likelihoods when prefixing target data points with non-member context. We conduct comprehensive experiments and show that ReCaLL achieves state-of-the-art performance on the WikiMIA dataset.
  arXiv  Detail & Related papers  (2024-06-23T00:23:13Z)
• CLAMBER: A Benchmark of Identifying and Clarifying Ambiguous Information Needs in Large Language Models [60.59638232596912]
  We introduce CLAMBER, a benchmark for evaluating large language models (LLMs) Building upon the taxonomy, we construct 12K high-quality data to assess the strengths, weaknesses, and potential risks of various off-the-shelf LLMs. Our findings indicate the limited practical utility of current LLMs in identifying and clarifying ambiguous user queries.
  arXiv  Detail & Related papers  (2024-05-20T14:34:01Z)
• Locally Differentially Private In-Context Learning [8.659575019965152]
  Large pretrained language models (LLMs) have shown surprising In-Context Learning (ICL) ability. This paper proposes a locally differentially private framework of in-context learning (LDP-ICL) Considering the mechanisms of in-context learning in Transformers by gradient descent, we provide an analysis of the trade-off between privacy and utility in such LDP-ICL.
  arXiv  Detail & Related papers  (2024-05-07T06:05:43Z)
• On Protecting the Data Privacy of Large Language Models (LLMs): A Survey [35.48984524483533]
  Large language models (LLMs) are complex artificial intelligence systems capable of understanding, generating and translating human language. LLMs process and generate large amounts of data, which may threaten data privacy.
  arXiv  Detail & Related papers  (2024-03-08T08:47:48Z)
• Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts When Knowledge Conflicts? [45.233517779029334]
  We identify whether responses are attributed to generated or retrieved contexts. Experiments reveal a significant bias in several LLMs to favor generated contexts, even when they provide incorrect information.
  arXiv  Detail & Related papers  (2024-01-22T12:54:04Z)
• Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentation [109.8527403904657]
  We show that large language models (LLMs) possess unwavering confidence in their knowledge and cannot handle the conflict between internal and external knowledge well. Retrieval augmentation proves to be an effective approach in enhancing LLMs' awareness of knowledge boundaries. We propose a simple method to dynamically utilize supporting documents with our judgement strategy.
  arXiv  Detail & Related papers  (2023-07-20T16:46:10Z)
• On the Risk of Misinformation Pollution with Large Language Models [127.1107824751703]
  We investigate the potential misuse of modern Large Language Models (LLMs) for generating credible-sounding misinformation. Our study reveals that LLMs can act as effective misinformation generators, leading to a significant degradation in the performance of Open-Domain Question Answering (ODQA) systems.
  arXiv  Detail & Related papers  (2023-05-23T04:10:26Z)
• Assessing Hidden Risks of LLMs: An Empirical Study on Robustness, Consistency, and Credibility [37.682136465784254]
  We conduct over a million queries to the mainstream large language models (LLMs) including ChatGPT, LLaMA, and OPT. We find that ChatGPT is still capable to yield the correct answer even when the input is polluted at an extreme level. We propose a novel index associated with a dataset that roughly decides the feasibility of using such data for LLM-involved evaluation.
  arXiv  Detail & Related papers  (2023-05-15T15:44:51Z)

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.