ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2502.12658
42
1
v1v2 (latest)

R.R.: Unveiling LLM Training Privacy through Recollection and Ranking

18 February 2025
Wenlong Meng
Zhenyuan Guo
Lenan Wu
Chen Gong
Wenyan Liu
Weixian Li
Chengkun Wei
Wenzhi Chen
    PILM
ArXiv (abs)PDFHTML
Main:8 Pages
10 Figures
Bibliography:2 Pages
7 Tables
Appendix:3 Pages
Abstract

Large Language Models (LLMs) pose significant privacy risks, potentially leaking training data due to implicit memorization. Existing privacy attacks primarily focus on membership inference attacks (MIAs) or data extraction attacks, but reconstructing specific personally identifiable information (PII) in LLMs' training data remains challenging. In this paper, we propose R.R. (Recollect and Rank), a novel two-step privacy stealing attack that enables attackers to reconstruct PII entities from scrubbed training data where the PII entities have been masked. In the first stage, we introduce a prompt paradigm named recollection, which instructs the LLM to repeat a masked text but fill in masks. Then we can use PII identifiers to extract recollected PII candidates. In the second stage, we design a new criterion to score each PII candidate and rank them. Motivated by membership inference, we leverage the reference model as a calibration to our criterion. Experiments across three popular PII datasets demonstrate that the R.R. achieves better PII identification performance than baselines. These results highlight the vulnerability of LLMs to PII leakage even when training data has been scrubbed. We release our code and datasets at GitHub.

View on arXiv
@article{meng2025_2502.12658,
  title={ R.R.: Unveiling LLM Training Privacy through Recollection and Ranking },
  author={ Wenlong Meng and Zhenyuan Guo and Lenan Wu and Chen Gong and Wenyan Liu and Weixian Li and Chengkun Wei and Wenzhi Chen },
  journal={arXiv preprint arXiv:2502.12658},
  year={ 2025 }
}
Comments on this paper