mlsanityData to 14 Oct 2026

Membership Inference Attack against Large Language Model-based Recommendation Systems: A New Distillation-based Paradigm

arXiv 2511.14763 ↗First posted 16 Sep 2025cs.IR cs.AIcited by 0

On the map: LLM safety and security → Membership inference attacks

Abstract (arXiv)

Membership Inference Attack (MIA) aims to determine whether a specific data sample was included in the training dataset of a target model. Traditional MIA approaches rely on shadow models to mimic target model behavior, but their effectiveness diminishes for Large Language Model (LLM)-based recommendation systems due to the scale and complexity of training data. This paper introduces a novel knowledge distillation-based MIA paradigm tailored for LLM-based recommendation systems. Our method constructs a reference model via distillation, applying distinct strategies for member and non-member data to enhance discriminative capabilities. The paradigm extracts fused features (e.g., confidence, entropy, loss, and hidden layer vectors) from the reference model to train an attack model, overcoming limitations of individual features. Extensive experiments on extended datasets (Last.FM, MovieLens, Book-Crossing, Delicious) and diverse LLMs (T5, GPT-2, LLaMA3) demonstrate that our approach significantly outperforms shadow model-based MIAs and individual-feature baselines. The results show its practicality for privacy attacks in LLM-driven recommender systems.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Membership inference attacks, one of 12 directions in LLM safety and security. Builds on 14 of its 21 references that are in the map. First posted on arXiv 16 Sep 2025.

Builds on

14 of 21 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025Membership Inference Attacks on Large-Scale Models: A Survey25 Mar
  2. 2024LLM Dataset Inference: Did you train on my dataset?10 Jun
  3. 2023Membership Inference Attacks against Language Models via Neighbourhood Comparison29 May
  4. 2021Membership Inference Attack Susceptibility of Clinical Language Models16 Apr

Other references · 10

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  4. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  5. 2020Extracting Training Data from Large Language Modelscited by 2,189
  6. 2014FitNets: Hints for Thin Deep Netscited by 1,653
  7. 2019Online Knowledge Distillation with Diverse Peerscited by 128
  8. 2023CoLLM: Integrating Collaborative Embeddings into Large Language Models for Recommendationcited by 100
  9. 2021Dynamic Knowledge Distillation for Pre-trained Language Modelscited by 38
  10. 2024Text-like Encoding of Collaborative Information in Large Language Models for Recommendationcited by 25

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Membership Inference Attacks on In-Context Examples in LLM-based Recommender Systems
  2. 2024Shadow-Free Membership Inference Attacks: Recommender Systems Are More Vulnerable Than You Thought
  3. 2022Debiasing Learning for Membership Inference Attacks Against Recommender Systems
  4. 2025Reference Recommendation based Membership Inference Attack against Hybrid-based Recommender Systems ↗
  5. 2025Membership Inference Attacks on Recommender System: A Survey
  6. 2022Membership Inference Attacks Against Latent Factor Model
  7. 2021Membership Inference Attacks Against Recommender Systems
  8. 2024EM-MIAs: Enhancing Membership Inference Attacks in Large Language Models through Ensemble Modeling