Stealthy LLM-Driven Data Poisoning Attacks Against Embedding-Based Retrieval-Augmented Recommender Systems

• URL: http://arxiv.org/abs/2505.05196v1
• Date: Thu, 08 May 2025 12:53:42 GMT
• Title: Stealthy LLM-Driven Data Poisoning Attacks Against Embedding-Based Retrieval-Augmented Recommender Systems
• Authors: Fatemeh Nazary, Yashar Deldjoo, Tommaso Di Noia, Eugenio Di Sciascio,
• Abstract summary: We study provider-side data poisoning in retrieval-augmented recommender systems (RAG)<n>By modifying only a small fraction of tokens within item descriptions, an attacker can significantly promote or demote targeted items.<n>Experiments on MovieLens, using two large language model (LLM) retrieval modules, show that even subtle attacks shift final rankings and item exposures while eluding naive detection.
• Score: 16.79952669254101
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present a systematic study of provider-side data poisoning in retrieval-augmented recommender systems (RAG-based). By modifying only a small fraction of tokens within item descriptions -- for instance, adding emotional keywords or borrowing phrases from semantically related items -- an attacker can significantly promote or demote targeted items. We formalize these attacks under token-edit and semantic-similarity constraints, and we examine their effectiveness in both promotion (long-tail items) and demotion (short-head items) scenarios. Our experiments on MovieLens, using two large language model (LLM) retrieval modules, show that even subtle attacks shift final rankings and item exposures while eluding naive detection. The results underscore the vulnerability of RAG-based pipelines to small-scale metadata rewrites and emphasize the need for robust textual consistency checks and provenance tracking to thwart stealthy provider-side poisoning.

Related papers

• TopicAttack: An Indirect Prompt Injection Attack via Topic Transition [71.81906608221038]
  Large language models (LLMs) are vulnerable to indirect prompt injection attacks.<n>We propose TopicAttack, which prompts the LLM to generate a fabricated transition prompt that gradually shifts the topic toward the injected instruction.<n>We find that a higher injected-to-original attention ratio leads to a greater success probability, and our method achieves a much higher ratio than the baseline methods.
  arXiv  Detail & Related papers  (2025-07-18T06:23:31Z)
• DATABench: Evaluating Dataset Auditing in Deep Learning from an Adversarial Perspective [59.66984417026933]
  We introduce a novel taxonomy, classifying existing methods based on their reliance on internal features (IF) (inherent to the data) versus external features (EF) (artificially introduced for auditing)<n>We formulate two primary attack types: evasion attacks, designed to conceal the use of a dataset, and forgery attacks, intending to falsely implicate an unused dataset.<n>Building on the understanding of existing methods and attack objectives, we further propose systematic attack strategies: decoupling, removal, and detection for evasion; adversarial example-based methods for forgery.<n>Our benchmark, DATABench, comprises 17 evasion attacks, 5 forgery attacks, and 9
  arXiv  Detail & Related papers  (2025-07-08T03:07:15Z)
• Benchmarking Misuse Mitigation Against Covert Adversaries [80.74502950627736]
  Existing language model safety evaluations focus on overt attacks and low-stakes tasks.<n>We develop Benchmarks for Stateful Defenses (BSD), a data generation pipeline that automates evaluations of covert attacks and corresponding defenses.<n>Our evaluations indicate that decomposition attacks are effective misuse enablers, and highlight stateful defenses as a countermeasure.
  arXiv  Detail & Related papers  (2025-06-06T17:33:33Z)
• The Silent Saboteur: Imperceptible Adversarial Attacks against Black-Box Retrieval-Augmented Generation Systems [101.68501850486179]
  We explore adversarial attacks against retrieval-augmented generation (RAG) systems to identify their vulnerabilities.<n>This task aims to find imperceptible perturbations that retrieve a target document, originally excluded from the initial top-$k$ candidate set.<n>We propose ReGENT, a reinforcement learning-based framework that tracks interactions between the attacker and the target RAG.
  arXiv  Detail & Related papers  (2025-05-24T08:19:25Z)
• CAMOUFLAGE: Exploiting Misinformation Detection Systems Through LLM-driven Adversarial Claim Transformation [4.02943411607022]
  Existing black-box text-based adversarial attacks are ill-suited for evidence-based misinformation detection systems.<n>We present CAMOUFLAGE, an iterative, LLM-driven approach that employs a two-agent system to create adversarial claim rewritings.<n>We evaluate CAMOUFLAGE on four systems, including two recent academic systems and two real-world APIs, with an average attack success rate of 46.92%.
  arXiv  Detail & Related papers  (2025-05-03T19:14:24Z)
• Illusions of Relevance: Using Content Injection Attacks to Deceive Retrievers, Rerankers, and LLM Judges [52.96987928118327]
  We find that embedding models for retrieval, rerankers, and large language model (LLM) relevance judges are vulnerable to content injection attacks.<n>We identify two primary threats: (1) inserting unrelated or harmful content within passages that still appear deceptively "relevant", and (2) inserting entire queries or key query terms into passages to boost their perceived relevance.<n>Our study systematically examines the factors that influence an attack's success, such as the placement of injected content and the balance between relevant and non-relevant material.
  arXiv  Detail & Related papers  (2025-01-30T18:02:15Z)
• Towards More Robust Retrieval-Augmented Generation: Evaluating RAG Under Adversarial Poisoning Attacks [45.07581174558107]
  Retrieval-Augmented Generation (RAG) systems have emerged as a promising solution to mitigate hallucinations.<n>RAG systems are vulnerable to adversarial poisoning attacks, where malicious passages injected into retrieval databases can mislead the model into generating factually incorrect outputs.<n>This paper investigates both the retrieval and the generation components of RAG systems to understand how to enhance their robustness against such attacks.
  arXiv  Detail & Related papers  (2024-12-21T17:31:52Z)
• Toward Improving Synthetic Audio Spoofing Detection Robustness via Meta-Learning and Disentangled Training With Adversarial Examples [33.445126880876415]
  We propose a reliable and robust spoofing detection system to filter out spoofing attacks instead of having them reach the automatic speaker verification system. A weighted additive angular margin loss is proposed to address the data imbalance issue, and different margins has been assigned to improve generalization to unseen spoofing attacks. We craft adversarial examples by adding imperceptible perturbations to spoofing speech as a data augmentation strategy, then we use an auxiliary batch normalization to guarantee that corresponding normalization statistics are performed exclusively on the adversarial examples.
  arXiv  Detail & Related papers  (2024-08-23T19:26:54Z)
• Corpus Poisoning via Approximate Greedy Gradient Descent [48.5847914481222]
  We propose Approximate Greedy Gradient Descent, a new attack on dense retrieval systems based on the widely used HotFlip method for generating adversarial passages. We show that our method achieves a high attack success rate on several datasets and using several retrievers, and can generalize to unseen queries and new domains.
  arXiv  Detail & Related papers  (2024-06-07T17:02:35Z)
• Defense of Adversarial Ranking Attack in Text Retrieval: Benchmark and Baseline via Detection [12.244543468021938]
  This paper introduces two types of detection tasks for adversarial documents. A benchmark dataset is established to facilitate the investigation of adversarial ranking defense. A comprehensive investigation of the performance of several detection baselines is conducted.
  arXiv  Detail & Related papers  (2023-07-31T16:31:24Z)
• Exploring Model Dynamics for Accumulative Poisoning Discovery [62.08553134316483]
  We propose a novel information measure, namely, Memorization Discrepancy, to explore the defense via the model-level information. By implicitly transferring the changes in the data manipulation to that in the model outputs, Memorization Discrepancy can discover the imperceptible poison samples. We thoroughly explore its properties and propose Discrepancy-aware Sample Correction (DSC) to defend against accumulative poisoning attacks.
  arXiv  Detail & Related papers  (2023-06-06T14:45:24Z)
• ADC: Adversarial attacks against object Detection that evade Context consistency checks [55.8459119462263]
  We show that even context consistency checks can be brittle to properly crafted adversarial examples. We propose an adaptive framework to generate examples that subvert such defenses. Our results suggest that how to robustly model context and check its consistency, is still an open problem.
  arXiv  Detail & Related papers  (2021-10-24T00:25:09Z)
• Exploring Robustness of Unsupervised Domain Adaptation in Semantic Segmentation [74.05906222376608]
  We propose adversarial self-supervision UDA (or ASSUDA) that maximizes the agreement between clean images and their adversarial examples by a contrastive loss in the output space. This paper is rooted in two observations: (i) the robustness of UDA methods in semantic segmentation remains unexplored, which pose a security concern in this field; and (ii) although commonly used self-supervision (e.g., rotation and jigsaw) benefits image tasks such as classification and recognition, they fail to provide the critical supervision signals that could learn discriminative representation for segmentation tasks.
  arXiv  Detail & Related papers  (2021-05-23T01:50:44Z)
• Temporal Action Detection with Multi-level Supervision [116.55596693897388]
  We introduce the Semi-supervised Action Detection (SSAD) task with a mixture of labeled and unlabeled data. We analyze different types of errors in the proposed SSAD baselines which are directly adapted from the semi-supervised classification task. We incorporate weakly-labeled data into SSAD and propose Omni-supervised Action Detection (OSAD) with three levels of supervision.
  arXiv  Detail & Related papers  (2020-11-24T04:45:17Z)

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.