mlsanityData to 7 Oct 2026

GRADA: Graph-based Reranking against Adversarial Documents Attack

arXiv 2505.07546 ↗First posted 12 May 2025cs.IR cs.AIcited by 10

On the map: Retrieval-augmented generation → Privacy and security of RAG

Abstract (arXiv)

Retrieval Augmented Generation (RAG) frameworks improve the accuracy of large language models (LLMs) by integrating external knowledge from retrieved documents, thereby overcoming the limitations of models' static intrinsic knowledge. However, these systems are susceptible to adversarial attacks that manipulate the retrieval process by introducing documents that are adversarial yet semantically similar to the query. Notably, while these adversarial documents resemble the query, they exhibit weak similarity to benign documents in the retrieval set. Thus, we propose a simple yet effective Graph-based Reranking against Adversarial Document Attacks (GRADA) framework aiming at preserving retrieval quality while significantly reducing the success of adversaries. Our study evaluates the effectiveness of our approach through experiments conducted on five LLMs: GPT-3.5-Turbo, GPT-4o, Llama3.1-8b, Llama3.1-70b, and Qwen2.5-7b. We use three datasets to assess performance, with results from the Natural Questions dataset demonstrating up to an 80% reduction in attack success rates while maintaining minimal loss in accuracy.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Placed in Privacy and security of RAG, one of 15 directions in Retrieval-augmented generation. Builds on 18 of its 34 references that are in the map. First posted on arXiv 12 May 2025.

Builds on

18 of 34 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025TrustRAG: Enhancing Robustness and Trustworthiness in Retrieval-Augmented Generation1 Jan

Other references · 17

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  5. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  6. 2016MS MARCO: A Human Generated MAchine Reading COmprehension Datasetcited by 1,591
  7. 2023Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injectioncited by 1,339
  8. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  9. 2022Ignore Previous Prompt: Attack Techniques For Language Modelscited by 760
  10. 2015A Neural Attention Model for Abstractive Sentence Summarizationcited by 394
  11. 2020Pretrained Transformers for Text Ranking: BERT and Beyondcited by 266
  12. 2023Poisoning Retrieval Corpora by Injecting Adversarial Passagescited by 153
5 more
  1. 2024BM25S: Orders of magnitude faster lexical search via eager sparse scoringcited by 87
  2. 2024Neural Exec: Learning (and Learning from) Execution Triggers for Prompt Injection Attackscited by 78
  3. 2022HLATR: Enhance Multi-stage Text Retrieval with Hybrid List Aware Transformer Rerankingcited by 19
  4. 2024Search Engines in an AI Era: The False Promise of Factual and Verifiable Source-Cited Responsescited by 15
  5. 2025Graph-Based Re-ranking in Information Retrieval and Beyond: A Surveycited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 10.

10 papers.

  1. 2026ProGRank: Probe-Gradient Reranking to Defend Dense-Retriever RAG from Corpus Poisoningcited by 2
  2. 2026Securing Retrieval-Augmented Generation: A Taxonomy of Attacks, Defenses, and Future Directionscited by 2
  3. 2026DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generationcited by 1
  4. 2026When Global Gating Is Enough: Admission-Time Hubness Control in Anisotropic Vector Retrieval Systemscited by 1
  5. 2026When Optimization Becomes Manipulation: Defending Generative Search against Malicious Generative Engine Optimizationcited by 1
  6. 2026Tracing Target Answers in Poisoned Retrieval Corpora via Token Influence Attributioncited by 0
  7. 2026TriShieldRAG: 3 Rings, One Blind Spot in Layered Defenses for Retrieval-Augmented Generationcited by 0
  8. 2026RAGSieve: Self-Referenced Local Contrast for Knowledge-Poison Detection in Retrieval-Augmented Generationcited by 0
  9. 2026Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generationcited by 0
  10. 2026Similarity Is Not Validity: Defending LLM Semantic Caches Against Poisoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025DeRAG: Black-box Adversarial Attacks on Multiple Retrieval-Augmented Generation Applications via Prompt Injection
  2. 2026AdversarialCoT: Single-Document Retrieval Poisoning for LLM Reasoning
  3. 2024On the Vulnerability of Applying Retrieval-Augmented Generation within Knowledge-Intensive Application Domains
  4. 2025Enhancing Retrieval Augmentation via Adversarial Collaboration
  5. 2024Towards More Robust Retrieval-Augmented Generation: Evaluating RAG Under Adversarial Poisoning Attacks
  6. 2025CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models
  7. 2025ReliabilityRAG: Effective and Provably Robust Defense for RAG-based Web-Search
  8. 2026SentAttack: A Sentence-Level Black-Box Adversarial Attack Method for Dense Retrieval Models