PURPOSE: Poisoning Conflict Resolution in RAG via Proxy-Fact-Grounded Updates
Zijian Wang, Yubo Zhu, Muzhi Dong, Yanjun Lou, Yisheng Li, ZiLiang Zhang, Wei Tong, Yuan Zhang, Jingyu Hua, Sheng Zhong
cs.CR, cs.AI
Submitted: 2026-08-05
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Parameters vs. Context: Fine-Grained Control of Knowledge Reliance in Language Models
- DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
- MS MARCO: A Human Generated MAchine Reading COmprehension Dataset
- Retrieval-Augmented Generation for Large Language Models: A Survey
- The Llama 3 Herd of Models
- LegalWiz: A Multi-Agent Generation Framework for Contradiction Detection in Legal Documents
- Architecture Matters: Comparing RAG Systems under Knowledge Base Poisoning
- Retrieval-Augmented Generation with Conflicting Evidence
- Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation
- Making Retrieval-Augmented Language Models Robust to Irrelevant Context
- LLM-based Corroborating and Refuting Evidence Retrieval for Scientific Claim Verification
- Certifiably Robust RAG against Retrieval Corruption
- HijackRAG: Hijacking Attacks against Retrieval-Augmented Large Language Models
- Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
- ArgRAG: Explainable Retrieval Augmented Generation using Quantitative Bipolar Argumentation
- BadRAG: Identifying Vulnerabilities in Retrieval Augmented Generation of Large Language Models
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Estimating the Social Cost of Corporate Data Breaches
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills