ATM: Adversarial Tuning Multi-agent System Makes a Robust Retrieval-Augmented Generator

2024/05/28 by Junda Zhu, Lingyong Yan, Zhu, Junda +7 · 7 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning and Algorithms #Neural Networks and Applications

paper · pdf · doi:10.48550/arxiv.2405.18111

openalex publication_date 2024/05/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Large language models (LLMs) are proven to benefit a lot from retrieval-augmented generation (RAG) in alleviating hallucinations confronted with knowledge-intensive questions. RAG adopts information retrieval techniques to inject external knowledge from semantic-relevant documents as input contexts. However, since today's Internet is flooded with numerous noisy and fabricating content, it is inevitable that RAG systems are vulnerable to these noises and prone to respond incorrectly. To this end, we propose to optimize the retrieval-augmented Generator with an Adversarial Tuning Multi-agent system (ATM). The ATM steers the Generator to have a robust perspective of useful documents for question answering with the help of an auxiliary Attacker agent through adversarially tuning the agents for several iterations. After rounds of multi-agent iterative tuning, the Generator can eventually better discriminate useful documents amongst fabrications. The experimental results verify the effectiveness of ATM and we also observe that the Generator can achieve better performance compared to the state-of-the-art baselines.

Cited by

• Can Synthetic Query Rewrites Capture User Intent Better than Humans in Retrieval-Augmented Generation?
• Bridging the Capability Gap: Joint Alignment Tuning for Harmonizing LLM-based Multi-Agent Systems
• AgentArmor: Enforcing Program Analysis on Agent Runtime Trace to Defend Against Prompt Injection
• RAGentA: Multi-Agent Retrieval-Augmented Generation for Attributed Question Answering
• Iterative Self-Incentivization Empowers Large Language Models as Agentic Searchers
• Direct Retrieval-augmented Optimization: Synergizing Knowledge Selection and Language Models
• Scalable Defense against In-the-wild Jailbreaking Attacks with Safety Context Retrieval

Related

• Enhancing Retrieval Augmentation via Adversarial Collaboration
• MAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation
• Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
• Supervising the search process produces reliable and generalizable information-seeking agents
• Towards More Robust Retrieval-Augmented Generation: Evaluating RAG Under Adversarial Poisoning Attacks