Enhancing Retrieval Augmentation via Adversarial Collaboration

2025-09-18

AI Summary

To address retrieval hallucination in Retrieval-Augmented Generation (RAG), where domain-specific large language models are misled by low-quality retrieval results, this paper proposes an adversarial collaboration-based dual-agent framework. The framework comprises a general-purpose detector and a domain-specific parser, coordinated by a mediation module that orchestrates iterative questioning–responding interactions to dynamically identify knowledge gaps, verify retrieval relevance, and refine retrieval queries. Its key innovation lies in introducing, for the first time, an adversarial collaboration paradigm between heterogeneous agents—integrating knowledge-gap detection, domain-expert reasoning, and closed-loop retrieval feedback. Experiments across multiple vertical domains demonstrate significant improvements: +12.7% in retrieval accuracy and +19.3% in factual consistency of generated outputs, outperforming state-of-the-art RAG baselines.

Technology Category

• Multiagent Systems: Adversarial Agents
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Retrieval-augmented Generation (RAG) is a prevalent approach for domain-specific LLMs, yet it is often plagued by "Retrieval Hallucinations"--a phenomenon where fine-tuned models fail to recognize and act upon poor-quality retrieved documents, thus undermining performance. To address this, we propose the Adversarial Collaboration RAG (AC-RAG) framework. AC-RAG employs two heterogeneous agents: a generalist Detector that identifies knowledge gaps, and a domain-specialized Resolver that provides precise solutions. Guided by a moderator, these agents engage in an adversarial collaboration, where the Detector's persistent questioning challenges the Resolver's expertise. This dynamic process allows for iterative problem dissection and refined knowledge retrieval. Extensive experiments show that AC-RAG significantly improves retrieval accuracy and outperforms state-of-the-art RAG methods across various vertical domains.

Problem

• Addresses retrieval hallucinations in RAG systems
• Improves recognition of poor-quality retrieved documents
• Enhances domain-specific knowledge retrieval accuracy

Innovation

• Adversarial collaboration framework with two agents
• Detector identifies gaps, Resolver provides solutions
• Iterative process improves retrieval accuracy significantly