MERMAID: Memory-Enhanced Retrieval and Reasoning with Multi-Agent Iterative Knowledge Grounding for Veracity Assessment

TL;DR

MERMAID is a novel multi-agent framework that enhances veracity assessment by integrating dynamic evidence retrieval, reasoning, and persistent memory, leading to improved accuracy and efficiency in fact-checking tasks.

Contribution

This work introduces MERMAID, a memory-enhanced, multi-agent system that tightly couples retrieval and reasoning, enabling dynamic evidence reuse and reducing redundant searches in veracity assessment.

Findings

MERMAID achieves state-of-the-art performance on multiple benchmarks.

The framework improves verification efficiency through evidence memory.

Experimental results demonstrate the effectiveness of integrating retrieval, reasoning, and memory.

Abstract

Assessing the veracity of online content has become increasingly critical. Large language models (LLMs) have recently enabled substantial progress in automated veracity assessment, including automated fact-checking and claim verification systems. Typical veracity assessment pipelines break down complex claims into sub-claims, retrieve external evidence, and then apply LLM reasoning to assess veracity. However, existing methods often treat evidence retrieval as a static, isolated step and do not effectively manage or reuse retrieved evidence across claims. In this work, we propose MERMAID, a memory-enhanced multi-agent veracity assessment framework that tightly couples the retrieval and reasoning processes. MERMAID integrates agent-driven search, structured knowledge representations, and a persistent memory module within a Reason-Action style iterative process, enabling dynamic evidence acquisition and cross-claim evidence reuse. By retaining retrieved evidence in an evidence memory, the framework reduces redundant searches and improves verification efficiency and consistency. We evaluate MERMAID on three fact-checking benchmarks and two claim-verification datasets using multiple LLMs, including GPT, LLaMA, and Qwen families. Experimental results show that MERMAID achieves state-of-the-art performance while improving the search efficiency, demonstrating the effectiveness of synergizing retrieval, reasoning, and memory for reliable veracity assessment.