Trification: A Comprehensive Tree-based Strategy Planner and Structural Verification for Fact-Checking

TL;DR

Trification introduces a tree-based strategy planner and a dependency graph to overhaul automated fact-checking. By generating a complete set of verification actions, structuring them as a DAG, and enabling dynamic graph modification, it enables parallel, conditional, and adaptive verification. Empirical results on FEVEROUS and HOVER show state-of-the-art macro-F1 and robust performance across multi-hop claims, with clear benefits from the REFINE and THINK components and dynamic planning. The framework advances robustness by mitigating premature termination and improving reasoning through explicit dependencies and recoverable graph updates.

Abstract

Technological advancement allows information to be shared in just a single click, which has enabled the rapid spread of false information. This makes automated fact-checking system necessary to ensure the safety and integrity of our online media ecosystem. Previous methods have demonstrated the effectiveness of decomposing the claim into simpler sub-tasks and utilizing LLM-based multi agent system to execute them. However, those models faces two limitations: they often fail to verify every component in the claim and lack of structured framework to logically connect the results of sub-tasks for a final prediction. In this work, we propose a novel automated fact-checking framework called Trification. Our framework begins by generating a comprehensive set of verification actions to ensure complete coverage of the claim. It then structured these actions into a dependency graph to model the logical interaction between actions. Furthermore, the graph can be dynamically modified, allowing the system to adapt its verification strategy. Experimental results on two challenging benchmarks demonstrate that our framework significantly enhances fact-checking accuracy, thereby advancing current state-of-the-art in automated fact-checking system.

Figures (2)

• Figure 1: Case Comparison: Chain Search vs. Tree Verification
• Figure 2: Dynamic Tree Planner: insufficient evidence triggers dynamic graph modification and integration of a newly generated subtree