Improving the fact-checking performance of language models by relying on their entailment ability
This work addresses the challenge of building robust fact-checking systems for real-world deployment, though it appears incremental as it builds on existing strategies like fine-tuning and prompting.
The authors tackled the problem of low accuracy in automated fact-checking by using entailed justifications from LLMs to train encoder-only language models, resulting in improved performance compared to existing methods.
Automated fact-checking has been a challenging task for the research community. Past works tried various strategies, such as end-to-end training, retrieval-augmented generation, and prompt engineering, to build robust fact-checking systems. However, their accuracy has not been very high for real-world deployment. We, on the other hand, propose a simple yet effective strategy, where entailed justifications generated by LLMs are used to train encoder-only language models (ELMs) for fact-checking. We conducted a rigorous set of experiments, comparing our approach with recent works and various prompting and fine-tuning strategies to demonstrate the superiority of our approach. Additionally, we did quality analysis of model explanations, ablation studies, and error analysis to provide a comprehensive understanding of our approach.