KG-CRAFT: Knowledge Graph-based Contrastive Reasoning with LLMs for Enhancing Automated Fact-checking

Dimensions overall score 9.0

GitHub Code Pulse

No public code linked for this paper yet.

• KG-CRAFT improves automatic claim verification by using LLMs augmented with contrastive questions grounded in a knowledge graph.method 90%
• KG-CRAFT constructs a knowledge graph from claims and associated reports.method 95%
• KG-CRAFT formulates contextually relevant contrastive questions based on the knowledge graph structure.method 95%
• KG-CRAFT achieves state-of-the-art predictive performance on LIAR-RAW and RAWFC datasets.result 90%
• KG-CRAFT distills evidence-based reports into concise summaries for veracity assessment by LLMs.method 95%
• The system may require significant computational resources.limitation 70%
• Generating relevant contrastive questions for all contexts can be challenging.limitation 70%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card