Follow the Path: Reasoning over Knowledge Graph Paths to Improve Large Language Model Factuality

arXiv:2505.11140v3 Announce Type: replace-cross Abstract: We introduce fs1, a simple yet effective method that improves the factuality of reasoning traces by collecting them from large reasoning models and grounding them in knowledge graph (KG) paths. We fine-tune eight instruction-tuned Large Language Models (LLMs) on 3.9K factually grounded reasoning traces and rigorously evaluate them on six complex open-domain question-answering (QA) benchmarks encompassing 23.9K questions. Our results demon