AI RESEARCH
Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality
arXiv CS.LG
•
ArXi:2509.23765v3 Announce Type: replace-cross Hallucination in large language models (LLMs) during long-form generation remains difficult to address under existing reinforcement learning from human feedback (RLHF) frameworks, as their preference rewards often overlook the model's own knowledge boundaries. In this paper, we propose the $\textbf{K}$nowledge-$\textbf{L}$evel $\textbf{C}$onsistency Reinforcement Learning $\textbf{F}$ramework ($\textbf{KLCF}$), which re-examines this problem from a distribution alignment perspective.