Tag
Proposes TEMPO, a policy optimization method that trains LLMs to reason exclusively from pre-cutoff information by using a two-mode reward and GRPO-based training, reducing knowledge leakage by 2–13% while improving task performance by 6–13%.
This paper introduces SeedRG, a semi-synthetic benchmark generation pipeline designed to eliminate knowledge leakage in Retrieval-Augmented Generation (RAG) evaluation by creating novel examples that preserve reasoning structures but are absent from model parametric memory.