“TrustRL-Eval: An Interpretable Reinforcement Learning Framework for Predicting and Optimizing Large Language Model Response Quality in Multi-Step Reasoning Tasks”, JDIAS, vol. 1, no. 1, Jun. 2026, Accessed: Jul. 18, 2026. [Online]. Available: https://www.jdataai.org/index.php/home/article/view/44