Wang Z, Ma Y, Xu C. Eliciting Chain-of-Thought in Base LLMs via Gradient-Based Representation Optimization. AAAI [Internet]. 2026 Mar. 14 [cited 2026 May 15];40(40):33782-90. Available from: https://ojs.aaai.org/index.php/AAAI/article/view/40669