Add black-box training link

Author: hyeondo-jang

research.html

@@ -529,7 +529,7 @@ <h2>Research Focus</h2>
                         We study the role of loss landscape curvature in generalization and develop flatness-oriented optimization strategies with both theoretical advantages and scalability for modern deep neural networks.
                     </div>
                     <div class="fold-item">
-                        <span class="fold-item-title">Zeroth-Order Optimization for Black-Box Model Training</span>
+                        <span class="fold-item-title">Zeroth-Order Optimization for Black-Box Model Training <span class="paper-tag">(<a href="https://arxiv.org/abs/2504.06838" target="_blank">ICLR 2025</a>)</span></span>
                         We develop methods that operate without direct gradient access, motivated by settings such as proprietary model APIs and decentralized learning systems where gradients are unavailable.
                     </div>
                 </div>