Implicit Geometric-Semantic Fusion for Task-Oriented 6-DoF Grasp Detection via Visual Affordance and Stability Learning. ESIT- International Journal 2026, 2 (2), 24-38. https://doi.org/10.66279/apg4me53.