Learning from Feedback: Semantic Enhancement for Object SLAM Using Foundation Models