Compositional generalization remains a foundational weakness of modern neural networks, limiting their robustness and applicability in domains requiring out-of-distribution reasoning. A central, yet unverified, assumption in neuro-symbolic AI is that compositional reasoning will emerge as a byproduct of successful symbol grounding. This work presents the first systematic empirical analysis to challenge this assumption by disentangling the contributions of grounding and reasoning. To operationalize this investigation, we introduce the Iterative Logic Tensor Network ($i$LTN), a fully differentiable architecture designed for multi-step deduction. Using a formal taxonomy of generalization -- probing for novel entities, unseen relations, and complex rule compositions -- we demonstrate that a model trained solely on a grounding objective fails to generalize. In contrast, our full $i$LTN, trained jointly on perceptual grounding and multi-step reasoning, achieves high zero-shot accuracy across all tasks. Our findings provide conclusive evidence that symbol grounding, while necessary, is insufficient for generalization, establishing that reasoning is not an emergent property but a distinct capability that requires an explicit learning objective.

翻译：组合泛化依然是现代神经网络的根本性弱点，限制了其在需分布外推理领域的鲁棒性与适用性。神经符号人工智能中一个核心但未经实证的假设是：组合推理将作为成功符号接地性的副产品自然涌现。本研究通过分离接地性与推理各自的作用，首次对这一假设开展了系统性实证分析。为具体化这一研究，我们提出了迭代逻辑张量网络（$i$LTN）——一种专为多步推理设计的全微分架构。基于形式化的泛化分类法（分别探测新实体、未见关系以及复杂规则组合），我们证明仅以接地性为目标训练的模型无法实现泛化。相反，我们的完整$i$LTN在感知接地性与多步推理上进行联合训练，在所有任务中均取得了高零样本准确率。研究结果提供确凿证据：符号接地性虽属必要，但不足以实现泛化，从而确立了推理并非一种涌现属性，而是一种需要显式学习目标的独立能力。