Guardat en:
Dades bibliogràfiques
Autors principals: Shahid, Mahnoor, Rothe, Hannes
Format: Preprint
Publicat: 2026
Matèries:
Accés en línia:https://arxiv.org/abs/2604.26521
Etiquetes: Afegir etiqueta
Sense etiquetes, Sigues el primer a etiquetar aquest registre!
_version_ 1866913072660086784
author Shahid, Mahnoor
Rothe, Hannes
author_facet Shahid, Mahnoor
Rothe, Hannes
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_26521
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Grounding vs. Compositionality: On the Non-Complementarity of Reasoning in Neuro-Symbolic Systems
Shahid, Mahnoor
Rothe, Hannes
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Logic in Computer Science
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.
title Grounding vs. Compositionality: On the Non-Complementarity of Reasoning in Neuro-Symbolic Systems
topic Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Logic in Computer Science
url https://arxiv.org/abs/2604.26521