When No Paths Lead to Rome: Benchmarking Systematic Neural Relational Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Das, Anirban, Khalid, Irtaza, Peñaloza, Rafael, Schockaert, Steven
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915580074786816
author Das, Anirban
Khalid, Irtaza
Peñaloza, Rafael
Schockaert, Steven
author_facet Das, Anirban
Khalid, Irtaza
Peñaloza, Rafael
Schockaert, Steven
contents Designing models that can learn to reason in a systematic way is an important and long-standing challenge. In recent years, a wide range of solutions have been proposed for the specific case of systematic relational reasoning, including Neuro-Symbolic approaches, variants of the Transformer architecture, and specialised Graph Neural Networks. However, existing benchmarks for systematic relational reasoning focus on an overly simplified setting, based on the assumption that reasoning can be reduced to composing relational paths. In fact, this assumption is hard-baked into the architecture of several recent models, leading to approaches that can perform well on existing benchmarks but are difficult to generalise to other settings. To support further progress in the field of systematic relational reasoning with neural networks, we introduce NoRA, a new benchmark which adds several levels of difficulty and requires models to go beyond path-based reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2510_23532
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle When No Paths Lead to Rome: Benchmarking Systematic Neural Relational Reasoning
Das, Anirban
Khalid, Irtaza
Peñaloza, Rafael
Schockaert, Steven
Artificial Intelligence
Machine Learning
Designing models that can learn to reason in a systematic way is an important and long-standing challenge. In recent years, a wide range of solutions have been proposed for the specific case of systematic relational reasoning, including Neuro-Symbolic approaches, variants of the Transformer architecture, and specialised Graph Neural Networks. However, existing benchmarks for systematic relational reasoning focus on an overly simplified setting, based on the assumption that reasoning can be reduced to composing relational paths. In fact, this assumption is hard-baked into the architecture of several recent models, leading to approaches that can perform well on existing benchmarks but are difficult to generalise to other settings. To support further progress in the field of systematic relational reasoning with neural networks, we introduce NoRA, a new benchmark which adds several levels of difficulty and requires models to go beyond path-based reasoning.
title When No Paths Lead to Rome: Benchmarking Systematic Neural Relational Reasoning
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2510.23532