Type-Less yet Type-Aware Inductive Link Prediction with Pretrained Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: De Bellis, Alessandro, Bufi, Salvatore, Servedio, Giovanni, Anelli, Vito Walter, Di Noia, Tommaso, Di Sciascio, Eugenio
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908568812257280
author De Bellis, Alessandro
Bufi, Salvatore
Servedio, Giovanni
Anelli, Vito Walter
Di Noia, Tommaso
Di Sciascio, Eugenio
author_facet De Bellis, Alessandro
Bufi, Salvatore
Servedio, Giovanni
Anelli, Vito Walter
Di Noia, Tommaso
Di Sciascio, Eugenio
contents Inductive link prediction is emerging as a key paradigm for real-world knowledge graphs (KGs), where new entities frequently appear and models must generalize to them without retraining. Predicting links in a KG faces the challenge of guessing previously unseen entities by leveraging generalizable node features such as subgraph structure, type annotations, and ontological constraints. However, explicit type information is often lacking or incomplete. Even when available, type information in most KGs is often coarse-grained, sparse, and prone to errors due to human annotation. In this work, we explore the potential of pre-trained language models (PLMs) to enrich node representations with implicit type signals. We introduce TyleR, a Type-less yet type-awaRe approach for subgraph-based inductive link prediction that leverages PLMs for semantic enrichment. Experiments on standard benchmarks demonstrate that TyleR outperforms state-of-the-art baselines in scenarios with scarce type annotations and sparse graph connectivity. To ensure reproducibility, we share our code at https://github.com/sisinflab/tyler .
format Preprint
id arxiv_https___arxiv_org_abs_2509_26224
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Type-Less yet Type-Aware Inductive Link Prediction with Pretrained Language Models
De Bellis, Alessandro
Bufi, Salvatore
Servedio, Giovanni
Anelli, Vito Walter
Di Noia, Tommaso
Di Sciascio, Eugenio
Computation and Language
Artificial Intelligence
Inductive link prediction is emerging as a key paradigm for real-world knowledge graphs (KGs), where new entities frequently appear and models must generalize to them without retraining. Predicting links in a KG faces the challenge of guessing previously unseen entities by leveraging generalizable node features such as subgraph structure, type annotations, and ontological constraints. However, explicit type information is often lacking or incomplete. Even when available, type information in most KGs is often coarse-grained, sparse, and prone to errors due to human annotation. In this work, we explore the potential of pre-trained language models (PLMs) to enrich node representations with implicit type signals. We introduce TyleR, a Type-less yet type-awaRe approach for subgraph-based inductive link prediction that leverages PLMs for semantic enrichment. Experiments on standard benchmarks demonstrate that TyleR outperforms state-of-the-art baselines in scenarios with scarce type annotations and sparse graph connectivity. To ensure reproducibility, we share our code at https://github.com/sisinflab/tyler .
title Type-Less yet Type-Aware Inductive Link Prediction with Pretrained Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2509.26224