TailNLG: A Multilingual Benchmark Addressing Verbalization of Long-Tail Entities

TL;DR

This paper introduces TailNLG, a multilingual benchmark for evaluating the verbalization of long-tail entities in Data-to-Text generation, revealing biases against rare entities across models and languages.

Contribution

It presents the first systematic study of long-tail entities in multilingual Data-to-Text generation and introduces a new benchmark built from Wikidata.

Findings

Models show bias against long-tail entities with lower embedding scores.

Model uncertainty is higher for rare entities across languages.

Existing metrics do not reliably capture performance differences on long-tail entities.

Abstract

The automatic verbalization of structured knowledge is a key task for making knowledge graphs accessible to non-expert users and supporting retrieval-augmented generation systems. Although recent advances in Data-to-Text generation have improved multilingual coverage, little attention has been paid to potential biases in the verbalization of rare entities, frequently known as long-tail entities. In this work, we present the first systematic study of long-tail entities in Data-to-Text generation. We introduce TailNLG, a new multilingual benchmark in English, Italian, and Spanish, built from Wikidata and covering entities with varying levels of popularity. We evaluate three different families of large language models in zero-shot settings and compare their performance on rare versus common entities, as well as against the established WebNLG benchmark. Our results reveal a consistent bias against long-tail entities: embedding-based scores are lower, and model uncertainty is higher for rare entities. We further show that the impact of long-tail entities varies across models and languages, and that existing evaluation metrics do not consistently capture these differences, highlighting the need for more reliable evaluation frameworks.