TailNLG: A Multilingual Benchmark Addressing Verbalization of Long-Tail Entities
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917366495969280 |
|---|---|
| author | Draetta, Lia Oliverio, Michael Ramón-Ferrer, Virginia Balestrucci, Pier Felice Corallo, Flaviana Badenes-Olmedo, Carlos Mazzei, Alessandro Stranisci, Marco Antonio Damiano, Rossana |
| author_facet | Draetta, Lia Oliverio, Michael Ramón-Ferrer, Virginia Balestrucci, Pier Felice Corallo, Flaviana Badenes-Olmedo, Carlos Mazzei, Alessandro Stranisci, Marco Antonio Damiano, Rossana |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_27768 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | TailNLG: A Multilingual Benchmark Addressing Verbalization of Long-Tail Entities Draetta, Lia Oliverio, Michael Ramón-Ferrer, Virginia Balestrucci, Pier Felice Corallo, Flaviana Badenes-Olmedo, Carlos Mazzei, Alessandro Stranisci, Marco Antonio Damiano, Rossana Computation and Language 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. |
| title | TailNLG: A Multilingual Benchmark Addressing Verbalization of Long-Tail Entities |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2603.27768 |