GLiNER-Relex: A Unified Framework for Joint Named Entity Recognition and Relation Extraction
Ihor Stepanov, Oleksandr Lukashov, Mykhailo Shtopko, Vivek Kalyanarangan · 2026
Read GLiNER-Relex: A Unified Framework for Joint Named Entity Recognition and Relation Extraction by Ihor Stepanov, Oleksandr Lukashov, Mykhailo Shtopko, Vivek Kalyanarangan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-10108.txt.
Description
Stepanov and three colleagues unify two language tasks that are usually run apart: finding named entities in text and extracting relations between them, the two steps behind building knowledge graphs. Their system, GLiNER-Relex, extends the GLiNER framework so one model does both jobs at once, instead of needing separate models for each.
A shared bidirectional transformer encoder represents text together with entity-type and relation-type labels, which lets the model extract arbitrary types specified at run time, even types never seen in training. Entity pairs are scored against relation embeddings by a dedicated module. Tests on CoNLL04, DocRED, FewRel and CrossRE show competitive results against specialised extractors and large language models, with GLiNER's trademark efficiency intact and code released openly.
Catalog record
| CREATOR | Ihor Stepanov, Oleksandr Lukashov, Mykhailo Shtopko, Vivek Kalyanarangan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.10108