English

TurkEmbed4Retrieval: Turkish Embedding Model for Retrieval Task

Information Retrieval 2025-11-12 v1

Abstract

In this work, we introduce TurkEmbed4Retrieval, a retrieval specialized variant of the TurkEmbed model originally designed for Natural Language Inference (NLI) and Semantic Textual Similarity (STS) tasks. By fine-tuning the base model on the MS MARCO TR dataset using advanced training techniques, including Matryoshka representation learning and a tailored multiple negatives ranking loss, we achieve SOTA performance for Turkish retrieval tasks. Extensive experiments demonstrate that our model outperforms Turkish colBERT by 19,26% on key retrieval metrics for the Scifact TR dataset, thereby establishing a new benchmark for Turkish information retrieval.

Keywords

Cite

@article{arxiv.2511.07595,
  title  = {TurkEmbed4Retrieval: Turkish Embedding Model for Retrieval Task},
  author = {Özay Ezerceli and Gizem Gümüşçekiçci and Tuğba Erkoç and Berke Özenç},
  journal= {arXiv preprint arXiv:2511.07595},
  year   = {2025}
}

Comments

4 pages, in Turkish language, 1 figure, conference

R2 v1 2026-07-01T07:30:48.052Z