# An End-to-End Ukrainian RAG for Local Deployment. Optimized Hybrid Search and Lightweight Generation

- **Authors:** Mykola Trokhymovych, Yana Oliinyk, Nazarii Nyzhnyk
- **Venue:** Proceedings of the Fifth Ukrainian Natural Language Processing Conference (UNLP 2026) (UNLP’26), 2026
- **Publisher:** Association for Computational Linguistics
- **Keywords:** retrieval-augmented generation, RAG, Ukrainian NLP, hybrid search, information retrieval, small language models, local deployment
- **Page:** https://www.trokhymovych.com/publications/an-end-to-end-ukrainian-rag-for-local-deployment-optimized-2026/

## Abstract

This paper presents a highly efficient Retrieval-Augmented Generation (RAG) system built specifically for Ukrainian document question answering, which achieved 2nd place in the UNLP 2026 Shared Task. Our solution features a custom two-stage search pipeline that retrieves relevant document pages, paired with a specialized Ukrainian language model fine-tuned on synthetic data to generate accurate, grounded answers. Finally, we compress the model for lightweight deployment. Evaluated under strict computational limits, our architecture demonstrates that high-quality, verifiable AI question answering can be achieved locally on resource-constrained hardware without sacrificing accuracy.

## Links

- arXiv: https://arxiv.org/abs/2604.22095
- HTML: https://aclanthology.org/2026.unlp-1.19/
- Data: https://github.com/unlp-workshop/unlp-2026-shared-task
- PDF: https://aclanthology.org/2026.unlp-1.19.pdf
- Code: https://github.com/trokhymovych/unlp-2026-shared-task
- Slides: https://www.trokhymovych.com/assets/pdf/papers_slides/unlp_26.pdf

## Citation

```bibtex
@inproceedings{trokhymovych-etal-2026-end,
  title = {An End-to-End {U}krainian {RAG} for Local Deployment. Optimized Hybrid Search and Lightweight Generation},
  author = {Trokhymovych, Mykola and Oliinyk, Yana and Nyzhnyk, Nazarii},
  editor = {Romanyshyn, Mariana},
  booktitle = {Proceedings of the Fifth {U}krainian Natural Language Processing Conference ({UNLP} 2026)},
  month = may,
  year = {2026},
  address = {Lviv, Ukraine},
  publisher = {Association for Computational Linguistics},
  url = {https://aclanthology.org/2026.unlp-1.19/},
  pages = {223--229},
  isbn = {979-8-89176-359-3},
}
```