mlsanityData to 14 Oct 2026

Tatarstan Toponyms: A Bilingual Dataset and Hybrid RAG System for Geospatial Question Answering

arXiv 2605.05962 ↗First posted 7 May 2026cs.CLcited by 0

On the map: Retrieval-augmented generation → Generation and context integration

Abstract (arXiv)

This paper addresses end-to-end geospatial question answering over multilingual toponymic data. We introduce a bilingual (Russian-Tatar) dataset of 9,688 toponyms with linguistic, etymological, and coordinate information (93.1 percent georeferenced). Based on this, we construct about 39,000 question-context-answer triples with guaranteed answer localization. Our architecture combines a hybrid retriever (dense semantic indexing with multilingual-e5-large plus geospatial filtering/ranking using KD-trees and haversine distance) and an extractive reader fine-tuned on transformer models. On 500 test queries, hybrid search achieves Recall@1 = 0.988, Recall@5 = 1.000, MRR = 0.994, significantly outperforming BM25 and spatial-only methods. Among readers (RuBERT, XLM-RoBERTa-large, T5-RUS), XLM-RoBERTa-large gives best results: EM = 0.992, F1 = 0.994. RuBERT models fail on coordinate questions due to tokenization artifacts, but simple post-processing recovers 100 percent accuracy. Resources (dataset, QA corpus, models, web demo) are openly released on Hugging Face. Results are directly applicable to geospatial QA services, geocoding, and digital humanities projects.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Generation and context integration, one of 15 directions in Retrieval-augmented generation. Builds on 8 of its 35 references that are in the map. First posted on arXiv 7 May 2026.

Builds on

8 of 35 references.

References · 8

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  3. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  4. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
  5. 2017Billion-scale similarity search with GPUscited by 2,497
  6. 2022RuNNE-2022 Shared Task: Recognizing Nested Named Entitiescited by 8
  7. 2026Adapting Large Language Models to a Low-Resource Agglutinative Language: A Comparative Study of LoRA and QLoRA for Bashkircited by 3
  8. 2022Russian SuperGLUE 1.1: Revising the Lessons not Learned by Russian NLP modelscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026GS-QA: A Benchmark for Geospatial Question Answering
  2. 2025Spatial-RAG: Spatial Retrieval Augmented Generation for Real-World Geospatial Reasoning Questions
  3. 2024Geode: A Zero-shot Geospatial Question-Answering Agent with Explicit Reasoning and Precise Spatio-Temporal Retrieval
  4. 2025GeoRAG: A Question-Answering Approach from a Geographical Perspective
  5. 2026GeoRisk-RAG: A Hierarchy-Aware Risk Framework for Improving RAG Reliability through Selective Answering
  6. 2025MapQA: Open-domain Geospatial Question Answering on Map Data
  7. 2024Location Aware Modular Biencoder for Tourism Question Answering
  8. 2026DaPT: A Dual-Path Framework for Multilingual Multi-hop Question Answering