We present the first domain-specific text-to-SQL benchmark in Russian, targeting fields with high operational load where rapid decision-making is critical. The benchmark spans across 9 domains, including healthcare, aviation, and others, and comprises 409 curated query pairs. It is designed to test model generalization under domain shift, introducing challenges such as specialized terminology and complex schema structures. Evaluation of state-of-the-art large language models (LLM) reveals significant performance drop in comparison to open-domain academic benchmarks, highlighting the need for domain-aware approaches in text-to-SQL. The benchmark is available at: https://github.com/BrodskaiaIrina/functional-text2sql-subsets

This paper presents a system developed for the Clinical NLP 2024 Shared Task, focusing on reliable text-to-SQL modeling on Electronic Health Records (EHRs). The goal is to create a model that accurately generates SQL queries for answerable questions while avoiding incorrect responses and handling unanswerable queries. Our approach comprises three main components: a query correspondence model, a text-to-SQL model, and an SQL verifier.For the query correspondence model, we trained a logistic regression model using hand-crafted features to distinguish between answerable and unanswerable queries. As for the text-to-SQL model, we utilized T5-3B as a pretrained language model, further fine-tuned on pairs of natural language questions and corresponding SQL queries. Finally, we applied the SQL verifier to inspect the resulting SQL queries.During the evaluation stage of the shared task, our system achieved an accuracy of 68.9 % (metric version without penalty), positioning it at the fifth-place ranking. While our approach did not surpass solutions based on large language models (LMMs) like ChatGPT, it demonstrates the promising potential of domain-specific specialized models that are more resource-efficient. The code is publicly available at https://github.com/runnerup96/EHRSQL-text2sql-solution.

Semantic parsing plays a pivotal role in advancing the accessibility of human-computer interaction on a large scale. Spider, a widely recognized dataset for text2SQL, contains a wide range of natural language (NL) questions in English and corresponding SQL queries. Original splits of Spider and its adapted to Russian language and improved version, PAUQ, assume independence and identical distribution of training and testing data (i.i.d split). In this work, we propose a target length split and multilingual i.i.d split to measure compositionality and cross-language generalization. We present experimental results of popular text2SQL models on original, multilingual, and target length splits. We also construct a context-free grammar for the evaluation of compositionality in text2SQL in an out-of-distribution setting. We make the splits publicly available on HuggingFace hub via https://huggingface.co/datasets/composite/pauq

Semantic parsing is an important task that allows to democratize human-computer interaction. One of the most popular text-to-SQL datasets with complex and diverse natural language (NL) questions and SQL queries is Spider. We construct and complement a Spider dataset for Russian, thus creating the first publicly available text-to-SQL dataset for this language. While examining its components - NL questions, SQL queries and databases content - we identify limitations of the existing database structure, fill out missing values for tables and add new requests for underrepresented categories. We select thirty functional test sets with different features that can be used for the evaluation of neural models’ abilities. To conduct the experiments, we adapt baseline architectures RAT-SQL and BRIDGE and provide in-depth query component analysis. On the target language, both models demonstrate strong results with monolingual training and improved accuracy in multilingual scenario. In this paper, we also study trade-offs between machine-translated and manually-created NL queries. At present, Russian text-to-SQL is lacking in datasets as well as trained models, and we view this work as an important step towards filling this gap.