SANE Schema-aware Natural-language Evaluation of Biological Data
For biologists and database users lacking SQL expertise, SANE provides a scalable and reproducible evaluation method to assess LLM reliability in domain-specific text-to-SQL tasks.
SANE introduces a schema-aware, automatically generated benchmark for evaluating text-to-SQL systems in biological data domains, and shows that few-shot LLMs with structured prompting achieve accurate query generation without fine-tuning, with failures mainly due to ambiguous inputs rather than incorrect SQL.
High-throughput microscopy generates large, structured datasets capturing cellular responses to pharmacological perturbations, but accessing these datasets typically requires SQL expertise. Large language models offer a natural-language alternative, yet their tendency to hallucinate raises concerns about result reliability . We present SANE Schema-Aware Natural-language Evaluation, a novel paradigm for domain-specific text-to-SQL evaluation: schema-grounded, automatically generated benchmarks tied to real and specific experimental structure. SANE makes evaluation more scalable, systematic, and reproducible. Using SANE, we evaluate a few-shot large language model and show that, under constrained schemas with structured prompting and guardrails, accurate query generation is achievable without any model training or fine-tuning. Most failures stem from ambiguous or underspecified inputs and manifest as overly cautious clarification requests or answers to queries that should first be disambiguated, rather than incorrect SQL generation. These results indicate that few-shot large language models can provide reliable database access in well-defined domains when combined with schema-aware prompting.