Get Started
Home
Topics
Search
Library
Research questionHow can natural-language interfaces generate reliable SQL over nested, graph-like enterprise schemas at acceptable cost?Natural-language questions must be mapped onto relevant fields and relationships in schemas that may combine deeply nested data with links between records. These structures make query planning and correctness difficult while increasing deployment costs.
AI
AI Agents
Code Generation & Program Synthesis
Evaluation & Benchmarks
Natural Language Processing
Latest papersRecent research connected to this question, newest first.A Cost-Aware Agentic Architecture for NL-to-SQL over Nested Enterprise Schemas, with a New BenchmarkThe paper contributes an execution-verified enterprise benchmark and evaluates a cost-aware, single-generation agentic system on it and on a Snowflake setting from Spider 2.0. Its evidence is limited to the reported enterprise schemas, datasets, and operating point, so broader production behavior is not established.research paper · Sep 4, 2026
Related questions
How can natural-language interfaces generate reliable SQL and GraphQL without runtime LLM inference?How can Text-to-SQL systems learn from database execution when annotated question-SQL pairs are unavailable?How can governed language-model analytics preserve expressive queries while producing replayable, evidence-backed answers?How can relational databases support isolated, reproducible branching and merging for AI agents through SQL?