mlsanityData to 7 Oct 2026

A Cost-Aware Agentic Architecture for NL-to-SQL over Nested Enterprise Schemas, with a New Benchmark

arXiv 2609.04641 ↗First posted 4 Sep 2026cs.AIcited by 0

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

Natural-language-to-SQL systems have ad- vanced rapidly on academic benchmarks, yet production enterprise schemas exhibit graph- like, semi-structured, deeply nested structure that current benchmarks do not measure. We make two complementary contributions. First, we introduce the DevRev NL2SQL bench- mark: 900 execution-verified queries with nested-type and link-graph structure, accom- panied by the Semantic Depth Score (SDS), a schema-agnostic rubric for analytical reasoning depth. Second, we present a cost-aware single- generation agentic architecture whose schema- selection, metadata-retrieval, and error-repair components are designed for the requirements this regime imposes. On the DevRev NL2SQL benchmark the system attains 91.7% answer correctness, a margin of 54.6 percentage points over the next-best baseline; on the Spider 2.0 Snowflake public dataset, it is competitive with leading systems at a single-generation operating point.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. Builds on 11 of its 16 references that are in the map. First posted on arXiv 4 Sep 2026.

Builds on

11 of 16 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025ReFoRCE: A Text-to-SQL Agent with Self-Refinement, Consensus Enforcement, and Column Exploration2 Feb
  2. 2023MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQL18 Dec

Other references · 9

  1. 2018Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Taskcited by 1,120
  2. 2023Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLscited by 667
  3. 2023DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correctioncited by 465
  4. 2023Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluationcited by 378
  5. 2024CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQLcited by 132
  6. 2024A Preview of XiYan-SQL: A Multi-Generator Ensemble Framework for Text-to-SQLcited by 46
  7. 2025Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scalingcited by 20
  8. 2025SING-SQL: A Synthetic Data Generation Framework for In-Domain Text-to-SQL Translationcited by 5
  9. 2026FlexSQL: Flexible Exploration and Execution Make Better Text-to-SQL Agentscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026A Semantic-Layer-Mediated Agent for Natural Language to SQL over Heterogeneous Enterprise Databases
  2. 2026An Agentic System for Schema Aware NL2SQL Generation
  3. 2026APEX-SQL: Talking to the data via Agentic Exploration for Text-to-SQL
  4. 2025BAPPA: Benchmarking Agents, Plans, and Pipelines for Automated Text-to-SQL Generation
  5. 2026AgentNLQ: A General-Purpose Agent for Natural Language to SQL
  6. 2026AgentSM: Semantic Memory for Agentic Text-to-SQL
  7. 2025RubikSQL: Lifelong Learning Agentic Knowledge Base as an Industrial NL2SQL System
  8. 2026DataClawEval: A Benchmark for Data Engineering Agents in Real Industrial Harness