mlsanityData to 14 Oct 2026

Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scaling

arXiv 2509.24403 ↗First posted 29 Sep 2025cs.CL cs.DBcited by 24

On the map: LLM reasoning → Knowledge- and tool-augmented reasoning

Abstract (arXiv)

State-of-the-art (SOTA) Text-to-SQL methods still lag significantly behind human experts on challenging benchmarks like BIRD. Current approaches that explore test-time scaling lack an orchestrated strategy and neglect the model's internal reasoning process. To bridge this gap, we introduce Agentar-Scale-SQL, a novel framework leveraging scalable computation to improve performance. Agentar-Scale-SQL implements an Orchestrated Test-Time Scaling strategy that synergistically combines three distinct perspectives: i) Internal Scaling via RL-enhanced Intrinsic Reasoning, ii) Sequential Scaling through Iterative Refinement, and iii) Parallel Scaling using Diverse Synthesis and Tournament Selection. Agentar-Scale-SQL is a general-purpose framework designed for easy adaptation to new databases and more powerful language models. Extensive experiments show that Agentar-Scale-SQL achieves SOTA performance on the BIRD benchmark, reaching 81.67% execution accuracy on the test set and ranking first on the official leaderboard, demonstrating an effective path toward human-level performance.

Key facts. Cited by 24 papers in the map (data through 14 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 20 of its 42 references that are in the map. First posted on arXiv 29 Sep 2025.

Builds on

20 of 42 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL22 May
  2. 2025CSC-SQL: Corrective Self-Consistency in Text-to-SQL via Reinforcement Learning19 May
  3. 2025Reward-SQL: Boosting Text-to-SQL via Stepwise Execution-Aware Reasoning and Process-Supervised Rewards7 May
  4. 2025Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL29 Mar

Other references · 16

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2020Scaling Laws for Neural Language Modelscited by 6,586
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  4. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  5. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  6. 2024OpenAI o1 System Cardcited by 1,244
  7. 2024Qwen2.5-Coder Technical Reportcited by 1,203
  8. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?cited by 223
  9. 2023C3: Zero-shot Text-to-SQL with ChatGPTcited by 164
  10. 2025OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scalecited by 91
  11. 2024A Survey of Text-to-SQL in the Era of LLMs: Where are we, and where are we going?cited by 56
  12. 2025XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQLcited by 40
4 more
  1. 2025Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Searchcited by 34
  2. 2024SQLBench: A Comprehensive Evaluation for Text-to-SQL Capabilities of Large Language Modelscited by 32
  3. 2025Automatic Metadata Extraction for Text-to-SQLcited by 27
  4. 2025A State-of-the-Art SQL Reasoning Model using RLVRcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 24.

24 papers.

Sort:

By area:

  1. 2026Arming Data Agents with Tribal Knowledgecited by 7
  2. 2026Text2GraphQuery-Bench: A Text to Graph Query Benchmarkcited by 3
  3. 2026TRUST-SQL: Tool-Integrated Multi-Turn Reinforcement Learning for Text-to-SQL over Unknown Schemascited by 3
  4. 2026ELT-Bench-Verified: Benchmark Quality Issues Underestimate AI Agent Capabilitiescited by 3
  5. 2026Human-Level Text-to-SQL via Reinforcement Learning on Verified Data, Without Pipeline Engineeringcited by 1
  6. 2026R3-SQL: Ranking Reward and Resampling for Text-to-SQLcited by 1
  7. 2026Every Step Counts: Step-Level Credit Assignment for Tool-Integrated Text-to-SQLcited by 1
  8. 2026TAHOE: Text-to-SQL with Automated Hint Optimization from Experiencecited by 1
  9. 2026SQLConductor: Search-to-Policy Learning for Step-wise Text-to-SQL Orchestrationcited by 1
  10. 2026IESR:Efficient MCTS-Based Modular Reasoning for Text-to-SQL with Large Language Modelscited by 0
  11. 2026Dial: A Knowledge-Grounded Dialect-Specific NL2SQL Systemcited by 0
  12. 2026SIRIUS-SQL: Anchoring Multi-Candidate Text-to-SQL in Execution Feedbackcited by 0
12 more
  1. 2026Data Intelligence Agents: Interpreting, Modeling, and Querying Enterprise Data via Autonomous Coding Agentscited by 0
  2. 2026SERL-SQL: Selective Hindsight Distillation for Text-to-SQL Reinforcement Agentic Learningcited by 0
  3. 2026ReToolSQL: Agentic Reinforcement Learning for Robust Text-to-SQLcited by 0
  4. 2026Are These Modules Worth Their Cost? A Paradigm-Level Accuracy-Cost Analysis of In-context Learning Text-to-SQLcited by 0
  5. 2026Evaluating LLMs on Conversational Text-to-SQL under Chain Ambiguity and Intent Driftcited by 0
  6. 2026Replacing Training with Memory: Listwise Selection for Text-to-SQLcited by 0
  7. 2026A Cost-Aware Agentic Architecture for NL-to-SQL over Nested Enterprise Schemas, with a New Benchmarkcited by 0
  8. 2026DualSQL: Text-to-SQL with Multi-Agent Reinforcement Learningcited by 0
  9. 2026LIMIT: Less Is More for Instruction Tuning in Text-to-SQLcited by 0
  10. 2026AptMQL-Bench: From Text-to-SQL to Text-to-MQL via Access-Pattern Schema Design and Data-Preserving Migrationcited by 0
  11. 2026Why, Where, How: Taxonomy-guided Error Grounding for Code Repair in NL2SQLcited by 0
  12. 2026ARCS: Towards Precise Text-to-SQL via Structured Disambiguationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Rethinking Agentic Workflows: Evaluating Inference-Based Test-Time Scaling Strategies in Text2SQL Tasks
  2. 2025BAPPA: Benchmarking Agents, Plans, and Pipelines for Automated Text-to-SQL Generation
  3. 2025RAISE: Reasoning Agent for Interactive SQL Exploration
  4. 2026APEX-SQL: Talking to the data via Agentic Exploration for Text-to-SQL
  5. 2026PExA: Parallel Exploration Agent for Complex Text-to-SQL
  6. 2026RoboPhD: Self-Improving Text-to-SQL Through Autonomous Agent Evolution
  7. 2025JudgeSQL: Reasoning over SQL Candidates with Weighted Consensus Tournament
  8. 2026AgentSM: Semantic Memory for Agentic Text-to-SQL