Pith. sign in

REVIEW 6 cited by

Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.23157 v2 pith:MIR7ZKTJ submitted 2025-03-29 cs.LG cs.AIcs.DBcs.PL

classification cs.LGcs.AIcs.DBcs.PL
keywords modelsreasoningrewardstext-to-sqlpartialaccuracycapabilitiesexplicitly
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Text-to-SQL is a challenging task involving multiple reasoning-intensive subtasks, including natural language understanding, database schema comprehension, and precise SQL query formulation. Existing approaches often rely on handcrafted reasoning paths with inductive biases that can limit their overall effectiveness. Motivated by the recent success of reasoning-enhanced models such as DeepSeek R1 and OpenAI o1, which effectively leverage reward-driven self-exploration to enhance reasoning capabilities and generalization, we propose a novel set of partial rewards tailored specifically for the Text-to-SQL task. Our reward set includes schema-linking, AI feedback, n-gram similarity, and syntax check, explicitly designed to address the reward sparsity issue prevalent in reinforcement learning (RL). Leveraging group relative policy optimization (GRPO), our approach explicitly encourages large language models (LLMs) to develop intrinsic reasoning skills necessary for accurate SQL query generation. With models of different sizes, we demonstrate that RL-only training with our proposed rewards consistently achieves higher accuracy and superior generalization compared to supervised fine-tuning (SFT). Remarkably, our RL-trained 14B-parameter model significantly outperforms larger proprietary models, e.g. o3-mini by 4% and Gemini-1.5-Pro-002 by 3% on the BIRD benchmark. These highlight the efficacy of our proposed RL-training framework with partial rewards for enhancing both accuracy and reasoning capabilities in Text-to-SQL tasks.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. ReliableTableQA:How Much Supervision Does Reliability Annotation Need?

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Reliability annotation for tabular QA reaches near-perfect accuracy with about 200 schema-stratified SFT examples, and GRPO adds measurable value only when SFT is undertrained.

  2. SQLens: An End-to-End Framework for Error Detection and Correction in Text-to-SQL

    cs.CL 2025-06 conditional novelty 6.0 of 10

    SQLens uses weak supervision over database and LLM error signals to detect clause-level semantic errors in text-to-SQL and fixes them iteratively, improving execution accuracy on BIRD and Spider.

  3. RAISE: Reasoning Agent for Interactive SQL Exploration

    cs.AI 2025-06 conditional novelty 6.0 of 10

    An interactive reasoning agent that explores databases with live queries raises text-to-SQL execution accuracy from 44.8% to 56.5% and achieves 81.8% best-of-N coverage on BIRD.

  4. PaVeRL-SQL: Text-to-SQL via Partial-Match Rewards and Verbal Reinforcement Learning

    cs.AI 2025-09 reject novelty 5.0 of 10

    PaVeRL-SQL reports SOTA execution accuracy on Spider2.0-SQLite using partial-match rewards and verbal RL, but overclaims SOTA on Spider and BIRD.

  5. LogicCat: A Chain-of-Thought Text-to-SQL Benchmark for Complex Reasoning

    cs.CL 2025-05 conditional novelty 5.0 of 10

    The authors built a 4,038-question text-to-SQL benchmark with chain-of-thought annotations and found that state-of-the-art models reach only 33.20% execution accuracy.

  6. CogniSQL-R1-Zero: Lightweight Reinforced Reasoning for Efficient SQL Generation

    cs.AI 2025-07 reject novelty 3.0 of 10

    A 7B model trained with GRPO and a sparse execution-correctness reward reaches 59.97% execution accuracy on BIRD dev, though the evaluation protocol and baseline numbers contain inconsistencies.

Pith tools