arXiv:2505. 04671v3 Announce Type: replace-cross Abstract: Recent advances in large language models (LLMs) trained with reinforcement learning (RL) have improved Text-to-SQL performance.
State-of-the-art Text-to-SQL systems are typically multi-agent pipelines centered around two fundamental tasks: schema linking and SQL generation. However, existing work trains separate models for eac...
DualSQL is a Text-to-SQL system that uses two agents sharing a single model backbone, enabling joint optimization via multi-agent reinforcement learning. The approach incorporates three database access tools for multi-step reasoning, rollout guardrails to stabilize training, and a new SQL correctness metric called robust execution match (REX). Trained on only 3,755 examples, DualSQL-4B reaches 68.0% execution accuracy on the BIRD dev set, while DualSQL-8B achieves 71.1%, surpassing prior state‑of‑the‑art single‑model solutions with 32B parameters.
arXiv:2606. 06825v1 Announce Type: cross Abstract: Reinforcement learning has recently shown promise in improving large language models for Text-to-SQL generation, yet existing methods typically optimize one-shot rewards defined over a single SQL state.
arXiv:2606. 11209v1 Announce Type: cross Abstract: Visual question answering increasingly requires multi-step reasoning.
arXiv:2603.20004v4 Announce Type: replace-cross Abstract: Translating natural language questions to SQL queries (Text-to-SQL) is a long-standing problem in database research. Recent efforts have focu...
arXiv:2606. 04503v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has greatly advanced large reasoning models (LRMs), but it requires timely training on a huge fully-annotated dataset.
arXiv:2509. 06278v4 Announce Type: replace Abstract: Table reasoning requires models to jointly perform comprehensive semantic understanding and precise numerical operations.
SPOC-SQL introduces a stage-wise approach to Text-to-SQL, breaking the task into four sequential subtasks aligned with standard SQL execution logic. It applies fine-grained preference optimization at key decision points and a structured decomposition strategy, enabling explicit intermediate representations for stage-wise intervention and correction. The method yields more controllable and reliable SQL generation, with experiments showing that incorporating stage-wise human knowledge consistently improves performance.
arXiv:2606. 30851v1 Announce Type: cross Abstract: Improving the reliability of large language models (LLMs) at inference time is a central challenge in structured reasoning tasks such as Text-to-SQL.
The paper introduces Stepwise Marginal Information Gain (MIG), an intrinsic process reward that evaluates how each reasoning step of a large language model (LLM) or vision-language model (VLM) improves the likelihood of the reference answer. MIG rewards only new likelihood maxima, preventing duplicate credit, and is combined with outcome, format, and self‑distillation objectives to guide training. Experiments on eight benchmarks show that this method outperforms outcome‑only reinforcement learning and improves accuracy by up to 4.8 points over binary‑reward training, including a 12.6‑point gain on MathVerse and a 12.9‑point advantage on vision‑language transfer at 7B parameters.
arXiv:2609.20842v1 Announce Type: new Abstract: Text-to-SQL translates natural-language questions into executable SQL queries, but open-source large language models still require task-specific post-t...
arXiv:2606. 15866v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has become an effective post-training paradigm for improving the reasoning abilities of large language models.