nl2sql

ESQ-Bench is a new Oracle-first NL2SQL benchmark that evaluates model performance on complex enterprise schemas, revealing significant degradation and silent semantic divergence compared to academic benchmarks.

Polaris trains LLMs to generate table descriptions directly from retrieval feedback using DPO, significantly outperforming previous methods like AutoDDG in table retrieval tasks.

This paper introduces Schema-Aware Localisation (SAL), a lightweight middleware that grounds LLM-generated SQL queries in live Oracle schema catalog to eliminate hallucination errors, achieving 62.6% execution-grounded truth on TPC-H questions without manual schema curation.

This paper systematically analyzes multiple pipeline extension components for Natural Language to SQL translation, including intermediate representations, synthetic data, preprocessing, and a reranker, using SmBoP and RASAT architectures, and finds that their interactions matter more than simply combining all components.

This paper presents a semantic-layer-mediated NL2SQL agent that decouples intent from physical execution by reasoning over a curated semantic model, achieving 94.15% execution accuracy on the Spider2-snow benchmark.

Soma-SQL proposes an autonomous method to resolve multi-source ambiguity in natural language to SQL translation using synthetic query logs and ambiguity-driven execution probing, achieving 13% improvement in execution accuracy over state-of-the-art baselines.

This paper presents AgentNLQ, a multi-agent system for natural language to SQL conversion that achieves 78.1% semantic accuracy on the BIRD benchmark through schema enrichment and a self-correcting orchestrator.

FD-NL2SQL is a feedback-driven natural language to SQL system for clinical oncology databases that improves with use through clinician edits and logic-based SQL augmentation. The system decomposes natural language questions into predicates, retrieves expert-verified exemplars, and synthesizes executable SQL with continuous learning capabilities.

ROSE is a novel intent-centered evaluation metric for NL2SQL that uses a Prover-Refuter cascade to assess semantic correctness independently of ground-truth SQL, achieving 24% better agreement with human experts than existing metrics. The paper addresses limitations of Execution Accuracy and provides a re-evaluation of 19 NL2SQL methods with publicly released resources.