The field of natural language processing is witnessing significant advancements in the application of large language models (LLMs) to text-to-SQL and process analysis tasks. Recent developments indicate a shift towards more efficient, scalable, and interpretable approaches. Researchers are exploring novel architectures, such as agentic frameworks and multi-expert systems, to improve the accuracy and robustness of text-to-SQL systems. Additionally, there is a growing focus on reducing computational costs and improving the sustainability of LLM-based systems. Noteworthy papers in this area include AGENTIQL, which proposes an agent-inspired multi-expert framework for text-to-SQL generation, and Agentic NL2SQL, which introduces an agentic system to reduce computational costs in NL2SQL tasks. These innovative approaches are expected to have a significant impact on the field, enabling more efficient and effective processing of natural language queries and process models.
Advances in Large Language Models for Text-to-SQL and Process Analysis
Sources
Evaluating LLM-Based Process Explanations under Progressive Behavioral-Input Reduction
Agentic Troubleshooting Guide Automation for Incident Management
AGENTIQL: An Agent-Inspired Multi-Expert Framework for Text-to-SQL Generation
Rethinking Agentic Workflows: Evaluating Inference-Based Test-Time Scaling Strategies in Text2SQL Tasks
Task-Aware Reduction for Scalable LLM-Database Systems
Information Extraction from Conversation Transcripts: Neuro-Symbolic vs. LLM
Classifier-Augmented Generation for Structured Workflow Prediction
MTSQL-R1: Towards Long-Horizon Multi-Turn Text-to-SQL via Agentic Training
Bridging the Semantic Gap: Contrastive Rewards for Multilingual Text-to-SQL
BenchPress: A Human-in-the-Loop Annotation System for Rapid Text-to-SQL Benchmark Curation
FACTS: Table Summarization via Offline Template Generation with Agentic Workflows
FinAI Data Assistant: LLM-based Financial Database Query Processing with the OpenAI Function Calling API
Rethinking Schema Linking: A Context-Aware Bidirectional Retrieval Approach for Text-to-SQL
Agentic NL2SQL to Reduce Computational Costs