Chatting with your ERP: A Recipe

2025/07/31 by Lidia Andrés Susinos, Gómez, Jorge Ruiz, Jorge Alamo Olivé +5
Business, Management and Accounting · #68P20 #68T50 #Artificial Intelligence (cs.AI) #Databases (cs.DB) #ERP Systems Implementation and Impact #Emerging Technologies (cs.ET) #FOS: Computer and information sciences #H.2.5 #H.2.8 #H.5.m #Human-Computer Interaction (cs.HC) #I.2.7 #Multiagent Systems (cs.MA)

paper · pdf · doi:10.48550/arxiv.2507.23429

openalex publication_date 2025/07/31 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This paper presents the design, implementation, and evaluation behind a Large Language Model (LLM) agent that chats with an industrial production-grade ERP system. The agent is capable of interpreting natural language queries and translating them into executable SQL statements, leveraging open-weight LLMs. A novel dual-agent architecture combining reasoning and critique stages was proposed to improve query generation reliability.

Citations

• MAG-SQL: Multi-Agent Generative Approach with Soft Schema Linking and Iterative Sub-SQL Refinement for Text-to-SQL
• SQLFixAgent: Towards Semantic-Accurate Text-to-SQL Parsing via Consistency-Enhanced Multi-Agent Collaboration
• Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation
• Teaching Large Language Models to Self-Debug
• A comprehensive evaluation of ChatGPT's zero-shot Text-to-SQL capability
• Evaluating the Text-to-SQL Capabilities of Large Language Models
• Long Short-Term Memory

Related

• KnowThyself: An Agentic Assistant for LLM Interpretability
• CHOPS: CHat with custOmer Profile Systems for Customer Service with LLMs
• Chit-Chat or Deep Talk: Prompt Engineering for Process Mining
• BAPPA: Benchmarking Agents, Plans, and Pipelines for Automated Text-to-SQL Generation
• A Practical Approach for Building Production-Grade Conversational Agents with Workflow Graphs