{"slug": "spreadsheet-rl-advancing-llm-agents-on-realistic-spreadsheet-tasks", "title": "Spreadsheet-RL: Advancing LLM Agents on Realistic Spreadsheet Tasks", "summary": "Researchers have developed Spreadsheet-RL, a reinforcement learning framework that trains AI agents to perform complex, multi-step tasks in Microsoft Excel. The system improved the Qwen3-4B model's performance on spreadsheet benchmarks from 12.0% to 23.4% and on domain-specific tasks from 8.4% to 17.2%. This advance could enable more capable AI assistants for real-world data workflows in finance, supply chain management, and other spreadsheet-intensive fields.", "body_md": "# Computer Science > Artificial Intelligence\n\n[Submitted on 21 May 2026]\n\n# Title:Spreadsheet-RL: Advancing Large Language Model Agents on Realistic Spreadsheet Tasks via Reinforcement Learning\n\n[View PDF](/pdf/2605.22642)\n\n[HTML (experimental)](https://arxiv.org/html/2605.22642v1)\n\nAbstract:Spreadsheet systems (e.g., Microsoft Excel, Google Sheets) play a central role in modern data-centric workflows. As AI agents grow increasingly capable of automating complex tasks, such as controlling computers and generating presentations, building an AI-driven spreadsheet agent has emerged as a promising research direction. Most existing spreadsheet agents rely on specialized prompting over general-purpose LLMs; while this design has potentials on simple spreadsheet operations, it struggles to manage the complex, multi-step workflows typical of real-world applications.\n\nWe introduce Spreadsheet-RL, a reinforcement learning (RL) fine-tuning framework designed to train specialized spreadsheet agents within a realistic Microsoft Excel environment. Spreadsheet-RL features an automated pipeline for scalable collection of paired start-goal spreadsheets from online forums, as well as domain-specific evaluation tasks in areas such as finance and supply chain management, which we compile into the new Domain-Spreadsheet benchmark dataset. It also includes a Spreadsheet Gym environment designed for multi-turn RL: Spreadsheet Gym exposes extensive Excel functionality through a Python sandbox, along with a refined harness that incorporates a comprehensive tool set and carefully designed tool-routing rules for spreadsheet tasks. Through comprehensive experiments, we show that Spreadsheet-RL substantially enhances AI agent's performance on both general and domain-specific spreadsheet tasks: it improves Qwen3-4B-Thinking-2507's Pass@1 on SpreadsheetBench from 12.0% to 23.4%, and raises Pass@1 from 8.4% to 17.2% on our curated Domain-Spreadsheet dataset. These results highlight Spreadsheet-RL's strong potential for generalization and real-world adoption in spreadsheet automation, and broadly, its promise for advancing LLM-based interactions with data interfaces in everyday work.\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/spreadsheet-rl-advancing-llm-agents-on-realistic-spreadsheet-tasks", "canonical_source": "https://arxiv.org/abs/2605.22642", "published_at": "2026-05-27 12:26:23+00:00", "updated_at": "2026-05-27 12:46:11.294751+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-agents", "ai-research"], "entities": ["Microsoft Excel", "Google Sheets", "Spreadsheet-RL", "Domain-Spreadsheet"], "alternates": {"html": "https://wpnews.pro/news/spreadsheet-rl-advancing-llm-agents-on-realistic-spreadsheet-tasks", "markdown": "https://wpnews.pro/news/spreadsheet-rl-advancing-llm-agents-on-realistic-spreadsheet-tasks.md", "text": "https://wpnews.pro/news/spreadsheet-rl-advancing-llm-agents-on-realistic-spreadsheet-tasks.txt", "jsonld": "https://wpnews.pro/news/spreadsheet-rl-advancing-llm-agents-on-realistic-spreadsheet-tasks.jsonld"}}