{"ID":2834400,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.01434","arxiv_id":"2512.01434","title":"A Flexible Multi-Agent LLM-Human Framework for Fast Human Validated Tool Building","abstract":"We introduce CollabToolBuilder, a flexible multiagent LLM framework with expert-in-the-loop (HITL) guidance that iteratively learns to create tools for a target goal, aligning with human intent and process, while minimizing time for task/domain adaptation effort and human feedback capture. The architecture generates and validates tools via four specialized agents (Coach, Coder, Critic, Capitalizer) using a reinforced dynamic prompt and systematic human feedback integration to reinforce each agent's role toward goals and constraints. This work is best viewed as a system-level integration and methodology combining multi-agent in-context learning, HITL controls, and reusable tool capitalization for complex iterative problems such as scientific document generation. We illustrate it with preliminary experiments (e.g., generating state-of-the-art research papers or patents given an abstract) and discuss its applicability to other iterative problem-solving.","short_abstract":"We introduce CollabToolBuilder, a flexible multiagent LLM framework with expert-in-the-loop (HITL) guidance that iteratively learns to create tools for a target goal, aligning with human intent and process, while minimizing time for task/domain adaptation effort and human feedback capture. The architecture generates an...","url_abs":"https://arxiv.org/abs/2512.01434","url_pdf":"https://arxiv.org/pdf/2512.01434v1","authors":"[\"Daull Xavier\",\"Patrice Bellot\",\"Emmanuel Bruno\",\"Vincent Martin\",\"Elisabeth Murisasco\"]","published":"2025-12-01T09:19:18Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\"]","has_code":false}