A Hybrid Framework for Natural Language Querying of IFC Models with Relational and Graph Representations

2026-05-13

AI Summary

This work addresses the challenge that non-expert users face in querying complex and semantically abstract IFC-based BIM models directly. To overcome this, the authors propose IfcLLM, a hybrid framework that uniquely integrates both relational and graph-structured representations of IFC data and leverages large language models (LLMs) with iterative retry and refinement mechanisms to enable high-accuracy natural language querying. Built upon the open-source GPT-OSS 120B model, the approach unifies structured IFC attributes, geometric information, and topological relationships into a reproducible and deployable query pipeline. Experiments on three real-world IFC models demonstrate first-round query accuracy ranging from 93.3% to 100%, with all initially failed queries successfully recovered through a backup LLM, effectively supporting common BIM analysis tasks.

Abstract

Building Information Modeling (BIM) is widely used in the Architecture, Engineering, and Construction (AEC) industry, but the complexity of Industry Foundation Classes (IFC) limits accessibility for non-expert users. To address this, we introduce IfcLLM, a hybrid framework for natural language interaction with IFC-based BIM models. It transforms IFC models into complementary representations: a relational representation for structured element properties and geometry, and a graph representation for topological relationships. These representations are integrated through iterative retry-and-refine LLM reasoning. We implement the framework using an open-weight LLM (GPT OSS 120B), supporting reproducible and deployment-oriented workflows. Evaluation on three IFC models with queries derived from 30 scenarios shows first-attempt accuracy of 93.3%-100%, with all failures recovered using a fallback LLM. The results show that combining complementary representations with iterative reasoning enables more accessible natural language querying of IFC data while supporting routine BIM analysis tasks.

Problem

• Natural Language Querying
• IFC Models
• BIM
• Accessibility
• Non-expert Users

Innovation

• hybrid representation
• natural language querying
• IFC models
• iterative LLM reasoning
• BIM accessibility