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

2026-05-13 · Source: Takara TLDR - Daily AI Papers · Field: Construction & Real Estate — Construction Technology & Building, Architecture & Urban Planning · Depth: Expert, medium

Summary

IfcLLM is a new hybrid framework designed to enable natural language interaction with Industry Foundation Classes (IFC) models, which are complex Building Information Modeling (BIM) data structures. It addresses the challenge of limited accessibility for non-expert users by transforming IFC models into two complementary representations: a relational format for structured properties and geometry, and a graph format for topological relationships. These representations are then integrated through an iterative retry-and-refine reasoning process powered by a Large Language Model (LLM). The framework was implemented using an open-weight LLM (GPT OSS 120B) and evaluated on three IFC models with queries derived from 30 scenarios, achieving a first-attempt accuracy of 93.3%-100%, with all initial failures successfully recovered using a fallback LLM.

Key takeaway

For AI Scientists developing natural language interfaces for complex domain-specific data, IfcLLM demonstrates a robust approach. You should consider adopting a hybrid data representation strategy, combining relational and graph structures, to capture both property details and topological relationships. Furthermore, integrating iterative LLM reasoning with a fallback mechanism can significantly improve query accuracy and user experience, making complex data more accessible.

Key insights

Combining relational and graph representations with iterative LLM reasoning enhances IFC model accessibility.

Principles

• Complementary data representations improve LLM reasoning.
• Iterative refinement enhances LLM accuracy and robustness.

Method

IfcLLM transforms IFC models into relational and graph representations, then uses iterative retry-and-refine LLM reasoning to integrate them for natural language querying.

In practice

• Use hybrid data models for complex domain querying.
• Implement LLM fallback mechanisms for error recovery.

Topics

• IFC Models
• Natural Language Querying
• Building Information Modeling
• Large Language Models
• Hybrid Data Representation

Best for: AI Scientist, AI Engineer, NLP Engineer, Research Scientist

Related on AIssential

• Integrating Graphs, Large Language Models, and Agents: Reasoning and Retrieval — Integrating graphs with LLMs enhances reasoning, retrieval, and decision-making by combining structured knowledge with semantic understanding.
• Combining Semantic Embeddings and Knowledge Graphs for Identifying Decision Patterns in Brazilian Judicial Decisions — Hybrid representations combining semantic embeddings and knowledge graphs enhance judicial decision pattern identification.
• LLMs on Tabular Data with Limited Semantics: Evidence from Industrial Car Retrofit Prediction — LLMs complement, but do not replace, strong tabular models on privacy-constrained industrial data lacking semantic cues.
• Construction of Knowledge Graph based on Language Model — PLMs significantly enhance Knowledge Graph construction by automating entity and relation extraction from text.
• How SHACL Makes Your LLMs Hum — Integrating SHACL-defined knowledge graphs with LLMs significantly reduces hallucinations and improves response accuracy.
• Code-on-Graph: Iterative Programmatic Reasoning via Large Language Models on Knowledge Graphs — Code-on-Graph integrates LLMs with KGs using programmatic reasoning, representing schemas as Python classes for flexible, scalable knowledge utilization.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Takara TLDR - Daily AI Papers.