# BIM-GPT: a Prompt-Based Virtual Assistant Framework for BIM Information Retrieval

arXiv: https://arxiv.org/abs/2304.09333 · HTML: https://mlsanity.com/paper/2304.09333/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 66 references that are in the map.
- First posted on arXiv 18 Apr 2023.

## Builds on (most cited first, 13 of 13)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [Distributed Representations of Words and Phrases and their Compositionality](https://mlsanity.com/paper/1310.4546/) (2013)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://mlsanity.com/paper/2302.04023/) (2023)
- [ChatIE: Zero-Shot Information Extraction via Chatting with ChatGPT](https://mlsanity.com/paper/2302.10205/) (2023)
- [PromptChainer: Chaining Large Language Model Prompts through Visual Programming](https://mlsanity.com/paper/2203.06566/) (2022)
- [Exploring the Limits of ChatGPT for Query or Aspect-based Text Summarization](https://mlsanity.com/paper/2302.08081/) (2023)
- [a survey on GPT-3](https://mlsanity.com/paper/2212.00857/) (2022)

## Built on by (most cited first, 12 of 12)

- [A Comprehensive Survey on Process-Oriented Automatic Text Summarization with Exploration of LLM-Based Methods](https://mlsanity.com/paper/2403.02901/) (2024)
- [Information Retrieval Meets Large Language Models: A Strategic Report from Chinese IR Community](https://mlsanity.com/paper/2307.09751/) (2023)
- [GPT Models in Construction Industry: Opportunities, Limitations, and a Use Case Validation](https://mlsanity.com/paper/2305.18997/) (2023)
- [Generative AI in the Construction Industry: Opportunities & Challenges](https://mlsanity.com/paper/2310.04427/) (2023)
- [Core Building Blocks: Next Gen Geo Spatial GPT Application](https://mlsanity.com/paper/2310.11029/) (2023)
- [LLM-assisted Graph-RAG Information Extraction from IFC Data](https://mlsanity.com/paper/2504.16813/) (2025)
- [Context-aware LLM-based AI Agents for Human-centered Energy Management Systems in Smart Buildings](https://mlsanity.com/paper/2512.25055/) (2025)
- [Foundation Models for Autonomous Robots in Unstructured Environments](https://mlsanity.com/paper/2407.14296/) (2024)
- [Prompt to Protection: A Comparative Study of Multimodal LLMs in Construction Hazard Recognition](https://mlsanity.com/paper/2506.07436/) (2025)
- [Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements](https://mlsanity.com/paper/2605.22079/) (2026)
- [BIM-Edit: Benchmarking Large Language Models for IFC-Based Building Information Modeling](https://mlsanity.com/paper/2606.20146/) (2026)
- [A Building as a Repository: KIR, a Typed Intermediate Representation for Agent-Authored Building Information Models](https://mlsanity.com/paper/2609.14578/) (2026)

## Abstract (arXiv)

Efficient information retrieval (IR) from building information models (BIMs) poses significant challenges due to the necessity for deep BIM knowledge or extensive engineering efforts for automation. We introduce BIM-GPT, a prompt-based virtual assistant (VA) framework integrating BIM and generative pre-trained transformer (GPT) technologies to support NL-based IR. A prompt manager and dynamic template generate prompts for GPT models, enabling interpretation of NL queries, summarization of retrieved information, and answering BIM-related questions. In tests on a BIM IR dataset, our approach achieved 83.5% and 99.5% accuracy rates for classifying NL queries with no data and 2% data incorporated in prompts, respectively. Additionally, we validated the functionality of BIM-GPT through a VA prototype for a hospital building. This research contributes to the development of effective and versatile VAs for BIM IR in the construction industry, significantly enhancing BIM accessibility and reducing engineering efforts and training data requirements for processing NL queries.