Higharc Wins Best Industrial Paper Award at EUVIP 2026

Higharc, a leading homebuilding AI company, has won the Best Industrial Paper Award at EUVIP 2026 for its groundbreaking research. The award recognizes the company's innovative approach to generating room layouts within architectural constraints using a BIM-native approach.

The research, titled "Small Building Model (SBM)," was authored by Manuel Rodríguez Ladrón de Guevara, Jinmo Rhee, Ardavan Bidgoli, Vaidas Razgaitis, and Michael Bergin. It introduces a domain-specific Transformer model that encodes rooms as structured sequences of building elements, preserving categorical and continuous attributes such as wall types, dimensions, positions, and spatial relationships. This novel representation enables the model to analyze and retrieve similar room configurations and generate layouts by predicting the placement of room contents within an existing architectural envelope.

The Higharc team spent approximately a year developing the research results. Their work demonstrated stronger inventory coverage and navigability compared to two specialized layout-generation baselines trained on the same data. The results support the researchers' central finding that preserving architectural information in the model's input representation plays a crucial role in generating usable layouts.

This award-winning paper was published on arXiv and can be accessed at arxiv.org/abs/2512.04832. Higharc's Generative Building Model (GBM), which powers generative design in the company's platform, draws on this research. GBM uses structured building data to generate interior layouts within existing BIM room geometry, accounting for measured walls, doors, windows, and the relationships between building elements.

Higharc's commitment to advancing the field of AI in the housing industry is evident in its rapid rise. The company recently ranked in the top 50 of the Deloitte Technology Fast 500 and was named one of Fast Company’s Most Innovative Companies of 2026. These accolades underscore the company's innovative approach and its potential to transform the homebuilding industry through the integration of AI technologies.

The research addresses a fundamental challenge in AI-generated architectural layouts: how to represent building geometry and constraints so that a machine learning model can produce usable designs within BIM environments. The hardest part, according to Rodríguez Ladrón de Guevara, was deciding how to represent a room so that the model could learn from its geometry and produce something a designer could actually use. This representation problem, more than the ML model itself, became the central question of the research.

Higharc's work is significant not just for its technical achievement but also for its potential to build the foundational infrastructure that makes housing a principal domain for machine learning. The company hopes that this recognition will inspire more researchers to tackle the enormous challenges and opportunities ahead in the field of AI and homebuilding.