Architectural Critique with Artificial Intelligence: Generating Architectural Reviews through Vision-Language Models DOI Creative Commons
Selen Çiçek,

Mehmet Sadık Aksu,

E. Öztürk

et al.

Journal of Computational Design, Journal Year: 2025, Volume and Issue: 6(1), P. 165 - 190

Published: March 28, 2025

Artificial Intelligence (AI) offers a potent opportunity to rethink architectural critique, in cases such as design competitions. The challenge lies capturing the interpretive depth required for evaluation—an inherently human process that connects intuition, reasoning, and contextual sensitivity. Building on this premise, proposed approach uses domain-specific dataset, curated validated by experienced architects domain experts, train context-aware Visual-Language Model (VLM) capable of delivering nuanced critique. model development follows two distinct phases: an initial prototype (v1) explores feasibility through classification visual attributes, while second phase (v2) evolves into structure generating detailed critique texts guided predefined criteria context, form, programmatic considerations. aims bridge gap between computational precision complexities judgment, offering structured yet adaptable framework utilizing AI evaluative aspects design.By integrating ecological intelligence framework, can also assess designs based their environmental impact sustainability practices, encouraging holistic aligns innovation with responsibility. Although still its early stages, work opens pathway complement traditional review processes reliable, scalable, context-sensitive feedback, laying foundation incorporating patterns tacit knowledge process.

Language: Английский

Architectural Critique with Artificial Intelligence: Generating Architectural Reviews through Vision-Language Models DOI Creative Commons
Selen Çiçek,

Mehmet Sadık Aksu,

E. Öztürk

et al.

Journal of Computational Design, Journal Year: 2025, Volume and Issue: 6(1), P. 165 - 190

Published: March 28, 2025

Artificial Intelligence (AI) offers a potent opportunity to rethink architectural critique, in cases such as design competitions. The challenge lies capturing the interpretive depth required for evaluation—an inherently human process that connects intuition, reasoning, and contextual sensitivity. Building on this premise, proposed approach uses domain-specific dataset, curated validated by experienced architects domain experts, train context-aware Visual-Language Model (VLM) capable of delivering nuanced critique. model development follows two distinct phases: an initial prototype (v1) explores feasibility through classification visual attributes, while second phase (v2) evolves into structure generating detailed critique texts guided predefined criteria context, form, programmatic considerations. aims bridge gap between computational precision complexities judgment, offering structured yet adaptable framework utilizing AI evaluative aspects design.By integrating ecological intelligence framework, can also assess designs based their environmental impact sustainability practices, encouraging holistic aligns innovation with responsibility. Although still its early stages, work opens pathway complement traditional review processes reliable, scalable, context-sensitive feedback, laying foundation incorporating patterns tacit knowledge process.

Language: Английский

Citations

0