Cultural Alien Sampler: Open-ended art generation balancing originality and coherence

2025-10-21

AI Summary

In open-domain artistic creation, large language models (LLMs) struggle to simultaneously achieve originality and internal consistency—often defaulting to cultural clichés or sacrificing coherence for novelty. This paper proposes *cultural heterosampling*, a method that decouples conceptual plausibility (internal consistency) from cultural typicality (external familiarity). It employs a dual-model architecture: a fine-tuned GPT-2 model trained on WikiArt for concept consistency assessment, and a cultural context model estimating typicality. Together, they navigate the creative space toward high-consistency, low-typicality concept combinations, enabling controllable cultural alienation. Human evaluation shows significant improvements in both originality and harmony over random baselines and GPT-4o, approaching the performance of art-major students. Quantitative analysis further confirms broader conceptual coverage and higher distributional diversity in generated outputs.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: Generation

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

In open-ended domains like art, autonomous agents must generate ideas that are both original and internally coherent, yet current Large Language Models (LLMs) either default to familiar cultural patterns or sacrifice coherence when pushed toward novelty. We address this by introducing the Cultural Alien Sampler (CAS), a concept-selection method that explicitly separates compositional fit from cultural typicality. CAS uses two GPT-2 models fine-tuned on WikiArt concepts: a Concept Coherence Model that scores whether concepts plausibly co-occur within artworks, and a Cultural Context Model that estimates how typical those combinations are within individual artists' bodies of work. CAS targets combinations that are high in coherence and low in typicality, yielding ideas that maintain internal consistency while deviating from learned conventions and embedded cultural context. In a human evaluation (N = 100), our approach outperforms random selection and GPT-4o baselines and achieves performance comparable to human art students in both perceived originality and harmony. Additionally, a quantitative study shows that our method produces more diverse outputs and explores a broader conceptual space than its GPT-4o counterpart, demonstrating that artificial cultural alienness can unlock creative potential in autonomous agents.

Problem

• Balancing originality and coherence in open-ended art generation
• Addressing LLMs' tendency toward cultural patterns or incoherent novelty
• Separating compositional fit from cultural typicality for creative ideas

Innovation

• CAS separates compositional fit from cultural typicality
• Uses two fine-tuned GPT-2 models for concept scoring
• Targets high coherence and low typicality combinations