Training Emergent Joint Associations: A Reinforcement Learning Approach to Creative Thinking in Language Models

Mukul Singh, Ananya Singha, Aishni Parab, Mansi Uniyal, Pronita Mehrotra, Sumit Gulwani

Keywords: Creativity, LLM, Associativity, RL, Human Cognition, Human and AI

TL;DR: We show that reinforcement learning guided by human-inspired associative thinking improves creativity and generalization in language models, producing more novel stories and better abstraction across programming and visualization tasks

Abstract: Associative thinking is the ability to connect seemingly unrelated ideas and is a foundational element of human creativity and problem-solving. This paper explores whether reinforcement learning (RL) guided by associative thinking principles can enhance a model's performance across diverse generative tasks, including story writing, code generation, and chart creation. We introduce a reinforcement learning framework that uses a prompt-based evaluation mechanism, incorporating established divergent thinking metrics from creativity research. A base language model is fine-tuned using this framework to reward outputs demonstrating higher novelty through higher degrees of conceptual connectivity. Interestingly, the experimental results suggest that RL-based associative thinking-trained models not only generate more original and coherent stories but also exhibit improved abstraction and flexibility in tasks such as programming and data visualization. Our findings provide initial evidence that modeling cognitive creativity principles through reinforcement learning can yield more adaptive and generative AI.

Paper Type: New Full Paper

Submission Number: 64