Pith. sign in

Probing and Inducing Combinational Creativity in Vision-Language Models

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

The ability to combine existing concepts into novel ideas stands as a fundamental hallmark of human intelligence. Recent advances in Vision-Language Models (VLMs) like GPT-4V and DALLE-3 have sparked debate about whether their outputs reflect combinational creativity--defined by M. A. Boden (1998) as synthesizing novel ideas through combining existing concepts--or sophisticated pattern matching of training data. Drawing inspiration from cognitive science, we investigate the combinational creativity of VLMs from the lens of concept blending. We propose the Identification-Explanation-Implication (IEI) framework, which decomposes creative processes into three levels: identifying input spaces, extracting shared attributes, and deriving novel semantic implications. To validate this framework, we curate CreativeMashup, a high-quality dataset of 666 artist-generated visual mashups annotated according to the IEI framework. Through extensive experiments, we demonstrate that in comprehension tasks, best VLMs have surpassed average human performance while falling short of expert-level understanding; in generation tasks, incorporating our IEI framework into the generation pipeline significantly enhances the creative quality of VLMs' outputs. Our findings establish both a theoretical foundation for evaluating artificial creativity and practical guidelines for improving creative generation in VLMs.

citation-role summary

background 1

citation-polarity summary

fields

cs.CL 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

background 1

representative citing papers

GuessBench: Sensemaking Multimodal Creativity in the Wild

cs.CL · 2025-06-01 · conditional · novelty 7.0

A Minecraft-based benchmark shows vision-language models often fail to decode player-built creations, with accuracy falling sharply for rare concepts and low-resource languages.

citing papers explorer

Showing 1 of 1 citing paper.

  • GuessBench: Sensemaking Multimodal Creativity in the Wild cs.CL · 2025-06-01 · conditional · none · ref 63 · internal anchor

    A Minecraft-based benchmark shows vision-language models often fail to decode player-built creations, with accuracy falling sharply for rare concepts and low-resource languages.