On Semiotic-Grounded Interpretive Evaluation of Generative Art
Summary: arXiv:2604.08641v1 Announce Type: cross
Abstract: Interpretation is essential to deciphering the language of art: audiences communicate with artists by recovering meaning from visual artifacts. However, current Generative Art (GenArt) evaluators remain fixated on surface-level image quality or literal prompt adherence, failing to assess the deeper symbolic or abstract meaning intended by the creator.
We address this gap by formalizing a Peircean computational semiotic theory that models Human-GenArt Interaction (HGI) as cascaded semiosis. This framework reveals that artistic meaning is conveyed through three modes – iconic, symbolic, and indexical – yet existing evaluators operate heavily within the iconic mode, remaining structurally blind to the latter two.
The Need for a New Evaluator
Current methods of evaluating Generative Art have significant limitations. Most evaluators prioritize:
- Surface-level image quality
- Literal adherence to prompts
This approach overlooks the rich layers of meaning that can exist in artistic works, leading to a narrow understanding of artistic intent and expression.
Introducing SemJudge
To overcome this structural blindness, we propose SemJudge, a novel evaluator designed to assess symbolic and indexical meanings in Human-GenArt Interaction. SemJudge employs a Hierarchical Semiosis Graph (HSG) that reconstructs the meaning-making process from prompt to generated artifact. This innovative approach aims to provide a more nuanced understanding of the artistic process.
Key Features of SemJudge
- Comprehensive Evaluation: SemJudge evaluates art beyond mere aesthetics, taking into account the symbolic and indexical dimensions of meaning.
- Alignment with Human Judgments: Extensive quantitative experiments indicate that SemJudge aligns more closely with human judgments compared to prior evaluators, especially in interpretation-intensive fine-art benchmarks.
- Enhanced Insights: User studies reveal that SemJudge produces deeper and more insightful artistic interpretations, allowing Generative Art to evolve as a medium.
The Future of Generative Art
The introduction of SemJudge paves the way for Generative Art to transcend the creation of visually appealing images. It offers a pathway for artists to express complex human experiences and emotions, enriching the interaction between creators and audiences. The deeper understanding fostered by SemJudge may lead to a renaissance in how art is created, appreciated, and interpreted in the digital age.
Conclusion
The development of SemJudge signifies a crucial step towards a more profound evaluation of Generative Art. By incorporating semiotic theory into the evaluation process, we can unlock new dimensions of meaning and enhance the dialogue between artists and their audience. For further information and to access the project page, visit SemJudge on GitHub.