ArtMine: Discovering and Formalizing Artistic Processes

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Kaustubh Kumar, Ashutosh Ranjan, Vivek Srivastava, Blessin Varkey, Shirish Karande
• Venue: arXiv: cs.LG
• Published: 2026-07-09
• Collected: 2026-07-10
• URL: http://arxiv.org/abs/2607.08331v1
• PDF: https://arxiv.org/pdf/2607.08331v1
• DOI: UNKNOWN

Plain-Language Brief

ArtMine is a framework designed to discover and formalize artistic processes from various historical sources. It uses an abductive agent to infer evidence-based production steps, which are then optimized to generate artworks. The goal is to create a system that models creative processes instead of just final artifacts, with potential applications in co-creativity systems for interpretation, education, collaboration, and cultural studies.

What the Document Says

• ArtMine is a framework for discovering and formalizing artistic processes.
• It uses a Peircean abductive agent to infer evidence-based production steps from heterogeneous artwork evidence.
• These steps are converted into a compositional graph and rendering prompt, then optimized through self-reflection over deviations between the generated and reference artworks.
• The framework was tested with open-domain historical sources across multiple artists and artistic movements.
• The goal is to create a system that models creative processes rather than just final artifacts.
• Potential applications include co-creativity systems for interpretation, education, collaboration, and cultural studies.

Key Concepts Mentioned

• ArtMine
• artistic processes
• Peircean abductive agent
• heterogeneous artwork evidence
• compositional graph
• rendering prompt
• self-reflection
• open-domain historical sources
• co-creativity systems
• interpretation
• education
• collaboration
• cultural studies

Claims Made by the Document

• ArtMine can discover and formalize artistic processes.
• It can optimize the generated artworks to match reference artworks.

Evidence Used

• The introduction of ArtMine as a framework for discovering and formalizing artistic processes.
• The explanation of how the abductive agent infers evidence-based production steps.

Methods / Approach

• Synthesizing heterogeneous artwork evidence into a structured repository
• Inferring evidence-based production steps using a Peircean abductive agent
• Converting the steps into a compositional graph and rendering prompt
• Optimizing through self-reflection over deviations between the generated and reference artworks

Limitations Stated by the Authors

• The authors note that artistic workflows are only partially documented, making process-level understanding difficult to formalize computationally.

Limitations Noted by the Agent

Agent interpretation — not from the source.

The agent may face challenges in accurately inferring production steps when the available evidence is sparse or ambiguous.

Open Questions

• What is the accuracy of ArtMine's inferred production steps compared to manual analysis?
• How does the optimization process affect the creative output and authenticity of the generated artworks?

Notes for Future Q&A

May include interpretation. Labeled where it does.

• The authors provide a case study as a preliminary proof-of-concept, but it is unclear how extensively the system has been tested or validated.

Raw Abstract / Excerpt

Understanding how artworks are created requires reasoning about the iterative decisions, material operations, and contextual influences that shape artistic production. While recent generative AI systems can synthesize artworks with high fidelity, they primarily model distributions over finished artifacts rather than the creative processes underlying their creation. In practice, artistic workflows are only partially documented through fragmented sources such as archival records, preparatory studies, correspondence, etc., making process-level understanding difficult to formalize computationally. In this work, we introduce ArtMine, a framework for discovering and formalizing artistic processes from heterogeneous historical evidence. Our approach synthesizes heterogeneous artwork evidence into a structured repository, from which a Peircean abductive agent infers evidence-grounded production steps. These steps are converted into a compositional graph and rendering prompt, then optimized through self-reflection over deviations between the generated and reference artworks. We provide a preliminary proof-of-concept case study using open-domain historical sources across multiple artists and artistic movements, demonstrating that fragmented documentary evidence can support coherent, interpretable, and auditable representations of artistic workflows. By modeling creative processes rather than only final artifacts, our work moves toward process-centred human-AI co-creativity systems that can support artistic interpretation, creative education, reflective collaboration, and computational studies of cultural production.