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

The paper introduces ArtMine, a framework for discovering and formalizing artistic processes from various historical sources. It uses an abductive agent to infer evidence-based production steps which are then optimized through self-reflection.

What the Document Says

• ArtMine is a framework for discovering and formalizing artistic processes
• The framework synthesizes heterogeneous artwork evidence into a structured repository
• An abductive agent infers evidence-grounded production steps from the repository
• The steps are converted into a compositional graph and rendering prompt
• The generated and reference artworks are optimized through self-reflection
• A case study is provided using open-domain historical sources across multiple artists and artistic movements

Key Concepts Mentioned

• ArtMine
• abductive agent
• heterogeneous artwork evidence
• production steps
• compositional graph
• rendering prompt
• self-reflection

Claims Made by the Document

• ArtMine can discover and formalize artistic processes from various historical sources
• The approach supports coherent, interpretable, and auditable representations of artistic workflows

Evidence Used

• Description of ArtMine's methodology in the text

Methods / Approach

• Synthesizing heterogeneous artwork evidence
• Using an abductive agent to infer production steps
• Converting steps into compositional graph and rendering prompt
• Optimization through self-reflection

Limitations Stated by the Authors

The paper provides a preliminary proof-of-concept case study, so the effectiveness of ArtMine on a larger scale is unknown

Limitations Noted by the Agent

Agent interpretation — not from the source.

The agent's ability to accurately infer and optimize artistic processes may be limited by the quality and completeness of available historical sources

Open Questions

• How effective is ArtMine at discovering and formalizing artistic processes on a larger scale?
• To what extent can the agent accurately infer and optimize artistic processes?

Notes for Future Q&A

May include interpretation. Labeled where it does.

The paper does not discuss the potential impact of ArtMine on human-AI co-creativity systems, artistic interpretation, creative education, reflective collaboration, or computational studies of cultural production. These aspects could be explored in further research.

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.