AI

Beyond Explicit Intent: Discovering and Shaping Artistic Intent through Human-AI Co-Creation

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 20/10/2026, 04:03:24 UTC.

Briefings

This article proposes a human-AI co-creation paradigm where artistic intent is discovered through observation rather than explicit text prompts. By expanding model exploration and integrating curatorial judgments into reinforcement learning, the system creates a continuous loop of explore-identify-shape-re-explore. Three user studies examined latent-preference discovery, personalized preference alignment, and the externalization of vague creative intentions.

Key points

• The system uses a 'continuous explore-identify-shape-re-explore' loop to discover latent preferences through observation and selection rather than explicit text prompts.
• Artists' curatorial judgments are integrated into reinforcement learning, assigning different learning strengths to generation trajectories to align with individual aesthetics.

Context

Generative AI is increasingly used for artistic creation, but existing interactions often assume creators can specify intentions in advance. This paradigm shifts focus to discovering and shaping intent through human-AI co-creation.

Limitations

Evidence is limited to three user studies; specific participant numbers, sample sizes, and detailed demographic data are not provided. The study period is not specified, and the exact date of the arXiv preprint is 2026-10-20.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 20/10/2026, 04:13:50 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeiccxa7douh3wc5nrsex67bzd4emb6s5wcfbjy7jiitzydtrzf5u7m