When Comics Chat: a deep prior for text-balloon reading order
Uses deep learning on a corpus of comic text-balloons to learn reading-order priors, then generates endless chats between balloons that never appeared together, making comics speak to itself.
Comics Pages as Audit Devices for Generative AI: A Gutter/Ghost Framework
Proposes the Gutter/Ghost framework: comics pages as testbeds that expose generative AI's failure at sequential coherence, making visible the hidden labour of AI-assisted comic workflows.
Whitewash Cartoonization: Accidental Indexicality into Cartoon Space?
Iterative image-to-image diffusion with lossy brightness feedback transforms images into stylized sequential animations with emergent narrative dynamics, arising from a normalization bug.
Computational Contact Sheets: Synthetic Photography and the Shifting Technical Identity of Comics
Explores synthetic photography via iterative prompting and algorithmic sequencing as a new form of comics—grid-based sequences that reveal machine logics rather than resolve into storylines.
Image GenAI demonstrates the insignificance of writing morphologies outside systematic correspondences to a lexicogrammar
Argues that GenAI's inability to produce writing exposes writing as a systematic correspondence between spatial marks and lexicogrammar—a property GenAI's 'typification engine' cannot replicate.
Comparative analysis of the uses of AI in comics between the two spheres of production
Compares mass-market and experimental uses of generative AI in comics through the lens of seriality theory, examining what problems AI solves for production and at what creative cost.
Metaphorical Comics at the Interface of AI and Mental Health: Toward Patient-Centered Visual Expression
Proposes using LLMs and retrieval-augmented generation to help mental health patients externalize emotional states through metaphorical comics, bridging graphic medicine and generative AI.