Innocent Panels, Hateful Stories: Evaluating and Detecting Hateful Intent in Multi-Turn Visual Story Generation

Platformed Publics: Algorithms, Affordances, and the Recomposition of Authority

Investigating Social Bias in Narrative Image Generation

Text-to-image (T2I) generation models are increasingly embedded in applications such as media content creation and education, raising concerns about how their outputs may reproduce social biases. Prior work has shown that T2I models exhibit social biases, yet existing evaluations largely focus on a photo generation task. As a result, it remains unclear whether and how such biases manifest in more narrative visual formats, such as storyboards and comics, where characters and events are presented across multiple panels. In this work, we compare bias expression across photo, storyboard, and comic generation in six T2I models by adapting BBG, a text-based bias evaluation framework, to image generation. Our results show that proprietary models generate 25.9% biased outputs in photo generation on average, with biased outputs increasing by 9.6pp in storyboard generation and 18.2pp in comic generation. We also find that photos mainly encode biases through subtle visual cues, while storyboards and comics reveal them more explicitly through event sequencing, character positioning, narrative resolution, and textual elements. These findings show that biases that remain less visible in photo generation may surface in narrative visual formats, highlighting the importance of evaluating T2I systems with diverse visual formats beyond photo generation.

Computational Humor with Multimodal LLMs: Methods, Datasets, Evaluation, and Challenges

Multimodal humor in memes, cartoons, and comics remains difficult for AI systems because intended meaning depends on non-literal mechanisms, shared cultural knowledge, and communicative intent rather than literal scene description. This survey focuses on visual humor understanding in single-image and multi-panel artifacts, while treating humor generation as an emerging downstream frontier. We position the literature against prior humor, sarcasm, and general MLLM surveys and organize it using a capability-centric hierarchy spanning recognition, interpretation and reasoning, and generation. Under this lens, we synthesize benchmark design, evaluation protocols, and modeling paradigms, tracing the field's shift from task-specific fusion models to large-model approaches based on multimodal alignment, evidence-grounded reasoning, and controlled generation. We conclude by highlighting the main barriers to progress: shortcut-prone evaluation, limited cultural and narrative coverage, weak evidence grounding, and unresolved safety and ownership concerns.

From Screen Narratives to Prompt-Conditioned AI Myths: An Exploratory Structuralist Comparison

The study contributes a cautious comparative framework for distinguishing prompt-induced structural conformity from analytically emergent features in AI-generated narrative.

Evaluating Multimodal Narrative Understanding of Popular Hollywood Films

It is found that many vision-language models struggle on this task (with many performing at near-chance levels of accuracy), while audio-visual models (including those that use audio in captioning scenes) reach a maximum accuracy of 61.1%, well below human-level performance.

Narrative Transportation as the Mechanism behind the Persuasive Power of Stories: From Parables to Audiovisual Media

Stories persuade audiences whose explicit beliefs diverge, and the question of why this happens motivates a long line of work on narrative transportation. This article asks how the transportation mechanism operates when a story withholds a verdict and leaves its central conflict open, and it does so through a structured content analysis of one director’s catalog of short films produced between 2018 and 2026. The corpus comprises more than 15 short films that have collectively won 31 awards and been selected for 84 official festivals across five continents. Each film was coded for narrative features such as open endings, character-first framing, inclusive casting, and formal constraint, and these features were cross-tabulated with the framing of the juries that conferred recognition. The analysis shows that films built around an unresolved question secured awards from juries with opposing vocabularies, including explicitly religious festivals and secular genre festivals. These results indicate that a reduction in overt persuasive intent is associated with broader cross-audience recognition, which supports the prediction that transportation suppresses counterarguing while sustaining emotional and imagistic engagement. The single-author design and the use of festival outcomes as an indirect index of transportation constrain the strength of the causal claim, and the discussion frames these limits as a guide for the direct audience studies that should follow. The findings have practical implications for independent filmmakers who want their work to cross audience boundaries at feature length.