Narrative Generation Research Articles

Image of the Other by Al-Muhassin Al-Tanukhi

The research investigates studying the image of the other in the books of Al-Muhassin Al-Tanukhi. The research focused on defining the other and the beginnings of the formation of his image in Arab-Islamic culture, in the Qur'anic text, and in the writings of travelers and geographers. It then explains the image of the other in Al-Muhassin Al-Tanukhi's writings. The research concluded that his image of the other is not stereotypical according to a preconceived view because he acts and recognizes the difference between the other. The images presented in the study varied across three main perspectives. First, there was a portrayal of admiration for the other, which did not culminate in a sense of amazement due to perceived cultural superiority. Second, a positive image of the other was offered. Lastly, disapproval was expressed stemming from a lack of comprehension. Furthermore, the research examined the methodology of narrative presentation, categorizing it into two distinct structures. The first structure is characterized by its simplicity, relying solely on a linear narration of events or stories. In contrast, the second structure is more complex, involving what is referred to as the phenomenon of narrative generation. This approach incorporates intricate elements such as characters and situations. The research investigates studying the image of the other in the books of Al-Muhassin Al-Tanukhi. The research focused on defining the other and the beginnings of the formation of his image in Arab-Islamic culture, in the Qur'anic text, and in the writings of travelers and geographers. It then explains the image of the other in Al-Muhassin Al-Tanukhi's writings. The research concluded that his image of the other is not stereotypical according to a preconceived view because he acts and recognizes the difference between the other. The images presented in the study varied across three main perspectives. First, there was a portrayal of admiration for the other, which did not culminate in a sense of amazement due to perceived cultural superiority. Second, a positive image of the other was offered. Lastly, disapproval was expressed stemming from a lack of comprehension. Furthermore, the research examined the methodology of narrative presentation, categorizing it into two distinct structures. The first structure is characterized by its simplicity, relying solely on a linear narration of events or stories. In contrast, the second structure is more complex, involving what is referred to as the phenomenon of narrative generation. This approach incorporates intricate elements such as characters and situations.

Report on the 7th International Workshop on Narrative Extraction from Texts (Text2Story 2024) at ECIR 2024

The Seventh International Workshop on Narrative Extraction from Texts (Text2Story'24) was held on March 24 th , 2024, in conjunction with the 46 th European Conference on Information Retrieval (ECIR 2024) in Glasgow, Scotland. Over the day, more than 50 attendees engaged in discussions and presentations focused on recent advancements in narrative representation, extraction, and generation. The workshop featured two invited keynote addresses, fourteen research paper presentations, and a poster session. The workshop proceedings are available online. 1 Date : 24 March 2024. Website : https://text2story24.inesctec.pt/.

Intérêt d’une intervention de soutien langagier basée sur la narration et menée par l’enseignante auprès d’enfants de grande section de maternelle avec difficultés langagières.

Context: Narrative intervention enhances narrative abilities by addressing macrostructural elements within storytelling tasks for instance and can be an effective way of supporting the language development of preschool-aged children. Moreover, language support in educational settings plays a preventive role in addressing potential difficulties in language skills due to a lack of linguistic stimulation. Objectives: This study assesses the implementation fidelity of a tier 2 intervention with a pre-school teacher, measures the effectiveness of this intervention with children with language difficulties, and determines whether these children can catch up the level of narrative skills of the rest of the class, following the intervention Methods: Children with language difficulties were selected by the master student based on an inclusion test determining their linguistic profile, which was administered to all pupils in the class. Narrative skills were assessed for the entire class in a pre-test, considering elements of macrostructure and microstructure elicited during narrative generation and retelling activities. A Rapid Automa-tized Naming (RAN) control task was also included to evaluate the specificity of the intervention. The teacher was trained by the master student for the intervention, according to profession-al development principles. Subsequently, the intervention, lasting 45 minutes once a week for 6 weeks, was delivered by the teacher to five children identified as having language difficulties, aged 5;5 to 5;10. The intervention aimed to improve the number of macrostructural elements in a retelling activity through the use of icons and sequential picture stories. Finally, the RAN task and narrative skills (retelling and narrative generation) of the entire class were reassessed in a post-test. Results: At the end of the intervention, macro- and microstructural improvements were observed in the story generation activity in the experimental group. Moreover, following the teacher's intervention, the experimental group did not differ anymore from the rest of the class in retelling (when considering the trained story) nor in story generation. However, data analysis revealed weaknesses in the implementation of the intervention, both in content and dosage. Conclusion: The results contribute to the existing research supporting the effectiveness of group narrative intervention. Furthermore, they suggest the possibility of replicating a tier 2 indirect narrative intervention conducted by the teacher, by enhancing the professional development framework.

StoryForge: AI-Powered Narrative with Dynamic Imagery and Voice

StoryForge is an AI-powered narrative generation system that utilizes a combination of OpenAI’s GPT-3.5 Turbo model [1], Hugging Face’s Diffusion-based text-to-image gen- erative model Stable Diffusion XL (SDXL) [4], and Hugging Face’s Text-to-Speech Bark model [10] to create dynamic and engaging stories tailored to user-specified constraints. Users can input parameters such as genre, number of characters, character details, key events, mood, tone, point of view, and reader age to guide the narrative generation process. The system leverages the power of GPT-3.5 Turbo to generate the story text [1], SDXL to generate images for key events [4], and Bark to synthesize speech for the narrative [10], providing a comprehensive and immersive storytelling experience. A Streamlit-based user interface enables seamless interaction with the system, allowing users to input constraints, view generated text and images, and listen to the synthesized audio. It transforms user inputs into a multisen- sory narrative experience, marking a significant stride in the intersection of artificial intelligence and imaginative storytelling. StoryForge demonstrates the potential of AI to revolutionize storytelling, empowering users to create personalized and cap- tivating narratives with ease and embark on a journey where words, images, and voices converge to create truly one-of-a-kind tales. Index Terms—StoryForge, GPT-3.5 Turbo, Bark, Stable Dif- fusion XL, Streamlit, Hugging Face, Generative Model, Story- telling, Multisensory

An Exploration of the Impact of Quality Illustrations in Children’s Picture Books on Preschool Student Narrative Ability

Narrative language development is one of the early literacy skills necessary for proficient reading and writing achievement. The correlation between narrative skills and later writing ability indicates that more advanced and complex narrative generation and recall is linked to later writing output. In this experimental study, 47 preschool students participated in an intervention focusing on the impact of high-quality illustrations on narrative ability. Results show a significant effect of intervention participation on narrative language measures in language complexity, narrative retelling, story comprehension, vocabulary usage, and episodic components. These findings suggest that attention to artistic devices used in illustrations paired with picture book study can increase student narrative ability

It Takes a Fool to Remain Sane: How and When HR Executives Use Jesting Techniques to Trickle Up Paradoxical Tensions

Human resource (HR) experts are continuously confronted with paradoxical tensions, which they need to navigate to benefit the HR function and the organization. Because the members of the top management team (TMT) are important stakeholders for the HR function, the HR executives’ effective navigation involves visibilizing paradoxes and trickling them up during interactions with the TMT. However, how HR executives (as low-powered organizational actors compared to their peers) visibilize tensions for TMT members is less understood from either the paradox literature or the HR scholarship perspectives. In this interview study, we use a court jester prompt as a narrative generator to explore HR executives’ reflections on using techniques to visibilize tensions within the overarching paradox of social and business interests. Our findings enrich the literature on paradox salience and provide examples of what we term jesting techniques on cognitive (e.g., exaggerating,), emotional (e.g., expressing feelings), and behavioral (e.g, reordering) levels.

Does Robin Hood Use a Lightsaber?: Automated Planning for Storytelling

Humans have been using stories to entertain, educate, and persuade audiences for centuries. The advent of modern AI tools in the form of Large Language Models (LLMs) such as chatGPT continues to fulfill this purpose. However while recent work has shown that LLMs can successfully be used for narrative generation, they lack coherence and can be prone to repetition and stilted language. Automated Planning can therefore be combined with Natural Language text generation to create narratives (stories) that are logical, coherent, and believable. A planning model provides scaffolding to an LLM so that the LLM's language generation is context-dependent, in order to allow users to create more coherent, logical, and believable stories in a variety of domains.

The use of differing verb types in the oral narratives of school-age children

ABSTRACT The aim of this project was to examine age-related use of action, metacognitive, and metalinguistic verbs because factors related to verb transparency are thought to impact word learning. Performance on fictional, oral narratives elicited using a single-episode picture was evaluated for 84 children with typical language (TL) and 38 age-matched children with a Developmental Language Disorder (DLD) ranging in age from 5;1 to 14;4 years of age. Narrative samples were transcribed and coded for occurrences of action verbs (AV), metacognitive verbs (MCV), and metalinguistic verbs (MLV). The total number of verbs used and the number of different verbs produced in each category were examined across ages, as was the difference in verb usage patterns across language ability groups. A statistically significant increase in the overall use of action, MCV, and MLV with age was observed. TL children used a wider variety of each verb type as compared to those with DLD. Age-related progression for verb use was observed in the fictional narrative generation task by both groups of children. However, despite similar rates of total verb use, children with DLD produced fewer different verbs in the three categories. These findings indicate the total number of verbs used is similar in children with TL and DLD, but there is a difference in the number of different verbs used in functional tasks such as narratives for children with DLD. These results also indicate a need for ongoing examination of the factors not only impacting verb acquisition but also use in communication tasks.

Storytelling systems: Progress towards generation informed by models of the reader

The present paper attempts to identify specific points of contact between efforts to generate narrative in the field of artificial intelligence and the field of narratology. Such points of contact centre around the tasks of modelling processes related to the construction of narrative in computational terms. A number of existing research efforts to develop systems related to storytelling are reviewed. An analysis of the functionalities that they cover suggests that storytelling might be profitably studied in terms of at least four different sub-tasks: Fabula generation, narrative discourse composition, text generation, and narrative interpretation. An analysis from a computational point of view of the context in which narrative is used as a means of communication between author and reader is presented. This shows that parts of the cognitive processes involved are yet to be modelled, which suggests that the set of sub-tasks identified may continue to grow with further research.

Unveiling the War and Constructing Identities: Exploring Memes in Ukrainian and Russian Social Media during the Russian Invasion of Ukraine

The article examines the generation and deployment of visual narratives in Ukrainian and Russian digital participatory cultures, with a specific focus on internet memes in the context of the Russian invasion of Ukraine. It analyzes the form, content, and functions of these memes and highlights their similarity in mobilizing and conveying political messages despite variations in their visual components. The study indicates that Ukrainian memes are used not only to promote political agendas but also serve as trauma coping and collective identity construction mechanisms in times of crisis, helping to promote new war narratives that are engaged in the construction of the self and the other.

Design requirements for a digital storytelling application for people with mild cognitive impairment (MCI).

The current digital storytelling applications present advantages for individuals with Mild Cognitive Impairment (MCI); however, there exists a notable oversight regarding their potential to facilitate group-based storytelling activities with this population. This study endeavors to identify design requirements for a more inclusive and accessible digital storytelling tool for people with MCI. The methodological framework encompasses distinct stages, commencing with focus groups and interviews (Stage 1), followed by prototyping workshops (Stage 2) and qualitative prototype testing (Stage 3). The comprehensive three-stage research involved participants residing in Beijing, China, including 43 people with MCI aged 65-95 years (M = 79.09, SD = 8.99), with a mean Montreal Cognitive Assessment score of 21.91 (range = 18-26, SD = 2.40). Additionally, 17 care partners and 10 occupational or clinical therapists actively participated. The culmination of the three-stage research process has yielded 12 discernible key design requirements. Preferred storytelling themes center around narratives designed to elicit positive emotions. The narrative material generation process involves a systematic approach, unlocking memories through carefully formulated questions. In memory retrieval, users are provided with hints, bolstering confidence and perpetuating a semblance of face-to-face interaction. The focus in story sharing lies in transcending mere narration and extending it to a wider audience. This case study centers on crafting a digital storytelling application to enhance social connections for people with MCI. It delves into crucial design requirements addressing memory challenges, emphasizing individual preparation and group sharing. The developed digital storytelling application demonstrates potential to offer valuable memory support and foster personal and collective connections. Future research will focus on formal testing to evaluate these outcomes.

The Impact of AI-Driven Narrative Generation, Exemplified by ChatGPT, on the Preservation of Human Creative Originality and Uniqueness

The appearance of Artificial Intelligence (AI) systems have become more adept at generating original texts, becoming the realm of storytelling, as exemplified by ChatGPT and similar language models. This has sparked a growing debate about its impact on the originality and uniqueness of human creativity and thinking. Bringing up the question of whether AI-driven narrative generation is a benefit or a detriment to humanity, this paper delves into the evolution of AI storytelling, examines the capabilities and limitations of large language models like ChatGPT, even under their impressive fluency, and further explores the implications for human creativity and intellectual diversity. While AI storytelling tools have undoubtedly revolutionised content generation, this paper argues that they do not inherently erode human originality and uniqueness and more so cannot authentically replicate the distinctiveness of human thinking. Instead, with conscientious implementation, AI writing technologies may serve as powerful complementary tools that can enhance human creativity, expand the diversity of voices and perspectives, provide more widespread access to the means of storytelling and personal expression, and optimise our human literary activities in a manner that elevates rather than erodes the uniqueness of human literary work when used thoughtfully and ethically.

A Structured Narrative Prompt for Prompting Narratives from Large Language Models: Sentiment Assessment of ChatGPT-Generated Narratives and Real Tweets

Large language models (LLMs) excel in providing natural language responses that sound authoritative, reflect knowledge of the context area, and can present from a range of varied perspectives. Agent-based models and simulations consist of simulated agents that interact within a simulated environment to explore societal, social, and ethical, among other, problems. Simulated agents generate large volumes of data and discerning useful and relevant content is an onerous task. LLMs can help in communicating agents’ perspectives on key life events by providing natural language narratives. However, these narratives should be factual, transparent, and reproducible. Therefore, we present a structured narrative prompt for sending queries to LLMs, we experiment with the narrative generation process using OpenAI’s ChatGPT, and we assess statistically significant differences across 11 Positive and Negative Affect Schedule (PANAS) sentiment levels between the generated narratives and real tweets using chi-squared tests and Fisher’s exact tests. The narrative prompt structure effectively yields narratives with the desired components from ChatGPT. In four out of forty-four categories, ChatGPT generated narratives which have sentiment scores that were not discernibly different, in terms of statistical significance (alpha level α=0.05), from the sentiment expressed in real tweets. Three outcomes are provided: (1) a list of benefits and challenges for LLMs in narrative generation; (2) a structured prompt for requesting narratives of an LLM chatbot based on simulated agents’ information; (3) an assessment of statistical significance in the sentiment prevalence of the generated narratives compared to real tweets. This indicates significant promise in the utilization of LLMs for helping to connect a simulated agent’s experiences with real people.

Impact of Discourse Type and Elicitation Task on Language Sampling Outcomes.

Language sampling is a critical component of language assessments. However, there are many ways to elicit language samples that likely impact the results. The purpose of this study was to examine how different discourse types and elicitation tasks affect various language sampling outcomes. A diverse group of K-3 students (N = 1,037) contributed eight spoken language samples in four elicitation conditions: (a) expository generation, (b) expository retell, (c) narrative generation, and (d) narrative retell. Samples were audio-recorded, transcribed, and coded for number of total words, number of different words, mean length of utterance in words (MLUW), and number of clauses per communication unit (i.e., Subordination Index [SI]). Narrative retell and expository generation conditions yielded the largest samples with the greatest lexical diversity when compared to narrative generation and expository retell. MLUW was higher in expository conditions, but mean SI was higher in narrative conditions. For both measures of syntax, narrative retell and expository generation yielded the highest mean scores. For each outcome, there were expected increases corresponding to grades; however, the differences faded between second and third grade. As a component of language assessments, clinicians' selection of language sampling procedures will impact the sample length, lexical diversity, utterance length, and syntactical complexity of the samples. https://doi.org/10.23641/asha.24185649.

Anti-racism praxis in an emerging community-academic approach to environmental and climate justice

Abstract Background Racial injustices, environmental and climate threats, and COVID-19 inequities that disproportionately affect communities of color have contributed to social reckoning with racism. Public health is grappling with how to strengthen collaborative processes, research, and practice to be grounded in an understanding of racism as a public health issue and incorporate a structural racism lens into strategies to promote health equity. We describe a process for strengthening anti-racism praxis for an emerging academic center focused on community-academic collaborative approaches to advancing participatory action research for environmental and climate justice. Methods University-based faculty and staff and popular education designers and facilitators collaborated to design an anti-racism process for an emerging environmental and climate justice academic research center. The anti-racism process focused on building leadership and action through deep discussion, reflection, narrative generation, and visioning. This process involved clarifying the Center's purpose, anticipated short- and long-term outcomes, collaborative processes, and theories of change. Results The anti-racism process focused on building deep connections and capacity to practice and advance racial justice within the administrative, research, teaching, and practice spheres of academic and community participants identified as active and passive allies on the Spectrum of Allyship. Dialogues illuminated the centrality of storytelling for listening to and learning from a range of lived experiences and envisioning processes and ways of working together. We discuss themes, facilitating factors, and challenges that emerged from our dialogic process. Conclusions This case highlights the importance of anti-racism processes as a collective experience to inform racial justice approaches within emerging and sustained public health units - large and small - and in participants’ spheres of influence. Key messages • As the field of public health grapples with racism as a driver of health inequities, anti-racism approaches are critical to public health practice and research. • Anti-racism approaches that engage narratives have potential to inform guiding approaches for public health units committed to health equity.

SceneCraft: Automating Interactive Narrative Scene Generation in Digital Games with Large Language Models

Creating engaging interactive story-based experiences dynamically responding to individual player choices poses significant challenges for narrative-centered games. Recent advances in pre-trained large language models (LLMs) have the potential to revolutionize procedural content generation for narrative-centered games. Historically, interactive narrative generation has specified pivotal events in the storyline, often utilizing planning-based approaches toward achieving narrative coherence and maintaining the story arc. However, manual authorship is typically used to create detail and variety in non-player character (NPC) interaction to specify and instantiate plot events. This paper proposes SCENECRAFT, a narrative scene generation framework that automates NPC interaction crucial to unfolding plot events. SCENECRAFT interprets natural language instructions about scene objectives, NPC traits, location, and narrative variations. It then employs large language models to generate game scenes aligned with authorial intent. It generates branching conversation paths that adapt to player choices while adhering to the author’s interaction goals. LLMs generate interaction scripts, semantically extract character emotions and gestures to align with the script, and convert dialogues into a game scripting language. The generated script can then be played utilizing an existing narrative-centered game framework. Through empirical evaluation using automated and human assessments, we demonstrate SCENECRAFT’s effectiveness in creating narrative experiences based on creativity, adaptability, and alignment with intended author instructions.

Evolving Interactive Narrative Worlds

An interactive narrative is bound by the context of the world where its story takes place. However, most work in interactive narrative generation takes its story world design and mechanics as given, which abdicates a large part of story generation to an external world designer. In this paper, we close the story world design gap with an evolutionary search framework for generating interactive narrative worlds and mechanics. Our framework finds story world designs that accommodate multiple distinct player roles. We evaluate our system with an action agreement ratio analysis that shows worlds generated by our framework provide a greater number of in-role action opportunities compared to story worlds randomly sampled from the generative space.

“The Truthfulness Lies in the Process, Not the Outcome” Using Artistic Practices to Further Truth-Telling and Memorialization in the Philippines

The Philippines have never known a unified state-sanctioned narrative about the violence that happened during the Marcos dictatorship. In order to resist ongoing disinformation campaigns that seek to erase evidence and memories of past violence, various institutional and civil society actors are currently initiating interventions in the domain of truth and memorialization. Notably, artists, curators, and creative professionals are engaging in various kinds of so-called ‘narrative documentation’ and ‘narrative change-making’. Several of these initiatives mobilize spatial dynamics and co-created processes to facilitate more complex forms of truth-telling and memorialization, which foreground complexity and ambiguity, and which prompt more engaged forms of truth-listening. This article zooms in on a specific project that mobilizes traditional artisan and artistic techniques and forms to revisit women’s experiences of historical and ongoing violence by crafting layered and ambiguous narratives about harm. In doing so, the Weaving Women’s Words on Wounds of War project seeks to further memorialization, truth-telling, and truth-listening about gendered violence. Through an analysis embedded in scholarship on memory, truth, and artistic practice, I argue that it is the generation of ambiguous and complex narratives that invites an active and relational type of engagement and listening. This holds potential for resisting the erasure of complex forms of violence, both in the context of the Philippines, as well as in other contexts where truth or memorialization initiatives may be incapable of capturing the gravity of lived experiences of violence or of facilitating genuine listening.

Pattern-Based Target Selection for Treatment of Irregular Past Tense: A Single-Case Experimental Design for Children With Developmental Language Disorder.

Many children with developmental language disorder (DLD) demonstrate difficulty inflecting irregular past tense verbs. We evaluated using phonological-based patterns (i.e., schemas) to select targets for treatment of irregular verb inflection for children with DLD. We hypothesized this approach would result in acquisition of treated verbs, generalization to untreated verbs within the same phonological schema, and increased use of correct irregular verbs in naturalistic contexts. Treatment was provided to three 7-year-old participants with DLD in the context of multiple baseline designs across behaviors and participants. Phonological schemas included vowel change, final alveolar, and dual change irregular verbs. Treatment was provided using established therapeutic approaches, including narrative generation, sentence imitation, and naturalistic play activities. Acquisition of treated and untreated targets was assessed at the beginning of sessions, and use of verbs in naturalistic contexts was observed throughout treatment sessions. Positive acquisition effects were noted for two of three participants. Generalization to untreated items occurred within and across treatment sets for two participants, whereas increases in accurate irregular verb production in naturalistic contexts were not observed. Outcomes demonstrated support for a pattern-based approach to target selection for treatment of irregular past tense verbs. One participant, who demonstrated general difficulty with the probe and treatment tasks, showed no treatment effect, suggesting that the treatment may be effective for some children with DLD. Further research is warranted to determine what factors might have influenced these variable outcomes across participants. https://doi.org/10.23641/asha.23671500.

Shinrin yoku as a pedagogy for peace amidst violence: generating dynamic narratives of Palestine-Israel relations on college campuses

ABSTRACT Amidst violent conflict over Palestine-Israel relations at colleges across America, how might we use our classrooms and campus landscapes to generate dynamic narratives that facilitate peace? Moving beyond a chronological ordering of events, a narrative is a constructed cohesive account of occurrences used to make sense of experiences and motivate action. In violent settings, narratives tend to retrench into static accounts that increase prejudice and motivate greater acts of violence. Alternatively, dynamic narratives offer complex judgement, plot, character, and value assessments of the world thus encouraging more openness to others and peace. I propose a novel intervention for the generation of dynamic narratives. I use the practice of shinrin yoku or guided forest walks in a seminar about Palestine and Israel, to invite liminality, the experience of communal spaces where traditional markers of power and social obligations are stripped. I expected that increasing experiences of shinrin yoku, and in turn liminality, will induce dynamic understandings of Palestinian Israeli relations on campus. Digital diary responses from eleven student participants kept over twelve weeks in a Fall 2022 seminar reveal that even with the eruption of hostilities, 1 students adopted dynamic stories about Palestine and Israel relations when they spent increasing time engaged in shinrin yoku.