In text generation, generating long stories is still a challenge. Coherence tends to decrease rapidly as the output length increases. Especially for generated stories, coherence of the narrative is an important quality aspect of the output text. In this paper we examine how narrative coherence is attained in the submissions of NaNoGenMo 2018, an online text generation event where participants are challenged to generate a 50,000 word novel. We list the main approaches that were used to generate coherent narratives and link them to scientific literature. Finally, we give recommendations on when to use which approach.

GPT-2, a neural language model trained on a large dataset of English web text, has been used in a variety of natural language generation tasks because of the language quality and coherence of its outputs. In order to investigate the usability of GPT-2 for text generation for video games, we fine-tuned GPT-2 on a corpus of video game quests and used this model to generate dialogue lines for questgiver NPCs in a role-playing game. We show that the model learned the structure of quests and NPC dialogue, and investigate how the temperature parameter influences the language quality and creativity of the output artifacts. We evaluated our approach with a crowdsource experiment in which human judges were asked to rate hand-written and generated quest texts on language quality, coherence and creativity. CCS CONCEPTS• Applied computing → Computer games; • Computing methodologies → Natural language generation.

We describe a prototype of Churnalist, a headline generator for creating contextually-appropriate fictional headlines that can be used as flavor text in games. Churnalist creates new headlines by remixing existing headlines. It extracts seed words from free text input, searches for related words in a dataset of word embeddings and uses these words in the new headlines. The system requires no linguistic expertise or handcoded language models from the user.

No abstract

We propose using real-world datasets to generate textual game assets for serious games. As an example, we used a dataset of P2000 crisis event messages to generate descriptive texts that can be transformed into new game assets by game writers, thereby reducing the writing effort required during the development phase of an adaptive serious game. In this paper we describe this first attempt and we discuss the challenges and possibilities of using open data for textual asset generation.