Plug-and-Blend: A Framework for Controllable Story Generation with Blended Control Codes

Lin, Zhiyu; Riedl, Mark O.

doi:10.18653/v1/2021.nuse-1.7

Cited by 15 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CREATIVE-WAND instantiations are summarized in Table 1. In both conditions, we use the Plug and Blend generator (Lin and Riedl 2021) applied to GPT-J (Wang and Komatsuzaki 2021) because its sketch inputs provide global control of topics of sentences in a story. We implement two versions of CREATIVE-WAND, one for each condition in our study.…”

Section: Methodsmentioning

confidence: 99%

“…StoryCreativeContext A backend interfacing with an implementation of Plug and Blend (Lin and Riedl 2021) with GPT-J (Wang and Komatsuzaki 2021) as the base language model, supporting both prompts and "sketchbased" high-level control. ExperienceManager SimpleExperienceManager A turn and rule-based agent that shows all available Communications and allow the user to make a choice, or request for activation of Interrupted Communication when there is one.…”

Section: Instantiation Description Creativecontextmentioning

confidence: 99%

See 1 more Smart Citation

Creative Wand: A System to Study Effects of Communications in Co-Creative Settings

Lin¹,

Rohan²,

Riedl³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Recent neural generation systems have demonstrated the potential for procedurally generating game content, images, stories, and more. However, most neural generation algorithms are "uncontrolled" in the sense that the user has little say in creative decisions beyond the initial prompt specification. Cocreative, mixed-initiative systems require user-centric means of influencing the algorithm, especially when users are unlikely to have machine learning expertise. The key to cocreative systems is the ability to communicate ideas and intent from the user to the agent, as well as from the agent to the user. Key questions in co-creative AI include: How can users express their creative intentions? How can creative AI systems communicate their beliefs, explain their moves, or instruct users to act on their behalf? When should creative AI systems take initiative? The answer to such questions and more will enable us to develop better co-creative systems that make humans more capable of expressing their creative intents. We introduce CREATIVE-WAND, a customizable framework for investigating co-creative mixed-initiative generation. CREATIVE-WAND enables plug-and-play injection of generative models and human-agent communication channels into a chat-based interface. It provides a number of dimensions along which an AI generator and humans can communicate during the co-creative process. We illustrate the CREATIVE-WAND framework by using it to study one dimension of co-creative communication-global versus local creative intent specification by the user-in the context of storytelling.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Instantiation Description Creativecontextmentioning

confidence: 99%

Creative Wand: A System to Study Effects of Communications in Co-Creative Settings

Lin¹,

Rohan²,

Riedl³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, updating gradients at the token level results in slow inference. Instead of updating the hidden activations, Krause et al (2020); Yang and Klein (2021); Lin and Riedl (2021) introduce generative discriminators to re-weight the next token distributions on the fly during inference, thus improving the inference speed.…”

Section: Related Workmentioning

confidence: 99%

Controllable Natural Language Generation with Contrastive Prefixes

Jiang¹,

Liu²,

Shen³

et al. 2022

Findings of the Association for Computational Linguistics: ACL 2022

View full text Add to dashboard Cite

To guide the generation of large pretrained language models (LM), previous work has focused on directly fine-tuning the language model or utilizing an attribute discriminator. In this work, we propose a novel lightweight framework for controllable GPT2 (Radford et al., 2019) generation, which utilizes a set of small attribute-specific vectors, called prefixes (Li and Liang, 2021), to steer natural language generation. Different from Li and Liang (2021), where each prefix is trained independently, we take the relationship among prefixes into consideration and train multiple prefixes simultaneously, as illustrated in Figure 1. We propose a novel supervised method and also an unsupervised method to train the prefixes for single-aspect control while the combination of these two methods can achieve multi-aspect control. Experimental results on both singleaspect and multi-aspect control show that our methods can guide generation towards the desired attributes while keeping high linguistic quality.

show abstract

“…However, the natural language processing community [55,81,82] strives to gain more control over how the text is generated, and we see this as a promising next step. In particular, a limitation of the current system is that it does not incorporate any theory-based message design principles [1,83], such as barriers, cues to action, and norm or threat appeals.…”

Section: Principal Findingsmentioning

confidence: 99%

Harnessing Artificial Intelligence for Health Message Generation: The Folic Acid Message Engine

Schmälzle¹,

Wilcox²

2022

J Med Internet Res

View full text Add to dashboard Cite

Background Communication campaigns using social media can raise public awareness; however, they are difficult to sustain. A barrier is the need to generate and constantly post novel but on-topic messages, which creates a resource-intensive bottleneck. Objective In this study, we aim to harness the latest advances in artificial intelligence (AI) to build a pilot system that can generate many candidate messages, which could be used for a campaign to suggest novel, on-topic candidate messages. The issue of folic acid, a B-vitamin that helps prevent major birth defects, serves as an example; however, the system can work with other issues that could benefit from higher levels of public awareness. Methods We used the Generative Pretrained Transformer-2 architecture, a machine learning model trained on a large natural language corpus, and fine-tuned it using a data set of autodownloaded tweets about #folicacid. The fine-tuned model was then used as a message engine, that is, to create new messages about this topic. We conducted a web-based study to gauge how human raters evaluate AI-generated tweet messages compared with original, human-crafted messages. Results We found that the Folic Acid Message Engine can easily create several hundreds of new messages that appear natural to humans. Web-based raters evaluated the clarity and quality of a human-curated sample of AI-generated messages as on par with human-generated ones. Overall, these results showed that it is feasible to use such a message engine to suggest messages for web-based campaigns that focus on promoting awareness. Conclusions The message engine can serve as a starting point for more sophisticated AI-guided message creation systems for health communication. Beyond the practical potential of such systems for campaigns in the age of social media, they also hold great scientific potential for the quantitative analysis of message characteristics that promote successful communication. We discuss future developments and obvious ethical challenges that need to be addressed as AI technologies for health persuasion enter the stage.

show abstract

Plug-and-Blend: A Framework for Controllable Story Generation with Blended Control Codes

Cited by 15 publications

References 15 publications

Creative Wand: A System to Study Effects of Communications in Co-Creative Settings

Creative Wand: A System to Study Effects of Communications in Co-Creative Settings

Controllable Natural Language Generation with Contrastive Prefixes

Harnessing Artificial Intelligence for Health Message Generation: The Folic Acid Message Engine

Contact Info

Product

Resources

About