Wordplay: When Language Meets Games

$$\ $$\ $$\ $$\ $$ | $\ $$ | $$ | $$ | $$ |$$$\ $$ | $$$$$$\ $$$$$$\ $$$$$$$ | $$$$$$\ $$ | $$$$$$\ $$\ $$\ $$\ $$ $$ $$\$$ |$$ __$$\ $$ __$$\ $$ __$$ |$$ __$$\ $$ | \____$$\ $$ | $$ |\__| $$$$ _$$$$ |$$ / $$ |$$ | \__|$$ / $$ |$$ / $$ |$$ | $$$$$$$ |$$ | $$ | $$$ / \$$$ |$$ | $$ |$$ | $$ | $$ |$$ | $$ |$$ |$$ __$$ |$$ | $$ |$$\ $$ / \$$ |\$$$$$$ |$$ | \$$$$$$$ |$$$$$$$ |$$ |\$$$$$$$ |\$$$$$$$ |\__| \__/ \__| \______/ \__| \_______|$$ ____/ \__| \_______| \____$$ | $$ | $$\ $$ | $$\ $$\ $$\ $$\ $$ | \$$$$$$ | $$ | $\ $$ |$$ | $$ | \__| \______/ $$ |$$$\ $$ |$$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ $$$$$$\ $$\ $$\ $$$$$$\ $$$$$$\ $$$$$$\ $$ $$ $$\$$ |$$ __$$\ $$ __$$\ $$ __$$\ $$ | \____$$\ $$ __$$\ $$ __$$\ $$ | $$ | \____$$\ $$ __$$\ $$ __$$\ $$$$ _$$$$ |$$ | $$ |$$$$$$$$ |$$ | $$ | $$ | $$$$$$$ |$$ | $$ |$$ / $$ |$$ | $$ | $$$$$$$ |$$ / $$ |$$$$$$$$ | $$$ / \$$$ |$$ | $$ |$$ ____|$$ | $$ | $$ | $$ __$$ |$$ | $$ |$$ | $$ |$$ | $$ |$$ __$$ |$$ | $$ |$$ ____| $$ / \$$ |$$ | $$ |\$$$$$$$\ $$ | $$ | $$$$$$$$\ \$$$$$$$ |$$ | $$ |\$$$$$$$ |\$$$$$$ |\$$$$$$$ |\$$$$$$$ |\$$$$$$$\ \__/ \__|\__| \__| \_______|\__| \__| \________| \_______|\__| \__| \____$$ | \______/ \_______| \____$$ | \_______| $$\ $$ | $$\ $$ | $$\ $$\ $$\ $$$$$$\ \$$$$$$ | \$$$$$$ | $$$\ $$$ | $$ | $$ __$$\ \______/ \______/ $$$$\ $$$$ | $$$$$$\ $$$$$$\ $$$$$$\ $$$$$$$\ $$ / \__| $$$$$$\ $$$$$$\$$$$\ $$$$$$\ $$$$$$$\ $$\$$\$$ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____| $$ |$$$$\ \____$$\ $$ _$$ _$$\ $$ __$$\ $$ _____| $$ \$$$ $$ |$$$$$$$$ |$$$$$$$$ | $$ | \$$$$$$\ $$ |\_$$ | $$$$$$$ |$$ / $$ / $$ |$$$$$$$$ |\$$$$$$\ $$ |\$ /$$ |$$ ____|$$ ____| $$ |$$\ \____$$\ $$ | $$ |$$ __$$ |$$ | $$ | $$ |$$ ____| \____$$\ $$ | \_/ $$ |\$$$$$$$\ \$$$$$$$\ \$$$$ |$$$$$$$ | \$$$$$$ |\$$$$$$$ |$$ | $$ | $$ |\$$$$$$$\ $$$$$$$ | \__| \__| \_______| \_______| \____/ \_______/ \______/ \_______|\__| \__| \__| \_______|\_______/


Back to the future ›››



Wordplay: When Language Meets Games @ ACL 2024.

Date and time: August 16, 2024

Location: Bangkok, Thailand

Virtual links: TBD

One line version: Your one stop shop for all things interactive narrative + AI!

The slightly longer version

This workshop will focus on exploring the utility of interactive narratives, think everything from classic text-adventures like Zork to modern Twine games, to fill a role as the learning environments of choice for language-based tasks including but not limited to storytelling. A few previous iterations of this workshop took place very successfully with hundreds of attendees, at NeurIPS 2018, NeurIPS 2020, & NAACL 2022. Since then, the community of people working in this area has rapidly increased. This workshop aims to be a centralized place where all researchers involved across a breadth of fields can interact and learn from each other. Furthermore, it will act as a showcase to the wider NLP/RL/Game communities on interactive narrative’s place as a learning environment. The program will feature a collection of invited talks in addition to contributed talks and posters from each of these sections of the interactive narrative community and the wider NLP and RL communities.

We like all things:

The actual version

Interaction is a core component of learning. Humans learn various skills — including language, vision, and motor skills — more effectively through interactive media [Feldman and Narayanan, 2004; Barsalou, 2008; Tamari et al., 2020]. In the realm of machine learning, interactive environments have served as cornerstones in the quest to develop more robust algorithms for learning agents across many ML sub-communities. Environments like the Atari Learning Environment (ALE) [Bellemare et al., 2013] and Malmo [Johnson et al., 2016] have enabled the development of game-playing agents that perform complex tasks based on raw video inputs, and more recently, THOR [Kolve et al., 2017] and other embodied environments have extended that development to agents embodied in simulated 3D worlds. However, relatively few such environments ground observations or actions in language.

Recent work has shown that interactive narrative — a setting at the intersection of natural language processing and generation, storytelling, and sequential decision making—provides an opportunity to develop situated language learning agents that span these fields [Narasimhan et al., 2015; Côté et al., 2018; Fulda et al., 2017; Zahavy et al., 2018; Hausknecht et al., 2020; Wang et al., 2022]. Interactive narratives have likewise been used to probe large language models (LLM) capabilities to evaluate progress [Li et al., 2021; Tamari et al., 2022; Bubeck et al., 2023; Park et al., 2023]. We use ‘interactive narratives’ to refer to scenarios in which a narrative unfolds sequentially, driven by an agent’s interactions. These interactions may be fully language based — the agent “perceives” and “speaks” in the simulation using only text — or multimodal. Examples include but are not limited to text-adventure games like Zork [Anderson et al., 1979], improvisational storytelling [Martin et al., 2016, 2017; Mathewson and Mirowski, 2017], situated dialogue [Urbanek et al., 2019], tabletop roleplaying games like Dungeons and Dragons [Martin et al., 2018; Callison-Burch et al., 2022; Zhu et al., 2023], and interactive question answering in these worlds [Yuan et al., 2019]. To infer the context and objectives of these narratives, humans bring to bear competencies in natural language understanding, commonsense reasoning, and deduction. These are competencies that a learning agent must possess or acquire to master the domain.

Finally, interactive narrative enables us to study storytelling with LLMs. Storytelling is a powerful, age-old form of human communication that, if mastered by machines, could greatly enhance their ability to engage entertainingly with people. It features many of the challenges discussed previously, including long-term coherence, and genre-specific and everyday commonsense reasoning. Automated storytelling intersects with interactive narrative in several ways, including: generation of language-based environments and scenarios [Guzdial et al., 2015; Fan et al., 2019; Tamari et al., 2019; Womack and Freeman, 2019; Ammanabrolu et al., 2020a,b]; improvisational or collaborative in-game storytelling [Martin et al., 2016, 2017; Mirowski and Mathewson, 2019]; or persona-driven situated dialogue [Urbanek et al., 2019; Prabhumoye et al., 2020; Callison-Burch et al., 2022; Park et al., 2023] for NPCs.

Diversity and Inclusion

This workshop aims to provide an environment with open exchange of ideas, freedom of thought and expression, and respectful scientific debate. Thus harassment and hostile behavior (Including but not limited to harassment based on race, gender, religion, age, color, appearance, national origin, ancestry, disability, sexual orientation, or gender identity.) are unwelcome in the workshop.

During the workshop, any participant who experiences harassment or hostile behavior may contact any of our organizing committee members, the organizers will take actions upon the situation to make sure we have a diverse, inclusive and friendly environment.

References