View, browse and sort the ever-growing list of sessions by pass type, topic, format and viewing experience. Use the experience filter to view sessions and roundtables that will be available on the virtual GDC event platform. Virtual sessions will be available to watch on-demand through April 1st, 2022 at 5pm PT. All days/times are displayed in Pacific Time.
Use the Viewing Experience filter to see which sessions will be available on-site, on the virtual platform, or both:
All registered attendees will be able to build their personal schedule in the virtual GDC event platform and mobile app. Download the Informa Tech Events app to gain access to the Game Developers Conference 2022 virtually. Login with the email address you used to register.
Suma Bailis (Senior Machine Learning Engineer, Google)
Jane Friedhoff (Senior UX Engineer, Google)
Location: Room 2009, West Hall
Date: Monday, March 21
Time: 3:00 pm - 3:30 pm
Pass Type: All Access Pass, Summits Pass - Get your pass now!
Topic: Programming
Format: Session
Viewing Experience: In-Person
Vault Recording: Video
Audience Level: No
If you’ve played with Dungeon AI or LitRPG Adventures, you’ve seen the power of language models like GPT-3 and LaMDA in action. However, while large language models can generate story and dialogue that makes sense, their coherence often only lasts for a few hundred words before they lose the thread of what they’re talking about, making it difficult to use them out of the box for goal-oriented experiences like games. In this talk, we demonstrate how to use LLMs in a graph-based authorship tool inspired by Twine to create narrative games populated with coherent and immersive NPCs that use dialogue unique to each playthrough and tailored to the player's input. We will walk through the creation of a graph, demo playthrough of the resulting narrative game, and equip the audience with tools to build next generation narrative games using off the shelf language models.
The audience will develop an understanding of how large language models can be used for goal-oriented narrative and dialogue-based gameplay. They will learn techniques for combining large language models with graph-based structures to provide for highly open-ended and immersive experiences with NPCs that still stay "on rails".
The intended audience is all levels of AI/Gameplay programmers interested in large language models and how they can be used to build next generation narrative games. No prior machine learning knowledge or training is required.