Dark Shadows
Dark Shadows Human Computation Gaming Intelligent Agents Video Games

Dark Shadows

Address the real-world problem of human trafficking while playing a video game.

By Steph Buongiorno, Jake Klinkert, Tanishq Chawla, Zixin Zhaung, and Corey Clark at Southern Methodist University’s Guildhall (video game) and Computer Science departments.

Iterations:

I: Buongiorno conceived of Dark Shadows as a film noir-style “document thriller” (inspired by Papers, Please! and Night Call). Buongiorno designed and developed underlying “human-in-the-loop” mechanics to collect player feedback to train amachine learning model, which can disambiguate speakers, locations, and events in real human trafficking data. See Dark Shadows, as well as its relation to other dimensions of human trafficking research conducted at SMU, on CBS News.

II: Buongiorno designed NPCs that leverage large language models (LLMs) and a self-reflection-based validation system to enable dynamic, free-form interactions with the player aligned with a procedural game narrative (re: PANGeA). NPCs express traits from the Big 5 Personality Model in their responses, while the validation system ensures responses stay on narrative track. https://www.youtube.com/watch?v=ys7vTkUZ8ic Dark Shadows, Scene III

III: [Forthcoming] Dark Shadows will include a social engineering scene, where the player must investigate the NPCs’ personalities and use rhetorical devices to gain clues on the mystery.

You may also like

See all Dark Shadows
PANGeA

PANGeA

PANGeA is a system that uses large language models (LLMs) to create narrative content for turn-based RPGs based on game designers' high-level criteria. It introduces a novel validation system for handling free-form text input during development and gameplay, employing "self-reflection" techniques, enabling small/local LLMs to perform comparably to foundational models. It enriches player-NPC interactions by generating personality-biased non-playable characters (NPCs). It improves AI accuracy through crowdsourcing mechanics. PANGeA houses a server with a custom memory system that provides context for LLM generation. The server's REST interface enables integration with any game engine.

GAME-KG

GAME-KG

Knowledge graphs (KGs) can augment large language models (LLMs) while also providing an explainable set of facts that can be inspected by a human. Explainability is valuable for fields that may otherwise avoid LLMs due to hallucinations, such as human trafficking analysis. Creating KGs poses challenges, however. KGs parsed from documents may include explicit connections (those directly stated in a document) but miss implicit connections (those evident to a human, but not directly stated). This research introduces GAME-KG, an approach to modifying explicit and implicit KG connections by crowdsourcing feedback through video games.

Agent-Driven, Game-Based Learning: Personalized CS Education for Diverse Students

Agent-Driven, Game-Based Learning: Personalized CS Education for Diverse Students

AI agents can personalize education by identifying students’ strengths, weaknesses, and personalities to generate content tailored to them. This work presents "personalized education agents" deployed in an educational version of Minecraft. Agents bridge concepts from lessons to "big picture" thinking by creating connections between STEM and interdisciplinary topics, such as the Language Arts. Agents translate student progression and learning outcomes to teachers for their assessment of student progress.