Скачать с ютуб Generative Agents: Interactive Simulacra of Human Behavior в хорошем качестве

Generative Agents: Interactive Simulacra of Human Behavior

UCL Information Security Research Seminar on 25.05.23 Speaker: Joon Sung Park Abstract: Believable proxies of human behavior can empower interactive applications ranging from immersive environments to rehearsal spaces for interpersonal communication to prototyping tools. In this paper, we introduce generative agents–computational software agents that simulate believable human behavior. Generative agents wake up, cook breakfast, and head to work; artists paint, while authors write; they form opinions, notice each other, and initiate conversations; they remember and reflect on days past as they plan the next day. To enable generative agents, we describe an architecture that extends a large language model to store a complete record of the agent’s experiences using natural language, synthesize those memories over time into higher-level reflections, and retrieve them dynamically to plan behavior. We instantiate generative agents to populate an interactive sandbox environment inspired by The Sims, where end users can interact with a small town of twenty five agents using natural language. In an evaluation, these generative agents produce believable individual and emergent social behaviors: for example, starting with only a single user-specified notion that one agent wants to throw a Valentine’s Day party, the agents autonomously spread invitations to the party over the next two days, make new acquaintances, ask each other out on dates to the party, and coordinate to show up for the party together at the right time. We demonstrate through ablation that the components of our agent architecture–observation, planning, and reflection–each contribute critically to the believability of agent behavior. By fusing large language models with computational, interactive agents, this work introduces architectural and interaction patterns for enabling believable simulations of human behavior. Bio: Joon Sung Park is a third-year computer science PhD student in the Human-Computer Interaction and Natural Language Processing groups at Stanford University, advised by Michael S. Bernstein and Percy Liang. He explores how we can leverage advances in natural language processing and machine learning, such as the development of large language models, to enable new interactive opportunities. His work has won a best paper award at CHI as well as multiple best paper nominations and other paper awards at CHI, CSCW, and ASSETS, and has been reported in venues such as Nature Machine Intelligence and Communications of ACM. Joon is recognized with the Microsoft Research Ph.D. Fellowship (2022), Stanford School of Engineering Fellowship (2021), and Siebel Scholar Award (2019). He holds a bachelor’s degree in Computer Science from Swarthmore College, and a master’s degree in Computer Science under the supervision of Karrie Karahalios from UIUC.