Showing papers by "Michael S. Bernstein published in 2023"
TL;DR: In this paper , the authors 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.
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.
69 citations
06 Mar 2023
TL;DR: Model sketching as mentioned in this paper is a framework for iteratively and rapidly authoring functional approximations of a machine learning model's decision-making logic, which refocuses practitioner attention on composing high-level, human-understandable concepts that the model is expected to reason over (e.g., profanity, racism, or sarcasm in a content moderation task).
Abstract: Machine learning practitioners often end up tunneling on low-level technical details like model architectures and performance metrics. Could early model development instead focus on high-level questions of which factors a model ought to pay attention to? Inspired by the practice of sketching in design, which distills ideas to their minimal representation, we introduce model sketching: a technical framework for iteratively and rapidly authoring functional approximations of a machine learning model’s decision-making logic. Model sketching refocuses practitioner attention on composing high-level, human-understandable concepts that the model is expected to reason over (e.g., profanity, racism, or sarcasm in a content moderation task) using zero-shot concept instantiation. In an evaluation with 17 ML practitioners, model sketching reframed thinking from implementation to higher-level exploration, prompted iteration on a broader range of model designs, and helped identify gaps in the problem formulation—all in a fraction of the time ordinarily required to build a model.
5 citations
31 Jan 2023
TL;DR: The authors performed a large-scale measurement study primarily of 70, 000 patent citations to premier human-computer interaction research venues, tracing how HCI research are cited in United States patents over the last 30 years.
Abstract: What is the impact of human-computer interaction research on industry? While it is impossible to track all research impact pathways, the growing literature on translational research impact measurement offers patent citations as one measure of how industry recognizes and draws on research in its inventions. In this paper, we perform a large-scale measurement study primarily of 70, 000 patent citations to premier HCI research venues, tracing how HCI research are cited in United States patents over the last 30 years. We observe that 20.1% of papers from these venues, including 60–80% of papers at UIST and 13% of papers in a broader dataset of SIGCHI-sponsored venues overall, are cited by patents—far greater than premier venues in science overall (9.7%) and NLP (11%). However, the time lag between a patent and its paper citations is long (10.5 years) and getting longer, suggesting that HCI research and practice may not be efficiently connected.
2 citations
TL;DR: The authors make a first attempt to characterize image accessibility on Wikipedia across languages, present new experimental results that can inform efforts to assess description quality, and offer some strategies to improve Wikipedia's image accessibility.
Abstract: We make a first attempt to characterize image accessibility on Wikipedia across languages, present new experimental results that can inform efforts to assess description quality, and offer some strategies to improve Wikipedia's image accessibility.