Theory of Mind (ToM) can be used to assess the capabilities of Large Language Models (LLMs) in complex scenarios where social reasoning is required. While the research community has proposed many ToM benchmarks, their hardness varies greatly, and their complexity is not well defined. This work proposes a framework to measure the complexity of ToM tasks. We quantify a problem's complexity as the number of states necessary to solve it correctly. Our complexity measure also accounts for spurious states of a ToM problem designed to make it apparently harder. We use our method to assess the complexity of five widely adopted ToM benchmarks. On top of this framework, we design a prompting technique that augments the information available to a model with a description of how the environment changes with the agents' interactions. We name this technique Discrete World Models (DWM) and show how it elicits superior performance on ToM tasks.

The emergence of large language models (LLMs) has significantly advanced the simulation of believable interactive agents. However, the substantial cost on maintaining the prolonged agent interactions poses challenge over the deployment of believable LLM-based agents. Therefore, in this paper, we develop Affordable Generative Agents (AGA), a framework for enabling the generation of believable and low-cost interactions on both agent-environment and inter-agents levels. Specifically, for agent-environment interactions, we substitute repetitive LLM inferences with learned policies; while for inter-agent interactions, we model the social relationships between agents and compress auxiliary dialogue information. Extensive experiments on multiple environments show the effectiveness and efficiency of our proposed framework. Also, we delve into the mechanisms of emergent believable behaviors lying in LLM agents, demonstrating that agents can only generate finite behaviors in fixed environments, based upon which, we understand ways to facilitate emergent interaction behaviors. Our code is publicly available at: \url{https://github.com/AffordableGenerativeAgents/Affordable-Generative-Agents}.

When Virginia Woolf wrote this innocuous sentence in "A Room of One's Own," her foundational work of feminist criticism, she opened the door to another field, still decades in the future--that of queer literary criticism. Do not blush," Woolf cautioned her audience. "Let us admit in the privacy of our own society that these things sometimes happen. Sometimes women do like women." Chloe and Olivia are characters in a book that Woolf has invented, a mediocre novel by a writer she names Mary Carmichael. Ostensibly, the women are friends and colleagues, not lovers, but Woolf drops clues for attentive readers. At one point, she interrupts her train of thought to ask for reassurance that Sir Chartres Biron is not lurking somewhere in the room. When she gave her original talks, Biron had recently been appointed the chief magistrate in an obscenity case that had been brought against the publisher of Radclyffe Hall's "The Well of Loneliness," a novel about a girl named Stephen who wants to be ...

Fox News correspondent Matt Finn has the latest on the impact of AI technology that some say could outpace humans on'Special Report.' Stanford researchers have leveraged generative artificial intelligence (AI) to create a simulated town comprising various characters, each with unique identities, memories and behaviors. The simulation, discussed at length in the new research paper "Generative Agents: Interactive Simulacra of Human Behavior," has been compared to an advanced version of the life simulation videogame "The Sims," as well as the HBO sci-fi series "Westworld." The latter tells the story of a theme park where robots are preloaded with storylines and personalities for wealthy human guests to interact with. Each day the robots are reset to their core tasks, but until then, they act like real humans, remembering their experiences, what people said to them and how they relate to the world around them.

We kicked off 2022 with a wonderful post from a fifth grader, Isabella, who (via her father, Rod Fuentes) wrote up a report of her science fair project on optimizing garbage routes in order to reduce litter in her city using computer vision. The final product was a heat map to be offered up to city council for implementation. I particularly loved the way Isabella combined her love for our planet and coding to improve her space, and I especially enjoy reading posts from our young authors. I always appreciate posts that touch on the more ethical aspects of data science, and Aisulu Omar did a wonderful job of shedding light on the importance of representation within data itself, the workplace, and how the two are connected. Addressing systemic biases within data science is so important and this thorough write-up was very effective in breaking down what one needs to know.

KnightSpear intensifies Isabella's machine learning capability to provide a better and smarter AI work coach for project management. Isabella, KnightSpear's AI-enabled chatbot work coach, will soon have the ability to process data within the app, and send reports and suggestions to project managers to help them efficiently manage tasks, strategically plan projects and keep their team motivated and engaged. Utilizing machine learning capability, these enhancements enable Isabella to find patterns in a user's work behavior and translate it into data-based predictions. This means Isabella can gather and collate information about trends in the team's personality and productivity, project performance and the team's progress and use it to provide status reports and make suggestions on how a project manager can efficiently manage tasks and engage her team. Isabella will work closely with the project manager, which is a new concept for a project management app.

The pressure was on for Abdul Gouda and his classmates at George Mason University: Not only did their graduation depend on the success of their project, but so did the hopes of an impossibly cute 10-year-old girl. Fifth-grader Isabella Nicola wanted to play the violin, but she was born with no left hand and a severely abbreviated forearm. Her music teacher at Island Creek Elementary in Fairfax County had built her a prosthetic allowing her to move the bow with her left arm and finger the strings with her right -- the opposite of how violin is usually taught. But the prosthetic was heavy and he thought there might be a better option. He reached out to Mason, his alma mater.