The Ai Pin from Humane, a much-hyped startup founded by former Apple employees, will cost $699. That'a according to The Verge, which obtained documents about the device ahead of its official launch on November 9. In addition, the Pin will reportedly have a monthly $24 subscription fee for access to T-Mobile's cellular network and large language models from OpenAI and Microsoft to power its smarts. The Ai Pin is a device that's about the size of a large business card that clips on to your clothing magnetically and acts as a personalized assistant controlled via voice and touch. Notably, it doesn't have a screen.

Depending on the hardware you're using, training a large language model of any significant size can take weeks, months, even years to complete. That's no way to do business -- nobody has the electricity and time to be waiting that long. On Wednesday, NVIDIA unveiled the newest iteration of its Eos supercomputer, one powered by more than 10,000 H100 Tensor Core GPUs and capable of training a 175 billion-parameter GPT-3 model on 1 billion tokens in under four minutes. That's three times faster than the previous benchmark on the MLPerf AI industry standard, which NVIDIA set just six months ago. Eos represents an enormous amount of compute.

OpenAI's extremely popular ChatGPT service was down and non-functional for its 100 million weekly active users. The service went down just before 9AM ET. OpenAI has acknowledged the outage and said that it's also impacting the company's API services. However, the service was restored at around 10:50 AM ET. Instead of a working platform, ChatGPT users were greeted with a warning message that says it's "at capacity right now." OpenAI wrote in an error report that it had "identified an issue resulting in high error rates across the API and ChatGPT, and we are working on remediation."

Existing datasets for automated fact-checking have substantial limitations, such as relying on artificial claims, lacking annotations for evidence and intermediate reasoning, or including evidence published after the claim. In this paper we introduce AVeriTeC, a new dataset of 4,568 real-world claims covering fact-checks by 50 different organizations. Each claim is annotated with question-answer pairs supported by evidence available online, as well as textual justifications explaining how the evidence combines to produce a verdict. Through a multi-round annotation process, we avoid common pitfalls including context dependence, evidence insufficiency, and temporal leakage, and reach a substantial inter-annotator agreement of $\kappa=0.619$ on verdicts. We develop a baseline as well as an evaluation scheme for verifying claims through several question-answering steps against the open web.

Large Language Models (LLMs) do not differentially represent numbers, which are pervasive in text. In contrast, neuroscience research has identified distinct neural representations for numbers and words. In this work, we investigate how well popular LLMs capture the magnitudes of numbers (e.g., that $4 < 5$) from a behavioral lens. Prior research on the representational capabilities of LLMs evaluates whether they show human-level performance, for instance, high overall accuracy on standard benchmarks. Here, we ask a different question, one inspired by cognitive science: How closely do the number representations of LLMscorrespond to those of human language users, who typically demonstrate the distance, size, and ratio effects? We depend on a linking hypothesis to map the similarities among the model embeddings of number words and digits to human response times. The results reveal surprisingly human-like representations across language models of different architectures, despite the absence of the neural circuitry that directly supports these representations in the human brain. This research shows the utility of understanding LLMs using behavioral benchmarks and points the way to future work on the number representations of LLMs and their cognitive plausibility.

As research in human-centered NLP advances, there is a growing recognition of the importance of incorporating human and social factors into NLP models. At the same time, our NLP systems have become heavily reliant on LLMs, most of which do not model authors. To build NLP systems that can truly understand human language, we must better integrate human contexts into LLMs. This brings to the fore a range of design considerations and challenges in terms of what human aspects to capture, how to represent them, and what modeling strategies to pursue. To address these, we advocate for three positions toward creating large human language models (LHLMs) using concepts from psychological and behavioral sciences: First, LM training should include the human context. Second, LHLMs should recognize that people are more than their group(s). Third, LHLMs should be able to account for the dynamic and temporally-dependent nature of the human context. We refer to relevant advances and present open challenges that need to be addressed and their possible solutions in realizing these goals.

Mathematical understanding and reasoning are crucial tasks for assessing the capabilities of artificial intelligence (AI). However, existing benchmarks either require just a few steps of reasoning, or only contain a small amount of data in one specific topic, making it hard to analyse AI's behaviour with reference to different problems within a specific topic in detail. In this work, we propose Conic10K, a challenging math problem dataset on conic sections in Chinese senior high school education. Our dataset contains various problems with different reasoning depths, while only the knowledge from conic sections is required. Since the dataset only involves a narrow range of knowledge, it is easy to separately analyse the knowledge a model possesses and the reasoning ability it has. For each problem, we provide a high-quality formal representation, the reasoning steps, and the final solution. Experiments show that existing large language models, including GPT-4, exhibit weak performance on complex reasoning. We hope that our findings could inspire more advanced techniques for precise natural language understanding and reasoning. Our dataset and codes are available at https://github.com/whyNLP/Conic10K.

Task assignment and scheduling algorithms are powerful tools for autonomously coordinating large teams of robotic or AI agents. However, the decisions these system make often rely on components designed by domain experts, which can be difficult for non-technical end-users to understand or modify to their own ends. In this paper we propose a preliminary design for a flexible natural language interface for a task assignment system. The goal of our approach is both to grant users more control over a task assignment system's decision process, as well as render these decisions more transparent. Users can direct the task assignment system via natural language commands, which are applied as constraints to a mixed-integer linear program (MILP) using a large language model (LLM). Additionally, our proposed system can alert users to potential issues with their commands, and engage them in a corrective dialogue in order to find a viable solution. We conclude with a description of our planned user-evaluation in the simulated environment Overcooked and describe next steps towards developing a flexible and transparent task allocation system.

We introduce Lumos, a novel framework for training language agents that employs a unified data format and a modular architecture based on open-source large language models (LLMs). Lumos consists of three distinct modules: planning, grounding, and execution. The planning module breaks down a task into a series of high-level, tool-agnostic subgoals, which are then made specific by the grounding module through a set of low-level actions. These actions are subsequently executed by the execution module, utilizing a range of off-the-shelf tools and APIs. In order to train these modules effectively, high-quality annotations of subgoals and actions were collected and are made available for fine-tuning open-source LLMs for various tasks such as complex question answering, web tasks, and math problems. Leveraging this unified data and modular design, Lumos not only achieves comparable or superior performance to current, state-of-the-art agents, but also exhibits several key advantages: (1) Lumos surpasses GPT-4/3.5-based agents in complex question answering and web tasks, while equalling the performance of significantly larger LLM agents on math tasks; (2) Lumos outperforms open-source agents created through conventional training methods and those using chain-of-thoughts training; and (3) Lumos is capable of effectively generalizing to unseen interactive tasks, outperforming larger LLM-based agents and even exceeding performance of specialized agents.

Large Language Models (LLMs) are increasingly being used for interactive decision-making tasks requiring planning and adapting to the environment. Recent works employ LLMs-as-agents in broadly two ways: iteratively determining the next action (iterative executors) or generating plans and executing sub-tasks using LLMs (plan-and-execute). However, these methods struggle with task complexity, as the inability to execute any sub-task may lead to task failure. To address these shortcomings, we introduce As-Needed Decomposition and Planning for complex Tasks (ADaPT), an approach that explicitly plans and decomposes complex sub-tasks as-needed, i.e., when the LLM is unable to execute them. ADaPT recursively decomposes sub-tasks to adapt to both task complexity and LLM capability. Our results demonstrate that ADaPT substantially outperforms established strong baselines, achieving success rates up to 28.3% higher in ALFWorld, 27% in WebShop, and 33% in TextCraft -- a novel compositional dataset that we introduce. Through extensive analysis, we illustrate the importance of multilevel decomposition and establish that ADaPT dynamically adjusts to the capabilities of the executor LLM as well as to task complexity.