Large multi-agent systems are often driven by locally defined agent interactions, which is referred to as self-organization. Our primary objective is to determine when the propagation of such local interactions will reach a specific agent of interest. Although conventional approaches that reconstruct all agent states can be used, they may entail unnecessary computational costs. In this paper, we investigate a CNN-LSTM model to forecast the state of a particular agent in a large self-organizing multi-agent system without the reconstruction. The proposed model comprises a CNN encoder to represent the system in a low-dimensional vector, a LSTM module to learn agent dynamics in the vector space, and a MLP decoder to predict the future state of an agent. As an example, we consider a forest fire model where we aim to predict when a particular tree agent will start burning. We compare the proposed model with reconstruction-based approaches such as CNN-LSTM and ConvLSTM. The proposed model exhibits similar or slightly worse AUC but significantly reduces computational costs such as activation than ConvLSTM. Moreover, it achieves higher AUC with less computation than the recontruction-based CNN-LSTM.

Language models (LMs) now excel at many tasks such as few-shot learning, question answering, reasoning, and dialog. However, they sometimes generate unsupported or misleading content. A user cannot easily determine whether their outputs are trustworthy or not, because most LMs do not have any built-in mechanism for attribution to external evidence. To enable attribution while still preserving all the powerful advantages of recent generation models, we propose RARR (Retrofit Attribution using Research and Revision), a system that 1) automatically finds attribution for the output of any text generation model and 2) post-edits the output to fix unsupported content while preserving the original output as much as possible. When applied to the output of several state-of-the-art LMs on a diverse set of generation tasks, we find that RARR significantly improves attribution while otherwise preserving the original input to a much greater degree than previously explored edit models. Furthermore, the implementation of RARR requires only a handful of training examples, a large language model, and standard web search.

In Multi-Document Summarization (MDS), the input can be modeled as a set of documents, and the output is its summary. In this paper, we focus on pretraining objectives for MDS. Specifically, we introduce a novel pretraining objective, which involves selecting the ROUGE-based centroid of each document cluster as a proxy for its summary. Our objective thus does not require human written summaries and can be utilized for pretraining on a dataset consisting solely of document sets. Through zero-shot, few-shot, and fully supervised experiments on multiple MDS datasets, we show that our model Centrum is better or comparable to a state-of-the-art model. We make the pretrained and fine-tuned models freely available to the research community https://github.com/ratishsp/centrum.

There are just a handful of companies that have surpassed the $1 trillion market cap, including Google parent Alphabet, Microsoft, Saudi Aramco, Amazon and Apple. On Tuesday, California-based Nvidia's market cap jumped high enough to be among their ranks. The chipmaker, which makes graphics processing units (GPUs) that help power generative artificial intelligence platforms, has seen its stock price soar as more companies look to expand their AI offerings. Nvidia hit a market cap of $1 trillion Tuesday with shares opening at $405.95, although it eased below that milestone by midday after shares dipped below $404.86. The company's valuation puts it above Facebook parent company Meta, Warren Buffett's Berkshire Hathaway and Elon Musk's Tesla.

In February, Meta released its large language model: LLaMA. Unlike OpenAI and its ChatGPT, Meta didn't just give the world a chat window to play with. Instead, it released the code into the open-source community, and shortly thereafter the model itself was leaked. Researchers and programmers immediately started modifying it, improving it, and getting it to do things no one else anticipated. And their results have been immediate, innovative, and an indication of how the future of this technology is going to play out.

Speaking on Russian TV, Mr Putin said the attack had been a response to what he described as a Russian attack on Ukraine's military intelligence HQ in recent days. The BBC is unable to independently verify whether any such attack took place.

The drone attack on Moscow early Tuesday morning showed that the war is real and near, not just for Ukrainians but also for Russians--a message that can't be good for Vladimir Putin. At least eight drones flew over Russia's capital in the wee hours, almost certainly launched by Ukraine (or perhaps by Russian rebels sympathetic to Ukraine's cause). The Kremlin claims that air-defense crews shot down or electronically jammed all the drones and that the damage done to a few apartment buildings was caused by metal shards of the disabled airframes as they fell from the sky. Even if this claim is true, it doesn't matter. The attack demonstrates that Russia's skies are porous, that Russian civilians are vulnerable.

A group of leading technology experts from across the globe have warned that artificial intelligence technology should be considered a societal risk and prioritised in the same class as pandemics and nuclear wars. The brief statement, signed by hundreds of tech executives and academics, was released by the Center for AI Safety on Tuesday amid growing concerns over regulation and risks the technology poses to humanity. "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war," the statement said. Signatories included the chief executives from Google's DeepMind, the ChatGPT developer OpenAI and AI startup Anthropic. The statement comes as global leaders and industry experts – such as the leaders of OpenAI – have made calls for regulation of the technology amid existential fears the technology could significantly affect job markets, harm the health of millions, and weaponise disinformation, discrimination and impersonation.

Peter Etchells, Professor of Psychology and Science Communication at Bath Spa University said: "The new Video Games Research Framework provides a much-needed set of standards to advance the scientific study of games in an ethical, progressive and robust manner.

A new open letter calling for regulation to mitigate'the risk of extinction from AI' has been signed by more than 350 industry experts, including several developing the tech. The 22-word statement reads: 'Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.' The short letter was signed by OpenAI CEO Sam Altman, creator of ChatGPT, who called on Congress to establish regulations for AI. While the document does not provide details, the statement likely aims to convince policymakers to create plans for the event AI goes rogue, just as there are plans in place for pandemics and nuclear wars. Altman was joined by other known leaders in AI, including Demis Hassabis of Google DeepMind, Dario Amodei of Anthropic and executives from Microsoft and Google.