A company that has worked closely with the UK government on artificial intelligence safety, the NHS and education is also developing AI for military drones. The consultancy Faculty AI has "experience developing and deploying AI models on to UAVs", or unmanned aerial vehicles, according to a defence industry partner company. Faculty has emerged as one of the most active companies selling AI services in the UK. Unlike the likes of OpenAI, Deepmind or Anthropic, it does not develop models itself, instead focusing on reselling models, notably from OpenAI, and consulting on their use in government and industry. Faculty gained particular prominence in the UK after working on data analysis for the Vote Leave campaign before the Brexit vote.

Nvidia already sells boatloads of computer chips to every major company building proprietary artificial intelligence models. But now, at a moment when public interest in open source and do-it-yourself AI is soaring, the company announced it will also begin offering a "personal AI supercomputer," later this year starting at 3,000 that anyone can use in their own home or office. Nvidia's new desktop machine, dubbed Digits, will go on sale in May and is about the size of a small book. It contains an Nvidia "superchip" called GB10 Grace Blackwell optimized to accelerate the computations needed to train and run AI models, and comes equipped with 128GB of unified memory and up to 4TB of NVMe storage for handling especially large AI programs. Jensen Huang, founder and CEO of Nvidia, announced the new system, along with several other AI offerings, during a keynote speech today at CES, an annual confab for the computer industry held in Las Vegas (you can check out all of the biggest announcements on the WIRED CES live blog). "Placing an AI supercomputer on the desks of every data scientist, AI researcher and student empowers them to engage and shape the age of AI," Huang said in a statement released ahead of his keynote.

Data-driven soft sensors are crucial in predicting key performance indicators in industrial systems. However, current methods predominantly rely on the supervised learning paradigms of parameter updating, which inherently faces challenges such as high development costs, poor robustness, training instability, and lack of interpretability. Recently, large language models (LLMs) have demonstrated significant potential across various domains, notably through In-Context Learning (ICL), which enables high-performance task execution with minimal input-label demonstrations and no prior training. This paper aims to replace supervised learning with the emerging ICL paradigm for soft sensor modeling to address existing challenges and explore new avenues for advancement. To achieve this, we propose a novel framework called the Few-shot Uncertainty-aware and self-Explaining Soft Sensor (LLM-FUESS), which includes the Zero-shot Auxiliary Variable Selector (LLM-ZAVS) and the Uncertainty-aware Few-shot Soft Sensor (LLM-UFSS). The LLM-ZAVS retrieves from the Industrial Knowledge Vector Storage to enhance LLMs' domain-specific knowledge, enabling zero-shot auxiliary variable selection. In the LLM-UFSS, we utilize text-based context demonstrations of structured data to prompt LLMs to execute ICL for predicting and propose a context sample retrieval augmentation strategy to improve performance. Additionally, we explored LLMs' AIGC and probabilistic characteristics to propose self-explanation and uncertainty quantification methods for constructing a trustworthy soft sensor. Extensive experiments demonstrate that our method achieved state-of-the-art predictive performance, strong robustness, and flexibility, effectively mitigates training instability found in traditional methods. To the best of our knowledge, this is the first work to establish soft sensor utilizing LLMs.

Mixture-of-Expert (MoE) models outperform conventional models by selectively activating different subnets, named \emph{experts}, on a per-token basis. This gated computation generates dynamic communications that cannot be determined beforehand, challenging the existing GPU interconnects that remain \emph{static} during the distributed training process. In this paper, we advocate for a first-of-its-kind system, called mFabric, that unlocks topology reconfiguration \emph{during} distributed MoE training. Towards this vision, we first perform a production measurement study and show that the MoE dynamic communication pattern has \emph{strong locality}, alleviating the requirement of global reconfiguration. Based on this, we design and implement a \emph{regionally reconfigurable high-bandwidth domain} on top of existing electrical interconnects using optical circuit switching (OCS), achieving scalability while maintaining rapid adaptability. We have built a fully functional mFabric prototype with commodity hardware and a customized collective communication runtime that trains state-of-the-art MoE models with \emph{in-training} topology reconfiguration across 32 A100 GPUs. Large-scale packet-level simulations show that mFabric delivers comparable performance as the non-blocking fat-tree fabric while boosting the training cost efficiency (e.g., performance per dollar) of four representative MoE models by 1.2$\times$--1.5$\times$ and 1.9$\times$--2.3$\times$ at 100 Gbps and 400 Gbps link bandwidths, respectively.

Project Chrono [1] is an open-source, physics-based simulation framework that supports the modeling, simulation, and analysis of complex systems. It is designed for high-performance, high-fidelity simulations and is widely used in research and industry. PyChrono [2] is the Python wrapper for Project Chrono, providing a user-friendly interface to the core functionalities of Project Chrono. It allows users to leverage the power of Project Chrono using Python, making it accessible to a broader range of users who prefer scripting in Python over C++. Project Chrono encompasses a wide range of features, and PyChrono inherits a subset of these capabilities: 1. Chrono::Engine: Provides core functionality for multibody dynamics and nonlinear finite element analysis, with robust treatment of friction and contact using both the penalty method and the Lagrange-multiplier method.

Pre-training a language model and then fine-tuning it has shown to be an efficient and effective technique for a wide range of code intelligence tasks, such as code generation, code summarization, and vulnerability detection. However, pretraining language models on a large-scale code corpus is computationally expensive. Fortunately, many off-the-shelf Pre-trained Code Models (PCMs), such as CodeBERT, CodeT5, CodeGen, and Code Llama, have been released publicly. These models acquire general code understanding and generation capability during pretraining, which enhances their performance on downstream code intelligence tasks. With an increasing number of these public pre-trained models, selecting the most suitable one to reuse for a specific task is essential. In this paper, we systematically investigate the reusability of PCMs. We first explore three intuitive model selection methods that select by size, training data, or brute-force fine-tuning. Experimental results show that these straightforward techniques either perform poorly or suffer high costs. Motivated by these findings, we explore learning-based model selection strategies that utilize pre-trained models without altering their parameters. Specifically, we train proxy models to gauge the performance of pre-trained models, and measure the distribution deviation between a model's latent features and the task's labels, using their closeness as an indicator of model transferability. We conduct experiments on 100 widely-used opensource PCMs for code intelligence tasks, with sizes ranging from 42.5 million to 3 billion parameters. The results demonstrate that learning-based selection methods reduce selection time to 100 seconds, compared to 2,700 hours with brute-force fine-tuning, with less than 6% performance degradation across related tasks.

Human communication is a multifaceted and multimodal skill. Communication requires an understanding of both the surface-level textual content and the connotative intent of a piece of communication. In humans, learning to go beyond the surface level starts by learning communicative intent in speech. Once humans acquire these skills in spoken communication, they transfer those skills to written communication. In this paper, we assess the ability of speech+text models and text models trained with special emphasis on human-to-human conversations to make this multimodal transfer of skill. We specifically test these models on their ability to detect covert deceptive communication. We find that with no special prompting speech+text LLMs have an advantage over unimodal LLMs in performing this task. Likewise, we find that human-to-human conversation-trained LLMs are also advantaged in this skill.

With the rapid scaling of large language models (LLMs), structured pruning has become a widely used technique to learn efficient, smaller models from larger ones, delivering superior performance compared to training similarly sized models from scratch. In this paper, we move beyond the traditional static pruning approach of determining a fixed pruning mask for a model, and propose a dynamic approach to structured pruning. In our method, the pruning mask is input-dependent and adapts dynamically based on the information described in a user instruction. Our approach, termed "instruction-following pruning", introduces a sparse mask predictor that takes the user instruction as input and dynamically selects the most relevant model parameters for the given task. To identify and activate effective parameters, we jointly optimize the sparse mask predictor and the LLM, leveraging both instruction-following data and the pre-training corpus. Experimental results demonstrate the effectiveness of our approach on a wide range of evaluation benchmarks. For example, our 3B activated model improves over the 3B dense model by 5-8 points of absolute margin on domains such as math and coding, and rivals the performance of a 9B model.

This study advances the field of conflict forecasting by using text-based actor embeddings with transformer models to predict dynamic changes in violent conflict patterns at the actor level. More specifically, we combine newswire texts with structured conflict event data and leverage recent advances in Natural Language Processing (NLP) techniques to forecast escalations and de-escalations among conflicting actors, such as governments, militias, separatist movements, and terrorists. This new approach accurately and promptly captures the inherently volatile patterns of violent conflicts, which existing methods have not been able to achieve. To create this framework, we began by curating and annotating a vast international newswire corpus, leveraging hand-labeled event data from the Uppsala Conflict Data Program. By using this hybrid dataset, our models can incorporate the textual context of news sources along with the precision and detail of structured event data. This combination enables us to make both dynamic and granular predictions about conflict developments. We validate our approach through rigorous back-testing against historical events, demonstrating superior out-of-sample predictive power. We find that our approach is quite effective in identifying and predicting phases of conflict escalation and de-escalation, surpassing the capabilities of traditional models. By focusing on actor interactions, our explicit goal is to provide actionable insights to policymakers, humanitarian organizations, and peacekeeping operations in order to enable targeted and effective intervention strategies.

This paper presents a new Large Language Model (LLM)-based Smart Device Management framework, a pioneering approach designed to address the intricate challenges of managing intelligent devices within public facilities, with a particular emphasis on applications to libraries. Our framework leverages state-of-the-art LLMs to analyze and predict device failures, thereby enhancing operational efficiency and reliability. Through prototype validation in real-world library settings, we demonstrate the framework's practical applicability and its capacity to significantly reduce budgetary constraints on public facilities. The advanced and innovative nature of our model is evident from its successful implementation in prototype testing. We plan to extend the framework's scope to include a wider array of public facilities and to integrate it with cutting-edge cybersecurity technologies, such as Internet of Things (IoT) security and machine learning algorithms for threat detection and response. This will result in a comprehensive and proactive maintenance system that not only bolsters the security of intelligent devices but also utilizes machine learning for automated analysis and real-time threat mitigation. By incorporating these advanced cybersecurity elements, our framework will be well-positioned to tackle the dynamic challenges of modern public infrastructure, ensuring robust protection against potential threats and enabling facilities to anticipate and prevent failures, leading to substantial cost savings and enhanced service quality.