Elon Musk said on the social media platform X on Monday that the first human patient has received a brain implant developed by his company Neuralink. After years of delays, Neuralink started recruiting patients for a clinical trial in the fall after receiving approval from the US Food and Drug Administration and a hospital ethics board. The company is developing a device called a brain-computer interface. Musk has said that Neuralink's ultimate goal is to "achieve a symbiosis with artificial intelligence," but for now he's starting with a far more modest aim: allowing paralyzed people to control a cursor or keyboard with their brains. In a brochure about the study, Neuralink says it is recruiting participants with quadriplegia, or paralysis in all four limbs, due to cervical spinal cord injury or amyotrophic lateral sclerosis (ALS) and that are at least 22 years old.

Elon Musk, Neuralink's billionaire founder, said the first human received an implant from the brain-chip startup on Sunday and is recovering well, in a post on Twitter/X on Monday. The US Food and Drug Administration (FDA) had given the company clearance last year to conduct its first trial to test its implant on humans. "Initial results show promising neuron spike detection," Musk added. The startup's Prime study is a trial for its wireless brain-computer interface to evaluate the safety of the implant and surgical robot. The study will assess the functionality of the interface, which enables people with quadriplegia, or paralysis of all four limbs, to control devices with their thoughts, according to the company's website.

However, it is still challenging to solve practical problems such as blood flows, atmospheric and ocean currents, wildfires, and wind turbines because of their multiscale nature. Resolving all scales is computationally infeasible. As a result, physical modeling is typically carried out on a coarse grid using the appropriate subgrid-scale (SGS) models. For example, large eddy simulation (LES) resolves large-scale turbulent motion on a grid, which carries most of the flow energy, while modeling the small scales that have relatively little influence on the mean flow [8]. The goal of SGS models is to capture the effect of the small-scale structures that cannot be resolved in the grid on the resolved scales and to guarantee numerical stability [9]. The static Smagorinsky model [10] and dynamic Smagorinsky model [11], which predict the dissipation of SGS energy, are the most widely used SGS models for turbulence. In high-order DG methods, both Collis [12] and Sengupta et al. [13] successfully used the static Smagorinksy model and dynamic Smagorinsky model, respectively. However, these Smagorinsky models perform poorly for certain flows [14, 15] because they are based on the assumption that eddy viscosity is always purely dissipative and thus are unable to account for energy flow from small scales to large scales (backscatter) [16, 11]. Alternatively, numerical dissipation can be used for modeling unresolved scales as an implicit SGS model.

Large Vision-Language Models (VLMs) have demonstrated impressive performance on complex tasks involving visual input with natural language instructions. However, it remains unclear to what extent capabilities on natural images transfer to Earth observation (EO) data, which are predominantly satellite and aerial images less common in VLM training data. In this work, we propose a comprehensive benchmark to gauge the progress of VLMs toward being useful tools for EO data by assessing their abilities on scene understanding, localization and counting, and change detection tasks. Motivated by real-world applications, our benchmark includes scenarios like urban monitoring, disaster relief, land use, and conservation. We discover that, although state-of-the-art VLMs like GPT-4V possess extensive world knowledge that leads to strong performance on open-ended tasks like location understanding and image captioning, their poor spatial reasoning limits usefulness on object localization and counting tasks. Our benchmark will be made publicly available on this website and on Hugging Face for easy model evaluation.

We introduce a novel Interval Bound Propagation (IBP) approach for the formal verification of object detection models, specifically targeting the Intersection over Union (IoU) metric. The approach has been implemented in an open source code, named IBP IoU, compatible with popular abstract interpretation based verification tools. The resulting verifier is evaluated on landing approach runway detection and handwritten digit recognition case studies. Comparisons against a baseline (Vanilla IBP IoU) highlight the superior performance of IBP IoU in ensuring accuracy and stability, contributing to more secure and robust machine learning applications.

Large language models (LLMs) encapsulate a vast amount of factual information within their pre-trained weights, as evidenced by their ability to answer diverse questions across different domains. However, this knowledge is inherently limited, relying heavily on the characteristics of the training data. Consequently, using external datasets to incorporate new information or refine the capabilities of LLMs on previously seen information poses a significant challenge. In this study, we compare two common approaches: unsupervised fine-tuning and retrieval-augmented generation (RAG). We evaluate both approaches on a variety of knowledge-intensive tasks across different topics. Our findings reveal that while unsupervised fine-tuning offers some improvement, RAG consistently outperforms it, both for existing knowledge encountered during training and entirely new knowledge. Moreover, we find that LLMs struggle to learn new factual information through unsupervised fine-tuning, and that exposing them to numerous variations of the same fact during training could alleviate this problem.

Large language models (LLMs) have exhibited an array of reasoning capabilities but face challenges like error propagation and hallucination, particularly in specialised areas like finance, where data is heterogeneous, and precision is paramount. We explore the potential of language model augmentation with external tools to mitigate these limitations and offload certain reasoning steps to external tools that are more suited for the task, instead of solely depending on the LLM's inherent abilities. More concretely, using financial domain question-answering datasets, we apply supervised fine-tuning on a LLaMA-2 13B Chat model to act both as a 'task router' and 'task solver'. The 'task router' dynamically directs a question to either be answered internally by the LLM or externally via the right tool from the tool set. Our tool-equipped SFT model, Raven, demonstrates an improvement of 35.2% and 5.06% over the base model and SFT-only baselines, respectively, and is highly competitive with strong GPT-3.5 results. To the best of our knowledge, our work is the first that investigates tool augmentation of language models for the finance domain.

ActorSim is a goal reasoning framework developed at the Naval Research Laboratory. Originally, all goal reasoning rules were hand-crafted. This work extends ActorSim with the capability of learning by demonstration, that is, when a human trainer disagrees with a decision made by the system, the trainer can take over and show the system the correct decision. The learning component uses Ripple-Down Rules (RDR) to build new decision rules to correctly handle similar cases in the future. The system is demonstrated using the RoboCup Rescue Agent Simulation, which simulates a city-wide disaster, requiring emergency services, including fire, ambulance and police, to be dispatched to different sites to evacuate civilians from dangerous situations. The RDRs are implemented in a scripting language, FrameScript, which is used to mediate between ActorSim and the agent simulator. Using Ripple-Down Rules, ActorSim can scale to an order of magnitude more goals than the previous version.

Spatial computing is a technological advancement that facilitates the seamless integration of devices into the physical environment, resulting in a more natural and intuitive digital world user experience. Spatial computing has the potential to become a significant advancement in the field of computing. From GPS and location-based services to healthcare, spatial computing technologies have influenced and improved our interactions with the digital world. The use of spatial computing in creating interactive digital environments has become increasingly popular and effective. This is explained by its increasing significance among researchers and industrial organisations, which motivated us to conduct this review. This review provides a detailed overview of spatial computing, including its enabling technologies and its impact on various applications. Projects related to spatial computing are also discussed. In this review, we also explored the potential challenges and limitations of spatial computing. Furthermore, we discuss potential solutions and future directions. Overall, this paper aims to provide a comprehensive understanding of spatial computing, its enabling technologies, their impact on various applications, emerging challenges, and potential solutions.

This paper presents a theoretical analysis and practical approach to the moral responsibilities when developing AI systems for non-military applications that may nonetheless be used for conflict applications. We argue that AI represents a form of crossover technology that is different from previous historical examples of dual- or multi-use technology as it has a multiplicative effect across other technologies. As a result, existing analyses of ethical responsibilities around dual-use technologies do not necessarily work for AI systems. We instead argue that stakeholders involved in the AI system lifecycle are morally responsible for uses of their systems that are reasonably foreseeable. The core idea is that an agent's moral responsibility for some action is not necessarily determined by their intentions alone; we must also consider what the agent could reasonably have foreseen to be potential outcomes of their action, such as the potential use of a system in conflict even when it is not designed for that. In particular, we contend that it is reasonably foreseeable that: (1) civilian AI systems will be applied to active conflict, including conflict support activities, (2) the use of civilian AI systems in conflict will impact applications of the law of armed conflict, and (3) crossover AI technology will be applied to conflicts that fall short of armed conflict. Given these reasonably foreseeably outcomes, we present three technically feasible actions that developers of civilian AIs can take to potentially mitigate their moral responsibility: (a) establishing systematic approaches to multi-perspective capability testing, (b) integrating digital watermarking in model weight matrices, and (c) utilizing monitoring and reporting mechanisms for conflict-related AI applications.