The UK should concentrate on setting global standards for artificial intelligence testing instead of trying to carry out all the vetting itself, according to a company assisting the government's AI Safety Institute. Marc Warner, the chief executive of Faculty AI, said the newly established institute could end up "on the hook" for scrutinising an array of AI models – the technology that underpins chatbots like ChatGPT – owing to the government's world-leading work in AI safety. Rishi Sunak announced the formation of the AI Safety Institute (AISI) last year ahead of the global AI safety summit, which secured a commitment from big tech companies to cooperate with the EU and 10 countries, including the UK, US, France and Japan, on testing advanced AI models before and after their deployment. The UK has a prominent role in the agreement because of its advanced work on AI safety, underlined by the establishment of the institute. Warner, whose London-based company has contracts with the UK institute that include helping it test AI models on whether they can be prompted to breach their own safety guidelines, said the institute should be a world leader in setting test standards.

The frequent occurrence of cyber-attacks has made webshell attacks and defense gradually become a research hotspot in the field of network security. However, the lack of publicly available benchmark datasets and the over-reliance on manually defined rules for webshell escape sample generation have slowed down the progress of research related to webshell escape sample generation strategies and artificial intelligence-based webshell detection algorithms. To address the drawbacks of weak webshell sample escape capabilities, the lack of webshell datasets with complex malicious features, and to promote the development of webshell detection technology, we propose the Hybrid Prompt algorithm for webshell escape sample generation with the help of large language models. As a prompt algorithm specifically developed for webshell sample generation, the Hybrid Prompt algorithm not only combines various prompt ideas including Chain of Thought, Tree of Thought, but also incorporates various components such as webshell hierarchical module and few-shot example to facilitate the LLM in learning and reasoning webshell escape strategies. Experimental results show that the Hybrid Prompt algorithm can work with multiple LLMs with excellent code reasoning ability to generate high-quality webshell samples with high Escape Rate (88.61% with GPT-4 model on VIRUSTOTAL detection engine) and Survival Rate (54.98% with GPT-4 model).

This study evaluates the ability of Large Language Model (LLM)-based Subpopulation Representative Models (SRMs) to generalize from empirical data, utilizing in-context learning with data from the 2016 and 2020 American National Election Studies. We explore generalization across response variables and demographic subgroups. While conditioning with empirical data improves performance on the whole, the benefit of in-context learning varies considerably across demographics, sometimes hurting performance for one demographic while helping performance for others. The inequitable benefits of in-context learning for SRM present a challenge for practitioners implementing SRMs, and for decision-makers who might come to rely on them. Our work highlights a need for fine-grained benchmarks captured from diverse subpopulations that test not only fidelity but generalization.

Recent work has shown the defense of 01 loss sign activation neural networks against image classification adversarial attacks. A public challenge to attack the models on CIFAR10 dataset remains undefeated. We ask the following question in this study: are 01 loss sign activation neural networks hard to deceive with a popular black box text adversarial attack program called TextFooler? We study this question on four popular text classification datasets: IMDB reviews, Yelp reviews, MR sentiment classification, and AG news classification. We find that our 01 loss sign activation network is much harder to attack with TextFooler compared to sigmoid activation cross entropy and binary neural networks. We also study a 01 loss sign activation convolutional neural network with a novel global pooling step specific to sign activation networks. With this new variation we see a significant gain in adversarial accuracy rendering TextFooler practically useless against it. We make our code freely available at \url{https://github.com/zero-one-loss/wordcnn01} and \url{https://github.com/xyzacademic/mlp01example}. Our work here suggests that 01 loss sign activation networks could be further developed to create fool proof models against text adversarial attacks.

Optimal allocation of agricultural water in the event of droughts is an important global problem. In addressing this problem, many aspects, including the welfare of farmers, the economy, and the environment, must be considered. Under this backdrop, our work focuses on several resource-matching problems accounting for agents with multi-crop portfolios, geographic constraints, and fairness. First, we address a matching problem where the goal is to maximize a welfare function in two-sided markets where buyers' requirements and sellers' supplies are represented by value functions that assign prices (or costs) to specified volumes of water. For the setting where the value functions satisfy certain monotonicity properties, we present an efficient algorithm that maximizes a social welfare function. When there are minimum water requirement constraints, we present a randomized algorithm which ensures that the constraints are satisfied in expectation. For a single seller--multiple buyers setting with fairness constraints, we design an efficient algorithm that maximizes the minimum level of satisfaction of any buyer. We also present computational complexity results that highlight the limits on the generalizability of our results. We evaluate the algorithms developed in our work with experiments on both real-world and synthetic data sets with respect to drought severity, value functions, and seniority of agents.

While physics-informed neural networks (PINNs) have become a popular deep learning framework for tackling forward and inverse problems governed by partial differential equations (PDEs), their performance is known to degrade when larger and deeper neural network architectures are employed. Our study identifies that the root of this counter-intuitive behavior lies in the use of multi-layer perceptron (MLP) architectures with non-suitable initialization schemes, which result in poor trainablity for the network derivatives, and ultimately lead to an unstable minimization of the PDE residual loss. To address this, we introduce Physics-informed Residual Adaptive Networks (PirateNets), a novel architecture that is designed to facilitate stable and efficient training of deep PINN models. PirateNets leverage a novel adaptive residual connection, which allows the networks to be initialized as shallow networks that progressively deepen during training. We also show that the proposed initialization scheme allows us to encode appropriate inductive biases corresponding to a given PDE system into the network architecture. We provide comprehensive empirical evidence showing that PirateNets are easier to optimize and can gain accuracy from considerably increased depth, ultimately achieving state-of-the-art results across various benchmarks. All code and data accompanying this manuscript will be made publicly available at \url{https://github.com/PredictiveIntelligenceLab/jaxpi}.

In social choice theory with ordinal preferences, a voting method satisfies the axiom of positive involvement if adding to a preference profile a voter who ranks an alternative uniquely first cannot cause that alternative to go from winning to losing. In this note, we prove a new impossibility theorem concerning this axiom: there is no ordinal voting method satisfying positive involvement that also satisfies the Condorcet winner and loser criteria, resolvability, and a common invariance property for Condorcet methods, namely that the choice of winners depends only on the ordering of majority margins by size.

The groundbreaking artificial intelligence image-generating company Midjourney is considering banning people from using its software to make political images of Joe Biden and Donald Trump as part of an effort to avoid being used to distract from or misinform about the 2024 US presidential election. "I don't know how much I care about political speech for the next year for our platform," Midjourney's CEO, David Holz, said last week, adding that the company is close to "hammering" – or banning – political images, including those of the leading presidential candidates, "for the next 12 months". In a conversation with Midjourney users in a chatroom on Discord, as reported by Bloomberg, Holz went on to say: "I know it's fun to make Trump pictures – I make Trump pictures. Trump is aesthetically really interesting. However, probably better to just not, better to pull out a little bit during this election.

From The Terminator to the The Matrix, killer robots have long been a terrifying staple of science-fiction flicks. But, while they might be scare-worthy in the cinema, should we really be afraid of a big bad AI? From supercharged plagues to full-blown nuclear annihilation, experts say there are five ways AI could bring about the end of humanity. Ben Eisenpress, Director of Operations at the Future of Life Institute, warned MailOnline that'all catastrophic risks from AI are currently underestimated.' So, if you still think the AI-apocalypse is nothing more than an outdated movie trope, read on to see just how worried you should really be. When you think about AI leading to the destruction of humanity, killer robots are most likely what you have in mind.

Even in a booming seller's market, real estate remains an outstanding investment, even for those who are new to it. But you want to make sure you're making the right decisions, and that's where a tool like Mashvisor comes in handy. Mashvisor helps you perform smarter property analysis with accurate, real-time real estate market data sourced from the MLS, Zillow, Rentometer, Airbnb.com, and the US Census Bureau. Mashvisor's AI and machine learning algorithms turn the raw data from these sources into actionable analytics, helping you find markets and properties in your budget with high investment return potential. Mashvisor covers 95% of US markets, allowing you to cast a wide net in your investment search.