Geometric trees are characterized by their tree-structured layout and spatially constrained nodes and edges, which significantly impacts their topological attributes. This inherent hierarchical structure plays a crucial role in domains such as neuron morphology and river geomorphology, but traditional graph representation methods often overlook these specific characteristics of tree structures. To address this, we introduce a new representation learning framework tailored for geometric trees. It first features a unique message passing neural network, which is both provably geometrical structure-recoverable and rotation-translation invariant. To address the data label scarcity issue, our approach also includes two innovative training targets that reflect the hierarchical ordering and geometric structure of these geometric trees. This enables fully self-supervised learning without explicit labels. We validate our method's effectiveness on eight real-world datasets, demonstrating its capability to represent geometric trees.

Data-driven techniques are being increasingly applied to complement physics-based models in fire science. However, the lack of sufficiently large datasets continues to hinder the application of certain machine learning techniques. In this paper, we use simulated data to investigate the ability of neural networks to parameterize dynamics in fire science. In particular, we investigate neural networks that map five key parameters in fire spread to the first arrival time, and the corresponding inverse problem. By using simulated data, we are able to characterize the error, the required dataset size, and the convergence properties of these neural networks. For the inverse problem, we quantify the network's sensitivity in estimating each of the key parameters. The findings demonstrate the potential of machine learning in fire science, highlight the challenges associated with limited dataset sizes, and quantify the sensitivity of neural networks to estimate key parameters governing fire spread dynamics.

Recent studies demonstrate that prompting an appropriate role-playing persona to an LLM improves its reasoning capability. However, assigning a proper persona is difficult since an LLM's performance is extremely sensitive to assigned prompts; therefore, personas sometimes hinder LLMs and degrade their reasoning capabilities. In this paper, we propose a novel framework, Jekyll \& Hyde, which ensembles the results of role-playing and neutral prompts to eradicate performance degradation via unilateral use of role-playing prompted LLM and enhance the robustness of an LLM's reasoning ability. Specifically, Jekyll \& Hyde collects two potential solutions from both role-playing and neutral prompts and selects a better solution after cross-checking via an LLM evaluator. However, LLM-based evaluators tend to be affected by the order of those potential solutions within the prompt when selecting the proper solution; thus, we also propose a robust LLM evaluator to mitigate the position bias. The experimental analysis demonstrates that role-playing prompts distract LLMs and degrade their reasoning abilities in 4 out of 12 datasets, even when using GPT-4. In addition, we reveal that Jekyll \& Hyde improves reasoning capabilities by selecting better choices among the potential solutions on twelve widely-used reasoning datasets. We further show that our proposed LLM evaluator outperforms other baselines, proving the LLMs' position bias is successfully mitigated.

Decentralized optimization has become a standard paradigm for solving large-scale decision-making problems and training large machine learning models without centralizing data. However, this paradigm introduces new privacy and security risks, with malicious agents potentially able to infer private data or impair the model accuracy. Over the past decade, significant advancements have been made in developing secure decentralized optimization and learning frameworks and algorithms. This survey provides a comprehensive tutorial on these advancements. We begin with the fundamentals of decentralized optimization and learning, highlighting centralized aggregation and distributed consensus as key modules exposed to security risks in federated and distributed optimization, respectively. Next, we focus on privacy-preserving algorithms, detailing three cryptographic tools and their integration into decentralized optimization and learning systems. Additionally, we examine resilient algorithms, exploring the design and analysis of resilient aggregation and consensus protocols that support these systems. We conclude the survey by discussing current trends and potential future directions.

Large Transformer networks are increasingly used in settings where low inference latency can improve the end-user experience and enable new applications. However, autoregressive inference is resource intensive and requires parallelism for efficiency. Parallelism introduces collective communication that is both expensive and represents a phase when hardware resources are underutilized. Towards mitigating this, Kraken is an evolution of the standard Transformer architecture that is designed to complement existing tensor parallelism schemes for efficient inference on multi-device systems. By introducing a fixed degree of intra-layer model parallelism, the architecture allows collective operations to be overlapped with compute, decreasing latency and increasing hardware utilization. When trained on OpenWeb-Text, Kraken models reach a similar perplexity as standard Transformers while also preserving their language modeling capabilities when evaluated on the SuperGLUE benchmark. Importantly, when tested on multi-GPU systems using TensorRT-LLM engines, Kraken speeds up Time To First Token by a mean of 35.6% across a range of model sizes, context lengths, and degrees of tensor parallelism.

As the US government weighs its options following a landmark "monopolist" ruling against Google last week, online publications increasingly face a bleak future. Bloomberg reports that their choice now boils down to allowing Google to use their published content to produce inline AI-generated search "answers" or losing visibility in the company's search engine. The crux of the problem lies in the Googlebot, the crawler that scours and indexes the live web to produce the results you see when you enter search terms. If publishers block Google from using their content for the AI-produced answers you now see littered at the top of many search results, they also lose the privilege of appearing in other Google search programs like snippets and Discover. Google uses a separate crawler for its Gemini (formerly Bard) chatbot, but its AI Overviews are generated using data from its main crawler.

Russia is putting generative artificial intelligence to work in online deception campaigns, but its efforts have been unsuccessful, according to a Meta security report released on Thursday. The parent company of Facebook and Instagram found that so far AI-powered tactics "provide only incremental productivity and content-generation gains" for bad actors and Meta has been able to disrupt deceptive influence operations. Meta's efforts to combat "coordinated inauthentic behavior" on its platforms come as fears mount that generative AI will be used to trick or confuse people in elections in the United States and other countries. Russia remains the top source of "coordinated inauthentic behavior" using bogus Facebook and Instagram accounts, David Agranovich, Meta's security policy director, told reporters. Since Russia's invasion of Ukraine in 2022, those efforts have been concentrated on undermining Ukraine and its allies, according to the report.

Anti-vaccine activist Robert F. Kennedy Jr.'s presidential campaign hosted an online panel Wednesday on the future of AI moderated, for some reason, by Ian Carroll, a self-styled journalist with a history of antisemitic statements. In the course of the conversation, Kennedy admitted that he "gets manipulated by AI all the time." "Somebody will send me something and I'll go'Holy cow, did you see this?'," he said, describing how he credulously forwards fake content to his children, only for them to have to correct him. RFK Jr. said he regularly "gets manipulated by AI." While Carroll has no particular public profile on AI, his persona tracks with the campaign's focus on tech figures and influencers as it courts a young, male, and extremely online audience.

Update, August 15, 5:50PM ET: The US Department of Homeland Security has issued a statement disputing some of MIT Technology Review's reporting. We've updated our post below with its statement and more details. The US Department of Homeland Security (DHS), which is looking to improve its facial recognition algorithms, is reportedly planning to use the facial data of migrant children entering the country for training. According to MIT Technology Review, the agency intends to collect and analyze facial captures of kids younger than 14. John Boyd, the assistant director of Homeland Security's Office of Biometric Identity Management who's involved in the development of biometric services for the government, told the publication that the collection will include children "down to the infant."

The US Department of Homeland Security (DHS) plans to collect and analyze photos of the faces of migrant children at the border in a bid to improve facial recognition technology, MIT Technology Review can reveal. The technology has traditionally not been applied to children, largely because training data sets of real children's faces are few and far between, and consist of either low-quality images drawn from the internet or small sample sizes with little diversity. Such limitations reflect the significant sensitivities regarding privacy and consent when it comes to minors. In practice, the new DHS plan could effectively solve that problem. But, beyond concerns about privacy, transparency, and accountability, some experts also worry about testing and developing new technologies using data from a population that has little recourse to provide--or withhold--consent. What Japan's "megaquake" warning really tells us On August 8, at 16:42 local time, a magnitude-7.1 earthquake shook southern Japan.