--Graph Neural Networks (GNNs) have achieved significant success in machine learning, with wide applications in social networks, bioinformatics, knowledge graphs, and other fields. Most research assumes ideal closed-set environments. However, in real-world open-set environments, graph learning models face challenges in robustness and reliability due to unseen classes. This highlights the need for Graph Open-Set Recognition (GOSR) methods to address these issues and ensure effective GNN application in practical scenarios. Research in GOSR is in its early stages, with a lack of a comprehensive benchmark spanning diverse tasks and datasets to evaluate methods. Moreover, traditional methods, Graph Out-of-Distribution Detection (GOODD), GOSR, and Graph Anomaly Detection (GAD) have mostly evolved in isolation, with little exploration of their interconnections or potential applications to GOSR. T o fill these gaps, we introduce G-OSR, a comprehensive benchmark for evaluating GOSR methods at both the node and graph levels, using datasets from multiple domains to ensure fair and standardized comparisons of effectiveness and efficiency across traditional, GOODD, GOSR, and GAD methods. The results offer critical insights into the generalizability and limitations of current GOSR methods and provide valuable resources for advancing research in this field through systematic analysis of diverse approaches. RAPH learning, as a significant research direction in machine learning, has been widely applied in social network analysis, recommendation systems, bioinformatics, knowledge graphs, traffic planning, and the fields of chemistry and materials science [1]. Graph Neural Networks (GNNs) have demonstrated superior performance in various node classification and graph classification tasks [2]. These methods typically follow a closed-set setting, which assumes that all test classes are among the seen classes accessible during training [3]. However, in real-world scenarios, due to undersampling, out-of-distribution, or anomalous samples, it is highly likely to encounter samples belonging to novel unseen classes, which can significantly impact the safety and robustness of models [4], as illustrated in Figure 1. Guangyao Chen is with Cornell University, Ithaca, NY, USA. Wentao Zhang is with Peking University, Beijing, China. Zhongyi Han is with King Abdullah University of Science and Technology, Thuwal, Saudi Arabia. Rundong He and Yilong Yin are the corresponding authors. Closed-set classification cannot identify unseen classes, while open-set recognition can identify unseen classes and classify nodes belonging to seen classes.

Large Language Models (LLMs) have shown impressive performance across a wide array of tasks involving both structured and unstructured textual data. Recent results on various benchmarks for code generation, repair, or completion suggest that certain models have programming abilities comparable to or even surpass humans. In this work, we demonstrate that high performance on such benchmarks does not correlate to humans' innate ability to understand structural control flow in code. To this end, we extract solutions from the HumanEval benchmark, which the relevant models perform strongly on, and trace their execution path using function calls sampled from the respective test set. Using this dataset, we investigate the ability of seven state-of-the-art LLMs to match the execution trace and find that, despite their ability to generate semantically identical code, they possess limited ability to trace execution paths, especially for longer traces and specific control structures. We find that even the top-performing model, Gemini, can fully and correctly generate only 47% of HumanEval task traces. Additionally, we introduce a subset for three key structures not contained in HumanEval: Recursion, Parallel Processing, and Object-Oriented Programming, including concepts like Inheritance and Polymorphism. Besides OOP, we show that none of the investigated models achieve an accuracy over 5% on the relevant traces. Aggregating these specialized parts with HumanEval tasks, we present CoCoNUT: Code Control Flow for Navigation Understanding and Testing, which measures a model's ability to trace execution of code upon relevant calls, including advanced structural components. We conclude that current LLMs need significant improvement to enhance code reasoning abilities. We hope our dataset helps researchers bridge this gap.

Deep learning (DL) models have emerged as a powerful tool in avian bioacoustics to assess environmental health. To maximize the potential of cost-effective and minimal-invasive passive acoustic monitoring (PAM), DL models must analyze bird vocalizations across a wide range of species and environmental conditions. However, data fragmentation challenges a comprehensive evaluation of generalization performance. Therefore, we introduce the BirdSet dataset, comprising approximately 520,000 global bird recordings for training and over 400 hours of PAM recordings for testing. Our benchmark offers baselines for several DL models to enhance comparability and consolidate research across studies, along with code implementations that include comprehensive training and evaluation protocols.

Entropy measures quantify the amount of information and correlations present in a quantum system. In practice, when the quantum state is unknown and only copies thereof are available, one must resort to the estimation of such entropy measures. Here we propose a variational quantum algorithm for estimating the von Neumann and R\'enyi entropies, as well as the measured relative entropy and measured R\'enyi relative entropy. Our approach first parameterizes a variational formula for the measure of interest by a quantum circuit and a classical neural network, and then optimizes the resulting objective over parameter space. Numerical simulations of our quantum algorithm are provided, using a noiseless quantum simulator. The algorithm provides accurate estimates of the various entropy measures for the examples tested, which renders it as a promising approach for usage in downstream tasks.

1. Citizen and community-science (CS) datasets have great potential for estimating interannual patterns of population change given the large volumes of data collected globally every year. Yet, the flexible protocols that enable many CS projects to collect large volumes of data typically lack the structure necessary to keep consistent sampling across years. This leads to interannual confounding, as changes to the observation process over time are confounded with changes in species population sizes. 2. Here we describe a novel modeling approach designed to estimate species population trends while controlling for the interannual confounding common in citizen science data. The approach is based on Double Machine Learning, a statistical framework that uses machine learning methods to estimate population change and the propensity scores used to adjust for confounding discovered in the data. Additionally, we develop a simulation method to identify and adjust for residual confounding missed by the propensity scores. Using this new method, we can produce spatially detailed trend estimates from citizen science data. 3. To illustrate the approach, we estimated species trends using data from the CS project eBird. We used a simulation study to assess the ability of the method to estimate spatially varying trends in the face of real-world confounding. Results showed that the trend estimates distinguished between spatially constant and spatially varying trends at a 27km resolution. There were low error rates on the estimated direction of population change (increasing/decreasing) and high correlations on the estimated magnitude. 4. The ability to estimate spatially explicit trends while accounting for confounding in citizen science data has the potential to fill important information gaps, helping to estimate population trends for species, regions, or seasons without rigorous monitoring data.

Mintlify, a startup developing software to automate software documentation tasks, today announced that it raised $2.8 million in a seed round led by by Bain Capital Ventures with participation from TwentyTwo Ventures and Quinn Slack, Sourcegraph's co-founder. CEO Han Wang says that the proceeds will be put toward product development and doubling Mintlify's core, three-person team by the end of the year. Ithaca, New York-based Mintlify was co-founded in 2021 by Han Wang and Hahnbee Lee -- both software engineers by trade. Wang previously co-launched Foodful, an startup that developed a cloud-based monitoring system for cows, and Pe•ple, an online customer community platform that was acquired by Tribe in early 2021. Lee was a co-founder at Pe•ple before briefly joining Duolingo as an engineer.

A touch-sensitive glove made from stretchable fibre-optic sensors could be used in robotics, sport and medicine. "We made a sensor that can sense haptic interactions, in the same way that our own skin sensors interact with [the] environment," says Hedan Bai at Cornell University in Ithaca, New York. Bai and her team created the glove using optical fibres made from thin elastomeric polyurethane cables that transmit light from an LED. The light is interrupted when the cables are bent, stretched or put under pressure. The team dyed parts of the fibres with different colours, meaning that as they are distorted, the colour of light coming out of the fibres changes.

Until recently, the field of plant breeding looked a lot like it did in centuries past. A breeder might examine, for example, which tomato plants were most resistant to drought and then cross the most promising plants to produce the most drought-resistant offspring. This process would be repeated, plant generation after generation, until, over the course of roughly seven years, the breeder arrived at what seemed the optimal variety. Researchers at ETH Zürich use standard color images and thermal images collected by drone to determine how plots of wheat with different genotypes vary in grain ripeness. Image credit: Norbert Kirchgessner (ETH Zürich, Zürich, Switzerland). Now, with the global population expected to swell to nearly 10 billion by 2050 (1) and climate change shifting growing conditions (2), crop breeder and geneticist Steven Tanksley doesn’t think plant breeders have that kind of time. “We have to double the productivity per acre of our major crops if we’re going to stay on par with the world’s needs,” says Tanksley, a professor emeritus at Cornell University in Ithaca, NY. To speed up the process, Tanksley and others are turning to artificial intelligence (AI). Using computer science techniques, breeders can rapidly assess which plants grow the fastest in a particular climate, which genes help plants thrive there, and which plants, when crossed, produce an optimum combination of genes for a given location, opting for traits that boost yield and stave off the effects of a changing climate. Large seed companies in particular have been using components of AI for more than a decade. With computing power rapidly advancing, the techniques are now poised to accelerate breeding on a broader scale. AI is not, however, a panacea. Crop breeders still grapple with tradeoffs such as higher yield versus marketable appearance. And even the most sophisticated AI …

A wearable device consisting of two mini-cameras mounted on earphones can recognise your facial expressions and read your lips, even if your mouth is covered. The tool – called C-Face – was developed by Cheng Zhang at Cornell University in Ithaca, New York, and his colleagues. It looks at the sides of the wearer's head and uses machine learning to accurately visualise facial expressions by analysing small changes in cheek contour lines. "With previous technology to reconstruct facial expression, you had to put a camera in front of you. But that brings a lot of limitations," says Zhang. "Right now, many people are wearing a face mask, and standard facial tracking will not work. Our technology still works because it doesn't rely on what your face looks like."

Jenny Sabin is perched high on a scissor lift, her head poking through an opening of the porous fabric structure that she's struggling to stretch onto the exoskeleton of her installation piece, which is suspended in the airy atrium of building 99 on Microsoft's Redmond, Washington, campus. Momentarily defeated, she pauses and looks up. "It's going to be gorgeous," she says. "It" is a glowing, translucent and ethereal pavilion that Sabin and her Microsoft collaborators describe as both a research tool and a glimpse into a future in which architecture and artificial intelligence merge. "To my knowledge, this installation is the first architectural structure to be driven by artificial intelligence in real time," said Sabin, principal designer at Jenny Sabin Studio in Ithaca, New York, who designed and built the pavilion as part of Microsoft's Artist in Residence program.