Accurate and efficient global ocean state estimation remains a grand challenge for Earth system science, hindered by the dual bottlenecks of computational scalability and degraded data fidelity in traditional data assimilation (DA) and deep learning (DL) approaches. Here we present an AI-driven Data Assimilation Framework for Ocean (ADAF-Ocean) that directly assimilates multi-source and multi-scale observations, ranging from sparse in-situ measurements to 4 km satellite swaths, without any interpolation or data thinning. Inspired by Neural Processes, ADAF-Ocean learns a continuous mapping from heterogeneous inputs to ocean states, preserving native data fidelity. Through AI-driven super-resolution, it reconstructs 0.25$^\circ$ mesoscale dynamics from coarse 1$^\circ$ fields, which ensures both efficiency and scalability, with just 3.7\% more parameters than the 1$^\circ$ configuration. When coupled with a DL forecasting system, ADAF-Ocean extends global forecast skill by up to 20 days compared to baselines without assimilation. This framework establishes a computationally viable and scientifically rigorous pathway toward real-time, high-resolution Earth system monitoring.

Data science and engineering workflows often span multiple stages, from warehousing to orchestration, using tools like BigQuery, dbt, and Airbyte. As vision language models (VLMs) advance in multimodal understanding and code generation, VLM-based agents could potentially automate these workflows by generating SQL queries, Python code, and GUI operations. This automation can improve the productivity of experts while democratizing access to large-scale data analysis. In this paper, we introduce Spider2-V, the first multimodal agent benchmark focusing on professional data science and engineering workflows, featuring 494 real-world tasks in authentic computer environments and incorporating 20 enterprise-level professional applications. These tasks, derived from real-world use cases, evaluate the ability of a multimodal agent to perform data-related tasks by writing code and managing the GUI in enterprise data software systems. To balance realistic simulation with evaluation simplicity, we devote significant effort to developing automatic configurations for task setup and carefully crafting evaluation metrics for each task. Furthermore, we supplement multimodal agents with comprehensive documents of these enterprise data software systems. Our empirical evaluation reveals that existing state-of-the-art LLM/VLM-based agents do not reliably automate full data workflows (14.0% success). Even with step-by-step guidance, these agents still underperform in tasks that require fine-grained, knowledge-intensive GUI actions (16.2%) and involve remote cloud-hosted workspaces (10.6%). We hope that Spider2-V paves the way for autonomous multimodal agents to transform the automation of data science and engineering workflow. Our code and data are available at https://spider2-v.github.io.

Weather forecasting, a critical component in industries like transportation and manufacturing, relies heavily on Numerical Weather Prediction (NWP) systems, which are based on 3D physical models and dynamical equations [1, 2]. For NWP systems to predict future atmospheric states effectively, they require accurate current atmospheric states as initial values. This necessity underscores the importance of a data assimilation (DA) system, which approximates the true atmospheric states by merging observations with prediction results from dynamical models [3]. The integration of a wide range of observations, from sources like aircraft, radiosondes, and satellites, is crucial for enhancing the DA system's accuracy [4]. Traditional methods to assess the impact of observations on weather forecasts include forecast sensitivity to observation (FSO) and its variations, such as ensemble FSO and hybrid FSO [2, 5, 6].

Previous research into agent communication has shown that a pre-trained guide can speed up the learning process of an imitation learning agent. The guide achieves this by providing the agent with discrete messages in an emerged language about how to solve the task. We extend this one-directional communication by a one-bit communication channel from the learner back to the guide: It is able to ask the guide for help, and we limit the guidance by penalizing the learner for these requests. During training, the agent learns to control this gate based on its current observation. We find that the amount of requested guidance decreases over time and guidance is requested in situations of high uncertainty. We investigate the agent's performance in cases of open and closed gates and discuss potential motives for the observed gating behavior.

This paper uses Factored Latent Analysis (FLA) to learn a factorized, segmental representation for observations of tracked objects over time. Factored Latent Analysis is latent class analysis in which the observation space is subdivided and each aspect of the original space is represented by a separate latent class model. One could simply treat these factors as completely independent and ignore their interdependencies or one could concatenate them together and attempt to learn latent class structure for the complete observation space. Alternatively, FLA allows the interdependencies to be exploited in estimating an effective model, which is also capable of representing a factored latent state. In this paper, FLA is used to learn a set of factored latent classes to represent different modalities of observations of tracked objects. Different characteristics of the state of tracked objects are each represented by separate latent class models, including normalized size, normalized speed, normalized direction, and position. This model also enables effective temporal segmentation of these sequences. This method is data-driven, unsupervised using only pairwise observation statistics. This data-driven and unsupervised activity classi- fication technique exhibits good performance in multiple challenging environments.