Abstract
Economic complexity - a group of dimensionality-reduction methods that apply network science to trade data - represented a paradigm shift in development economics towards materializing the once-intangible concept of capabilities as inferrable and quantifiable. Measures such as the Economic Complexity Index (ECI) and the Product Space have proven their worth as robust estimators of an economy's subsequent growth; less obvious, however, is how they have come to be so. Despite ECI drawing its micro-foundations from a combinatorial model of capabilities, where a set of homogeneous capabilities combine to form products and the economies which can produce them, such a model is consistent with neither the fact that distinct product classes draw on distinct capabilities, nor the interrelations between different products in the Product Space which so much of economic complexity is based upon. In this paper, we extend the combinatorial model of economic complexity through two innovations: an underlying network which governs the relatedness between capabilities, and a production function which trades the original binary specialization function for a fine-grained, product-level output function. Using country-product trade data across 216 countries, 5000 products and two decades, we show that this model is able to accurately replicate both the characteristic topology of the Product Space and the complexity distribution of countries' export baskets. In particular, the model bridges the gap between the ECI and capabilities by transforming measures of economic complexity into direct measures of the capabilities held by an economy - a transformation shown to both improve the informativeness of the Economic Complexity Index in predicting economic growth and enable an interpretation of economic complexity as a proxy for productive structure in the form of capability substitutability.

Abstract
The US Physics community completed the Snowmass planning process in 2022, culminating in the HEPAP Particle Physics Project Prioritization Panel (P5) publishing its summary report at the end of 2023. Building on this, the US Magnet Development Program, a national accelerator magnet R&D program established by DOE-OHEP in 2016, has updated its strategic plan to align with the 2023 P5 report, resulting in this roadmap document.

Abstract
Imagine decision-makers uploading data and, within minutes, receiving clear, actionable insights delivered straight to their fingertips. That is the promise of the AI Data Scientist, an autonomous Agent powered by large language models (LLMs) that closes the gap between evidence and action. Rather than simply writing code or responding to prompts, it reasons through questions, tests ideas, and delivers end-to-end insights at a pace far beyond traditional workflows. Guided by the scientific tenet of the hypothesis, this Agent uncovers explanatory patterns in data, evaluates their statistical significance, and uses them to inform predictive modeling. It then translates these results into recommendations that are both rigorous and accessible. At the core of the AI Data Scientist is a team of specialized LLM Subagents, each responsible for a distinct task such as data cleaning, statistical testing, validation, and plain-language communication. These Subagents write their own code, reason about causality, and identify when additional data is needed to support sound conclusions. Together, they achieve in minutes what might otherwise take days or weeks, enabling a new kind of interaction that makes deep data science both accessible and actionable.

Abstract
The learning from practice paradigm is crucial for developing capable Agentic AI systems, yet it is severely hampered by inefficient experience generation, a bottleneck especially pronounced in complex benchmarks like GAIA. To address this, we introduce AWorld, an open-source system engineered for large-scale agent-environment interaction. By distributing tasks across a cluster, AWorld accelerates experience collection by 14.6x compared to standard single-node, sequential execution. This critical speedup makes extensive reinforcement learning practical and scalable. Leveraging this capability, we trained a Qwen3-32B-based agent that significantly outperforms its base model, increasing its overall GAIA accuracy from 21.59% to 32.23%. On the benchmark's most challenging levels, our agent achieves a score of 16.33%, surpassing the performance of leading proprietary models. Our open-source system and resulting agent provide a practical blueprint for a complete agentic AI training pipeline, from efficient interaction to demonstrable model improvement.

Abstract
Many technology companies aim to improve access and inclusion not only by making their products accessible but also by bringing people with disabilities into the tech workforce. We know less about how accessibility is experienced and negotiated by disabled workers within these organizations. Through interviews with 20 BLV workers across various tech companies, we uncover a persistent misalignment between organizational attempts at accessibility and the current realities of these employees. We introduce the concept of the accessibility paradox, which we define as the inherent tension between the productivity- and profit-driven nature of tech companies and their desire to hire and retain disabled workers. Focusing on the experiences of BLV workers, we show how the accessibility paradox manifests in their everyday workplace interactions, including digital infrastructure, accommodations processes and policies, ability assumptions, and competing priorities. We offer recommendations for future research and practice to understand and improve workplace accessibility and inclusion.

Abstract
Optical see-through augmented reality (OST-AR) systems like Microsoft HoloLens 2 hold promise for arm's distance guidance (e.g., surgery), but depth perception of the hologram and occlusion of real instruments remain challenging. We present an evaluation of how visualizing the target object with different transparencies and visualizing a tracked tool (virtual proxy vs. real tool vs. no tool tracking) affects depth perception and system usability. Ten participants performed two experiments on HoloLens 2. In Experiment 1, we compared high-transparency vs. low-transparency target rendering in a depth matching task at arm's length. In Experiment 2, participants performed a simulated surgical pinpoint task on a frontal bone target under six visualization conditions ($2 \times 3$: two target transparencies and three tool visualization modes: virtual tool hologram, real tool, or no tool tracking). We collected data on depth matching error, target localization error, system usability, task workload, and qualitative feedback. Results show that a more opaque target yields significantly lower depth estimation error than a highly transparent target at arm's distance. Moreover, showing the real tool (occluding the virtual target) led to the highest accuracy and usability with the lowest workload, while not tracking the tool yielded the worst performance and user ratings. However, making the target highly transparent, while allowing the real tool to remain visible, slightly impaired depth cues and did not improve usability. Our findings underscore that correct occlusion cues, rendering virtual content opaque and occluding it with real tools in real time, are critical for depth perception and precision in OST-AR. Designers of arm-distance AR systems should prioritize robust tool tracking and occlusion handling; if unavailable, cautiously use transparency to balance depth perception and tool visibility.

Abstract
This study addresses critical industrial challenges in e-commerce product categorization, namely platform heterogeneity and the structural limitations of existing taxonomies, by developing and deploying a multimodal hierarchical classification framework. Using a dataset of 271,700 products from 40 international fashion e-commerce platforms, we integrate textual features (RoBERTa), visual features (ViT), and joint vision--language representations (CLIP). We investigate fusion strategies, including early, late, and attention-based fusion within a hierarchical architecture enhanced by dynamic masking to ensure taxonomic consistency. Results show that CLIP embeddings combined via an MLP-based late-fusion strategy achieve the highest hierarchical F1 (98.59\%), outperforming unimodal baselines. To address shallow or inconsistent categories, we further introduce a self-supervised ``product recategorization'' pipeline using SimCLR, UMAP, and cascade clustering, which discovered new, fine-grained categories (e.g., subtypes of ``Shoes'') with cluster purities above 86\%. Cross-platform experiments reveal a deployment-relevant trade-off: complex late-fusion methods maximize accuracy with diverse training data, while simpler early-fusion methods generalize more effectively to unseen platforms. Finally, we demonstrate the framework's industrial scalability through deployment in EURWEB's commercial transaction intelligence platform via a two-stage inference pipeline, combining a lightweight RoBERTa stage with a GPU--accelerated multimodal stage to balance cost and accuracy.