2026· International Journal of Artificial Intelligence, Data Science, and Machine Learning· Vol 7, pp. 216-219· 0 citations
TL;DR
It is argued that shifting from a human-centric to an agent-first paradigm is a prerequisite for unlocking the true scaling laws of decentralized machine intelligence.
Abstract
Modern data architecture is fundamentally constrained by a legacy assumption: that the ultimate con- sumer of information is a human being. Consequently, cur- rent paradigms disproportionately allocate computational resources to human-legible interfaces, rigid API schemas, and highly abstracted intermediate data states designed primarily for manual oversight and debugging. With the rapid ascendancy of autonomous agentic workflows, this human-centric design introduces severe inefficiencies in latency, throughput, and structural complexity. (1) This paper introduces an Agent-First Data Architecture (AFDA), a paradigm shift that re-engineers data stor- age, transport, and synthesis exclusively for machine-to- machine optimization. We explore the systematic disman- tling of the traditional visual application layer, demon- strating how autonomous agents render fixed user in- terfaces and rigid middleware pipelines obsolete through on-the-fly, task-specific data computation. Furthermore, we analyze the efficiencies gained by transitioning from human-readable protocols (such as JSON or XML) to non- human-readable intermediate states. Finally, we address the architectural flattening of the modern software stack and confront the emerging challenges of this shift, specif- ically the ”black box” debugging crisis and the necessity of specialized Observer Agents for forensic translation. Ultimately, we argue that shifting from a human-centric to an agent-first paradigm is a prerequisite for unlocking the true scaling laws of decentralized machine intelligence. (2)
This work presents DeepEye, a workflow-centric agentic data system that turns user intents into transparent and steerable analytical workflows and develops DataMagic as the system’s Video Generator, a declarative multi-agent method that improves data-video quality.
As large multimodal models move from understanding content to operating on digital environments, mobile GUI has emerged as a challenging and consequential testbed for digital embodied intelligence. Mobile agents operate under three coupled constraints: precise perception of complex interfaces, scalable acquisition of high-quality interaction data, and robust long-horizon decision making under compounding execution errors. This report presents HyMobileAgent, a mobile GUI agent built on Hy3.0-VL-A3B, a vision-native foundation model featuring native any-resolution input, an A3B-scale deployment budget, and a 32K context window to model extended interaction histories. Rather than relying solely on model scaling, we develop a joint data and environment centric scaling framework to address the key bottlenecks of mobile interaction. Our framework integrates a GUI perception flywheel combining mock-interface synthesis, rejection sampling, and icon-specific augmentation; a knowledge pipeline that transforms tutorial videos into structured interaction data; a million-scale action data pipeline deployed across more than 2000 sandbox and real-device instances with automated failure attribution; the PhoneWorld Mock App Factory, providing a resettable training environment with 34 mock applications and over 34000 tasks; and a structured Planning-and-Reflection mechanism with explicit dead-loop detection for reliable long-horizon execution. We also introduce a progressive training recipe consisting of mid-training, supervised fine-tuning, and reinforcement learning with task-specific reward designs.
Hy Vision Team, Huawen Shen, Zhengyang Tang et al.· arXiv.org· 1 citation
Comparisons against stronger model and coding-agent competitors further indicate that both domain-specific agent runtime structure and foundation-model strength matter for autonomous data analysis.
Mixed-initiative Visual Analytics (VA) systems empower human users by interleaving human intuition with software agents and their machine intelligence. However, the development and rigorous evaluation of such systems remain constrained by engineering overhead. Developers must, e.g., implement complex, low-level state synchronization to manage asynchronous agent behaviors, while researchers struggle to capture the multimodal provenance required to study and evaluate human-AI collaboration. We present MIVAIS, a dual-layered research platform designed to abstract the structural complexities of mixed-initiative VA. First, it contributes a computational Infrastructure that standardizes human-software agent interaction, state synchronization, and communication between the agents. Second, it provides a declarative Study Environment that automatically logs multimodal human-AI telemetry - including application/system state, screen capture, audio, and additional sensor data - enabling seamless, in-situ user studies and post-session analysis. We technically validate our infrastructure by replicating three state-of-the-art systems (Podium, Voyager 2, and ProactiveVA). Furthermore, we evaluate the framework's expressiveness and efficiency through expert case studies with HCI and VA researchers, demonstrating how MIVAIS effectively lowers the barrier to prototyping and evaluating intelligent, co-adaptive interfaces.
Tobias St\"ahle, Simon Schneider, Rita Sevastjanova et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
