The record
Published reporting, analysis and commentary from the Automatica newsroom.
New research from arXiv:2602. 03839v2 reveals a critical architectural characteristic within large language model (LLM) operations: approximately 99% of per-step weight updates become effectively invisible after standard BF16 casting during training and inference forward passes a
Federated learning, often presented as a solution for privacy in AI development, carries a subtle but significant risk: it can amplify existing societal inequalities. New research published on arXiv CS.
Well, folks, it seems AI isn't just for writing your terrible romance novels or suggesting another streaming show you'll never watch. According to a couple of fresh papers dropped on arXiv today, the machines are now cleaning up after scientists and figuring out how to make thems
Today, cutting-edge AI research dropping on arXiv is redefining the very foundations of how machines perceive and interact with our world, promising to unlock unprecedented efficiency and capabilities in 3D content generation, robotic navigation, and smart environments. These thr
Recent publications on arXiv CS. LG, dated May 20, 2026, detail significant research breakthroughs in artificial intelligence applied to medical imaging analysis, directly addressing critical challenges related to interpretability, fairness, and diagnostic precision.
Just as Utah greenlights a data center stretching across 40,000 acres, xAI has committed to purchasing an additional $2. 8 billion worth of natural gas turbines, according to a recent SpaceX IPO filing TechCrunch.
New academic research published on arXiv CS. LG on May 20, 2026, details advanced machine learning models designed to address complex problems within financial market operations and dynamic resource allocation.
Google has unveiled significant advancements aimed at making AI agent deployment dramatically simpler for developers while simultaneously articulating grand ambitions for AI's role in scientific discovery and health, all against a backdrop of intensifying capital expenditure acro
Enterprise AI agents are exhibiting critical failures in memory and operational stability, frequently "forgetting" learned sequences of actions, a vulnerability that undermines their utility and introduces unpredictable behavior into automated systems. This fundamental flaw compe
OpenAI has reportedly recommenced preparations for an Initial Public Offering (IPO) that may occur as early as September, a development following the swift resolution of the lawsuit brought by Elon Musk TechCrunch. This rapid resumption of financial strategizing underscores a ren
Artificial intelligence concepts are increasingly employed as a framework for understanding human behavior and cognition, a development that signifies a profound shift in humanity's self-interpretation and analytical methodologies arXiv CS. AI.
A new wave of research documents, published on arXiv this week, unveils an evolving frontier in AI: the very benchmarks designed to measure its capabilities. These are not merely academic exercises; they are the unseen blueprints for the automated entities that will increasingly
Cutting-edge research published today on arXiv introduces several advancements poised to democratize AI’s role in medical diagnosis and drug discovery, significantly addressing deployment challenges and data limitations. Among the most impactful is the UltraSeg family of models,
A new family of AI models, Stable Audio 3, has been introduced, designed to offer flexible, variable-length audio generation and editing capabilities. This development, detailed in a recent arXiv paper, aims to make creating and refining audio more efficient and less resource-int
A new wave of AI research is tackling the critical barriers to widespread clinical deployment of artificial intelligence in medicine, moving beyond GPU-dependent models and addressing issues of data scarcity and interpretability. Recent papers on arXiv highlight breakthroughs ran
The fight for transparency in artificial intelligence just got two potent new weapons. New research published on arXiv CS.
A recent publication on arXiv, titled “ContextFlow: Hierarchical Task-State Alignment for Long-Horizon Embodied Agents,” identifies and proposes a solution for a critical bottleneck in the development of sophisticated AI. The paper argues that as embodied agents increasingly dele
Large Language Models, once heralded as controlled and alignable, are proving deeply vulnerable to manipulation. New research published today on arXiv CS.
Another round of human resources being 'optimized' out of existence has begun, this time at Meta. The social media giant is reportedly preparing to cut approximately 8,000 jobs, a move that has employees scrambling to cash in their benefits before the axe falls Wired.
A new research framework, Mega-ASR, has been proposed to directly confront the "acoustic robustness bottleneck" that limits the reliability of automatic speech recognition (ASR) systems in complex, real-world operational environments arXiv CS. AI.