Research & Evaluation Correspondent
Mira Sethi focuses on how models are measured, compared and understood. Her coverage examines evaluation design, reproducibility and the assumptions behind claims of progress. She looks for the caveat that changes the headline.
711 published articles · Showing 41–60
The audio generation app Huxe, born from the minds of former NotebookLM developers, has ceased operations, pulling its application from both the App Store and Play Store today, with services slated to end by month's end TechCrunch. This shutdown underscores the harsh realities fa...
Read article →The push to infuse artificial intelligence into every facet of our digital lives is revealing its complex, often raw, realities this week. Finnish phone-maker HMD, a formidable challenger in a landscape dominated by giants, is making a shrewd move, strategically bundling a locali...
Read article →A flurry of new AI benchmarks, published just yesterday on arXiv CS. AI, signals a critical pivot in artificial intelligence development: a concerted effort to move beyond elementary tasks and tackle the nuanced, high-stakes challenges of the real world....
Read article →Flipper Devices, the visionary company behind the widely popular—and often debated—Flipper Zero, has unveiled its latest creation: the Flipper One. This new device emerges as a pocket-sized, open-source Linux computer, specifically engineered to empower the global community of ha...
Read article →A groundbreaking paper published on arXiv today introduces a novel approach to federated learning, enabling the interpretation of complex, interdependent temporal dynamics across distributed industrial systems without compromising data privacy. This advancement is a crucial step ...
Read article →The AI sector is demonstrating a remarkable duality this week, with established players like Anthropic projecting their first profitable quarter, while innovative seed-stage ventures such as Imperagen and Clouted successfully close significant funding rounds. This landscape under...
Read article →A new pre-print surfacing on arXiv today, May 21, 2026, illuminates a critical bottleneck for the burgeoning field of LLM-based multi-agent systems (MAS): the labor-intensive, brittle, and often rigid process of designing their architectures. This research, titled “Evolutionary G...
Read article →A torrent of groundbreaking research, published today on arXiv CS. LG, signals a critical leap in AI's ability to move beyond mere object recognition to genuinely understanding complex, real-world data....
Read article →A torrent of cutting-edge machine learning research, published on May 21, 2026, on arXiv CS. LG, signals a critical shift in the AI landscape: a concerted effort to move beyond the "black box" and build fundamentally more interpretable, robust, and human-aligned intelligent syste...
Read article →The world stands on the cusp of a profound societal shift driven by artificial intelligence, with Google set to launch a radically redesigned search experience fueled by "agentic AI" and the Vatican preparing its first papal encyclical explicitly addressing the technology—both cr...
Read article →Intuit, the financial software giant, is executing a dramatic corporate restructuring, laying off over 3,000 employees with the explicit goal of refocusing its operations squarely on artificial intelligence TechCrunch. This isn't merely an efficiency play; it’s a visceral commitm...
Read article →A wave of new research from arXiv CS. AI, published just yesterday, May 20, 2026, reveals a dual thrust in artificial intelligence development: a monumental stride towards unifying complex climate modeling with a novel foundation model and a pragmatic breakthrough for dynamic gam...
Read article →A pair of groundbreaking research papers, both published on arXiv on May 20, 2026, signal a significant leap forward in AI's ability to generate dynamic video and realistically animate diverse characters. These breakthroughs directly address long-standing limitations that have ch...
Read article →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...
Read article →The fight for transparency in artificial intelligence just got two potent new weapons. New research published on arXiv CS....
Read article →A flurry of new research from arXiv CS. AI reveals a crucial shift in the AI landscape: the rise of highly specialized foundation models designed to tackle some of humanity's most pressing and unpredictable challenges, from mitigating devastating wildfires to stabilizing our ener...
Read article →The landscape for Automatic Speech Recognition (ASR) just shifted. A new dataset, PAREDA (PAper REading DAtaset), has emerged as a critical tool for advancing ASR systems, specifically tackling the persistent challenge of real-world speech variability....
Read article →Cutting-edge research published today on arXiv is pushing the boundaries of generative AI, offering breakthroughs in both counterfactual video foley generation and cross-modal recommendation systems. These papers signal a critical advancement in how AI understands and manipulates...
Read article →A groundbreaking new research paper from arXiv outlines an "agentic pipeline" designed to revolutionize kinematic monitoring for spinal cord injury (SCI) rehabilitation patients, addressing long-standing challenges in deploying multi-view markerless motion capture systems in real...
Read article →A crucial moment for scientific and engineering AI has arrived with the unveiling of two pivotal research developments: the HiLiftAeroML dataset, the first-ever open-source high-fidelity computational fluid dynamics (CFD) dataset for high-lift aircraft, and FieldFormer, a novel t...
Read article →