As artificial intelligence (AI) technologies, particularly generative and collaborative learning models— are increasingly integrated into healthcare and other sensitive domains, data privacy, security, and fairness concerns have grown significantly. This paper focuses on a thorough examination of current privacy-preserving AI models, including federated learning (FL), differential privacy (DP), homomorphic encryption, and generative adversarial networks (GANs). Key contributions are reviewed across recent works that explore privacy-preserving mechanisms within domains such as clinical diagnostics, drug discovery, Internet of Medical Things […]

Ver mais

arXiv:2601.04282v1 Announce Type: new Abstract: Machine unlearning has become a crucial role in enabling generative models trained on large datasets to remove sensitive, private, or copyright-protected data. However, existing machine unlearning methods face three challenges in learning to forget identity of generative models: 1) inefficient, where identity erasure requires fine-tuning all the model’s parameters; 2) limited controllability, where forgetting intensity cannot be controlled and explainability is lacking; 3) catastrophic collapse, where the model’s retention capability undergoes drastic degradation […]

Ver mais

The AI arms race is a trillion-dollar game of follow-the-leader, and everyone’s running in the wrong direction. While OpenAI, Google, and Anthropic pour billions into making models 0.3% more accurate on benchmarks nobody cares about, a quiet revolution is brewing in Discord servers, GitHub repos, and weekend hack projects. These indie developers aren’t trying to build better AI. They’re building AI that actually matters. The Benchmark Trap Here’s what Big AI won’t tell you: the race to dominate […]

Ver mais

arXiv:2601.04283v1 Announce Type: new Abstract: Building on insights from the grokking literature, we study character-level Transformers trained to compute modular addition from text, and focus on robustness under input-format variation rather than only in-distribution accuracy. We identify a previously under-emphasized failure mode: models that achieve high in-distribution accuracy can fail catastrophically when the same expression is shifted to different absolute character positions (“position shift”) or presented under out-of-distribution natural-language templates. Using a disjoint-pair split over all ordered pairs […]

Ver mais

Pioneering Predictive Intelligence: How One Architect Transformed Manufacturing Asset Management Through IoT and Oracle Cloud Innovation In an era where unplanned downtime can cost manufacturers millions and operational resilience determines market leadership, the convergence of IoT technology with enterprise asset management represents a paradigm shift in how organizations approach maintenance intelligence. The groundbreaking IoT-enabled predictive maintenance solution architected by Ramachandra Handaragal stands as a watershed moment in manufacturing operations—the industry’s first fully integrated IoT-Oracle eAM platform that transformed […]

Ver mais

Motivated by recent work on the experts problem in the streaming model, we consider the experts problem in the sliding window model. The sliding window model is a well-studied model that captures applications such as traffic monitoring, epidemic tracking, and automated trading, where recent information is more valuable than older data. Formally, we have $n$ experts, $T$ days, the ability to query the predictions of $q$ experts on each day, a limited amount of memory, and should achieve […]

Ver mais

Ever since the AI race picked up the pace, we have been witnessing some landmark investments within the industry. SoftBank, a major Japanese multinational conglomerate headquartered in Tokyo, has been among the most popular names involved in such investments. Earlier this month, there were reports hinting that Soft Bank is coming together with NVIDIA to invest at least $15 billion into Skild AI, which is already a well-funded startup that makes foundation models for robots. SoftBank boosts its […]

Ver mais

[Updated on 2018-09-30: thanks to Yoonju, we have this post translated in Korean!] [Updated on 2019-04-18: this post is also available on arXiv.] Generative adversarial network (GAN) has shown great results in many generative tasks to replicate the real-world rich content such as images, human language, and music. It is inspired by game theory: two models, a generator and a critic, are competing with each other while making each other stronger at the same time. However, it is […]

Ver mais

This study develops an adaptive workflow allocation mechanism for anti-money laundering (AML) operations, aiming to improve the accuracy and efficiency of suspicious-transaction review. A multi-agent simulation platform was constructed to model transaction flows, alert generation, and analyst decision behaviors. The system integrates model-confidence estimation, analyst-fatigue prediction, and real-time workload signals to dynamically route alerts. Experiments were conducted using 27.3 million historical transactions and 186,000 alerts from a large commercial financial dataset. Compared with fixed allocation rules, the adaptive […]

Ver mais