Fine-tuning Large Language Models (LLMs) for specialized domains is constrained by a fundamental challenge: the need for diverse, cross-organizational data conflicts with the principles of data privacy and sovereignty. While Federated Learning (FL) provides a framework for collaboration without raw data exchange, its classic centralized form introduces a single point of failure and remains vulnerable to model inversion attacks. Decentralized FL (DFL) mitigates this risk by removing the central aggregator but typically relies on inefficient, random peer-to-peer (P2P) […]

Ver mais

arXiv:2603.02214v1 Announce Type: new Abstract: Federated Inference (FI) studies how independently trained and privately owned models can collaborate at inference time without sharing data or model parameters. While recent work has explored secure and distributed inference from disparate perspectives, a unified abstraction and system-level understanding of FI remain lacking. This paper positions FI as a distinct collaborative paradigm, complementary to federated learning, and identifies two fundamental requirements that govern its feasibility: inference-time privacy preservation and meaningful performance gains […]

Ver mais

Google’s journey has been one of the most influential ones in the fast-evolving field of artificial intelligence. The company has come a long way from Bard to the highly developed Gemini AI models and has become an AI-first giant instead of just being a search-centric one. The Gemini AI Timeline over the years shows how AI has progressed from giving simple text feedback to its now multimodal reasoning ability, which supports a variety of applications from programming to […]

Ver mais

arXiv:2602.10464v1 Announce Type: cross Abstract: Recent advances in artificial intelligence have enabled the generation of large-scale, low-cost predictions with increasingly high fidelity. As a result, the primary challenge in statistical inference has shifted from data scarcity to data reliability. Prediction-powered inference methods seek to exploit such predictions to improve efficiency when labeled data are limited. However, existing approaches implicitly adopt a use-all philosophy, under which incorporating more predictions is presumed to improve inference. When prediction quality is heterogeneous, […]

Ver mais

arXiv:2602.02563v1 Announce Type: new Abstract: Prevailing spatiotemporal prediction models typically operate under a forward (unidirectional) learning paradigm, in which models extract spatiotemporal features from historical observation input and map them to target spatiotemporal space for future forecasting (label). However, these models frequently exhibit suboptimal performance when spatiotemporal discrepancies exist between inputs and labels, for instance, when nodes with similar time-series inputs manifest distinct future labels, or vice versa. To address this limitation, we propose explicitly incorporating label features […]

Ver mais

Over the last two years, something subtle has shifted in software engineering, and most of us have felt it even if we can’t name it. Teams report being dramatically more productive. Code is written faster, prototypes appear sooner, and the distance from idea to working system has collapsed. And yet, the software we rely on every day doesn’t feel proportionally better. In many cases, it feels harder to understand, harder to change, and more fragile than before. The […]

Ver mais

arXiv:2603.05618v1 Announce Type: new Abstract: Chain-of-Thought (CoT) prompting improves LLM reasoning but can increase privacy risk by resurfacing personally identifiable information (PII) from the prompt into reasoning traces and outputs, even under policies that instruct the model not to restate PII. We study such direct, inference-time PII leakage using a model-agnostic framework that (i) defines leakage as risk-weighted, token-level events across 11 PII types, (ii) traces leakage curves as a function of the allowed CoT budget, and (iii) […]

Ver mais

The transition from fifth-generation (5G) to sixth-generation (6G) mobile networks represents a fundamental shift in wireless communication paradigms, driven by the need for ultra-low latency, extreme data rates, native intelligence, and support for mission-critical and immersive applications. This paper presents the Rexhep Network Optimization Framework, a layered and AI-native architectural model designed to enable a smooth, efficient, and scalable evolution from 5G to 6G systems. The proposed framework integrates physical and spectrum intelligence, intelligent radio access networks (RAN) […]

Ver mais