Researchers propose adaptive AI contracts that selectively use detailed evaluations to reduce costs, but practical challenges remain. Empirical tests show promise in question-answering and code-genera…
Researchers have introduced the Data Relativistic Uncertainty (DRU) framework, which applies quantum concepts like wave-particle duality to AI-driven image enhancement, treating images as probabilisti…
A new approach to AI coordination uses field-programmable gate arrays (FPGAs) to enforce real-time safety and synchronization at the hardware level, addressing the limitations of software-only solutio…
A new Guard Rail Validation (GRV) framework aims to intercept and validate AI decisions in real-time before they affect live telecom networks, addressing risks of unchecked autonomous network operatio…
A new reinforcement learning approach is optimizing Formula 1 race strategies by adapting to real-time dynamics and competitors' behavior, giving teams a competitive edge through data-driven decision-…
Researchers applied hybrid quantum-classical neural networks to sentiment analysis of COVID-19 tweets, finding that the models matched classical accuracy and showed distinct learning dynamics. In tran…
Researchers introduced ReRanking Preference Optimization (RRPO), a method that uses reinforcement learning from LLM feedback to improve reranker relevance for AI-generated answers, outperforming stron…
Researchers in Singapore developed an AI model that improves rainfall prediction by accounting for the geometric differences between rain gauges, microwave links, and radar data. The model reduced pre…
Researchers introduced SelectTSL, a deep learning architecture for prompt-guided selective sound source localization that focuses only on user-specified targets in multi-source environments. The syste…
Researchers have developed a lightweight method for quantifying uncertainty in neural network predictions using two key approximations: a first-order Taylor expansion and an isotropy assumption. The a…
A survey of 377 Indian dermatologists reveals that while 49.9% use AI, it is primarily for administrative tasks rather than clinical diagnosis, and tools fail to address chronic disease management cha…
Researchers have developed a new backdoor attack on speech classification models called DRL-CLBA, which uses reinforcement learning and audio steganography to evade traditional defenses. The attack ac…
Researchers developed a two-stage AI pipeline that estimates wind conditions and uses reinforcement learning to improve drone trajectory tracking by 48% in turbulent winds. The method, tested in simul…
New research characterizes the intrinsic compressibility of KV caches in Transformer models, proposing a principled algorithm for efficient inference. The study uses minimax risk to determine when acc…
Researchers introduced Resonant Brane Splatting (RBS), a new framework for image super-resolution that uses dynamic Brane primitives to improve speed and quality over traditional methods. RBS achieves…
A study at a minority-serving university found that undergraduates use large language models in four distinct ways—Strategic, Instrumental, Dialogic, and Dependent—and that current assessment tools, w…
Researchers unveiled a latent world model using an equivariant encoder and predictor that achieves invariant predictions across orientations, with training loss symmetry ensuring consistent performanc…
Researchers introduced Locality-Aware Continual Unlearning (LACU), a framework for removing concepts from text-to-image diffusion models while preserving stability and protecting semantically related …
Researchers introduced the Semantic-Enhanced Patch Slimming (SEPS) framework to improve vision-language alignment by reducing patch redundancy and ambiguity, outperforming existing methods by 23% to 8…
Researchers have developed a novel AI method that uses safe reference policies as probabilistic regulators to balance exploration and safety, providing finite-sample guarantees and handling non-monoto…