Stay updated with the latest research and technology news
arXiv:2506.10371v2 Announce Type: replace Abstract: The self-attention mechanism, a cornerstone of Transformer-based state-of-the-art deep learning architectures, is largely heuristic-driven and fundamentally challenging to interpret. Establishing...
arXiv:2506.11389v3 Announce Type: replace Abstract: As the cost of pretraining large language models grows, there is continued interest in strategies to improve learning efficiency during...
arXiv:2506.12007v2 Announce Type: replace Abstract: Neural surrogates for Partial Differential Equations (PDEs) often suffer significant performance degradation when evaluated on problem configurations outside their training...
arXiv:2506.13564v2 Announce Type: replace Abstract: We propose an efficient framework to compress massive video-frame features before feeding them into large multimodal models, thereby mitigating the...
arXiv:2506.13880v2 Announce Type: replace Abstract: We introduce surface Minkowski tensors to characterize rotational symmetries of shapes embedded in curved surfaces. The definition is based on...
arXiv:2506.14734v4 Announce Type: replace Abstract: The suffix tree is arguably the most fundamental data structure on strings: introduced by Weiner (SWAT 1973) and McCreight (JACM...
arXiv:2506.15199v2 Announce Type: replace Abstract: While there are many applications of ML to scientific problems that look promising, visuals can be deceiving. Using numerical analysis...
arXiv:2506.16309v3 Announce Type: replace Abstract: Over the last few years, machine learning unlocked previously infeasible features for compression, such as providing guarantees for users' privacy...
arXiv:2506.16796v3 Announce Type: replace Abstract: Real-World Image Super-Resolution is one of the most challenging task in image restoration. However, existing methods struggle with an accurate...
arXiv:2506.16982v2 Announce Type: replace Abstract: Accurately assessing student knowledge is central to education. Cognitive Diagnosis (CD) models estimate student proficiency at a fixed point in...
arXiv:2506.18058v2 Announce Type: replace Abstract: Excessive computational times represent a major challenge in the solution of corrosion models, limiting their practical applicability, e.g., as a...
arXiv:2506.18221v3 Announce Type: replace Abstract: Transfer learning is widely used to adapt large pretrained models to new tasks with only a small amount of new...
arXiv:2506.18755v3 Announce Type: replace Abstract: We study the Universal Solvability of Robot Motion Planning on Graphs (USolR) problem: given an undirected graph $G = (V,...
arXiv:2506.19307v2 Announce Type: replace Abstract: Presbyopia, a common age-related vision condition affecting most people as they age, often remains inadequately understood by those unaffected. To...
arXiv:2506.19385v4 Announce Type: replace Abstract: We present CID-GraphRAG (Conversational Intent-Driven Graph Retrieval-Augmented Generation), a novel framework that addresses the limitations of existing dialogue systems in...
arXiv:2506.20487v5 Announce Type: replace Abstract: Humanoid robots are drawing significant attention as versatile platforms for complex motor control, human-robot interaction, and general-purpose physical intelligence. However,...
arXiv:2506.21550v2 Announce Type: replace Abstract: Anomaly detection in multivariate time series is essential across domains such as healthcare, cybersecurity, and industrial monitoring, yet remains fundamentally...
arXiv:2506.21910v3 Announce Type: replace Abstract: In language model training, it is desirable to equip models with capabilities from various tasks. However, it is not clear...
arXiv:2506.22956v2 Announce Type: replace Abstract: Exploring high-latitude lunar regions presents an extremely challenging visual environment for robots. The low sunlight elevation angle and minimal light...
arXiv:2507.00075v4 Announce Type: replace Abstract: Self-improvement is a significant techniques within the realm of large language model (LLM), aiming to enhance the LLM performance without...
arXiv:2507.01835v3 Announce Type: replace Abstract: Hyperspectral reconstruction (HSR) from RGB images is a fundamentally ill-posed problem due to severe spectral information loss. Existing approaches typically...
arXiv:2507.02187v2 Announce Type: replace Abstract: There is growing industry interest in creating unobtrusive designs for electrooculography (EOG) sensing of eye gestures on glasses (e.g. JINS...
arXiv:2507.03006v2 Announce Type: replace Abstract: This work presents a comparative evaluation of two fundamentally different feature extraction paradigms--Histogram of Oriented Gradients (HOG) and Topological Data...
arXiv:2507.03041v4 Announce Type: replace Abstract: Compound AI systems integrating multiple components, such as Large Language Models, specialized tools, and traditional machine learning models, are increasingly...