Zobrazeno 1 - 10
of 465 740
pro vyhledávání: '"Wolf, A."'
A common practice in large language model (LLM) usage for complex analytical tasks such as code generation, is to sample a solution for the entire task within the model's context window. Previous works have shown that subtask decomposition within the
Externí odkaz:
http://arxiv.org/abs/2409.18028
We show that a universe with a non-minimally coupled scalar field can fit current measurements of the expansion rate of the Universe better than the standard $\Lambda$-Cold Dark Matter model or other minimally coupled dark energy models. While we fin
Externí odkaz:
http://arxiv.org/abs/2409.17019
Autor:
Mai, Zheda, Chowdhury, Arpita, Zhang, Ping, Tu, Cheng-Hao, Chen, Hong-You, Pahuja, Vardaan, Berger-Wolf, Tanya, Gao, Song, Stewart, Charles, Su, Yu, Chao, Wei-Lun
Fine-tuning is arguably the most straightforward way to tailor a pre-trained model (e.g., a foundation model) to downstream applications, but it also comes with the risk of losing valuable knowledge the model had learned in pre-training. For example,
Externí odkaz:
http://arxiv.org/abs/2409.16223
We propose an extension of the reinforcement learning architecture that enables moral decision-making of reinforcement learning agents based on normative reasons. Central to this approach is a reason-based shield generator yielding a moral shield tha
Externí odkaz:
http://arxiv.org/abs/2409.15014
Autor:
Chung, Long Kiu, Jung, Wonsuhk, Pullabhotla, Srivatsank, Shinde, Parth, Sunil, Yadu, Kota, Saihari, Batista, Luis Felipe Wolf, Pradalier, Cédric, Kousik, Shreyas
In the classical reach-avoid problem, autonomous mobile robots are tasked to reach a goal while avoiding obstacles. However, it is difficult to provide guarantees on the robot's performance when the obstacles form a narrow gap and the robot is a blac
Externí odkaz:
http://arxiv.org/abs/2409.13195
Autor:
Liu, Weizhe, Fan, Xiaohui, Yang, Jinyi, Bañados, Eduardo, Wang, Feige, Wolf, Julien, Barth, Aaron J., Costa, Tiago, Decarli, Roberto, Eilers, Anna-Christina, Loiacono, Federica, Shen, Yue, Farina, Emanuele Paolo, Jin, Xiangyu, Jun, Hyunsung D., Li, Mingyu, Lupi, Alessandro, Marshall, Madeline A., Pan, Zhiwei, Pudoka, Maria, Zhuang, Ming-Yang, Champagne, Jaclyn B., Li, Huan, Sun, Fengwu, Tee, Wei Leong, Vayner, Andrey, Zhang, Haowen
James Webb Space Telescope opens a new window to directly probe luminous quasars powered by billion solar mass black holes in the epoch of reionization and their co-evolution with massive galaxies with unprecedented details. In this paper, we report
Externí odkaz:
http://arxiv.org/abs/2409.13189
Plastic waste in aquatic environments poses severe risks to marine life and human health. Autonomous robots can be utilized to collect floating waste, but they require accurate object identification capability. While deep learning has been widely use
Externí odkaz:
http://arxiv.org/abs/2409.12659
Surface phonon polaritons (SPhPs) have become a key ingredient for infrared nanophotonics, owing to their long lifetimes and the large number of polar dielectric crystals supporting them. While these evanescent modes have been thoroughly characterize
Externí odkaz:
http://arxiv.org/abs/2409.12035
Autor:
Fischer, Tobias, Paredes, Isabel, Batchelor, Michael, Beier, Thorsten, Haviland, Jesse, Traversaro, Silvio, Vollprecht, Wolf, Schmitz, Markus, Milford, Michael
The Robot Operating System (ROS) has become the de facto standard middleware in robotics, widely adopted across domains ranging from education to industrial applications. The RoboStack distribution has extended ROS's accessibility by facilitating ins
Externí odkaz:
http://arxiv.org/abs/2409.09941
The blind image deconvolution is a challenging, highly ill-posed nonlinear inverse problem. We introduce a Multiscale Hierarchical Decomposition Method (MHDM) that is iteratively solving variational problems with adaptive data and regularization para
Externí odkaz:
http://arxiv.org/abs/2409.08734