Zobrazeno 1 - 10
of 1 842
pro vyhledávání: '"Rodriguez, Pedro A."'
Autor:
Pagnoni, Artidoro, Pasunuru, Ram, Rodriguez, Pedro, Nguyen, John, Muller, Benjamin, Li, Margaret, Zhou, Chunting, Yu, Lili, Weston, Jason, Zettlemoyer, Luke, Ghosh, Gargi, Lewis, Mike, Holtzman, Ari, Iyer, Srinivasan
We introduce the Byte Latent Transformer (BLT), a new byte-level LLM architecture that, for the first time, matches tokenization-based LLM performance at scale with significant improvements in inference efficiency and robustness. BLT encodes bytes in
Externí odkaz:
http://arxiv.org/abs/2412.09871
A semidomain is a subsemiring of an integral domain. Within this class, a unique factorization semidomain (UFS) is characterized by the property that every nonzero, nonunit element can be factored into a product of finitely many prime elements. In th
Externí odkaz:
http://arxiv.org/abs/2412.05261
Autor:
Lee, Yejin, Sun, Anna, Hosmer, Basil, Acun, Bilge, Balioglu, Can, Wang, Changhan, Hernandez, Charles David, Puhrsch, Christian, Haziza, Daniel, Guessous, Driss, Massa, Francisco, Kahn, Jacob, Wan, Jeffrey, Reizenstein, Jeremy, Zhai, Jiaqi, Isaacson, Joe, Schlosser, Joel, Pino, Juan, Sadagopan, Kaushik Ram, Shamis, Leonid, Ma, Linjian, Hwang, Min-Jae, Chen, Mingda, Elhoushi, Mostafa, Rodriguez, Pedro, Pasunuru, Ram, Yih, Scott, Popuri, Sravya, Liu, Xing, Wu, Carole-Jean
Generative artificial intelligence (AI) technology is revolutionizing the computing industry. Not only its applications have broadened to various sectors but also poses new system design and optimization opportunities. The technology is capable of un
Externí odkaz:
http://arxiv.org/abs/2410.00215
Autor:
Kornjača, Milan, Hu, Hong-Ye, Zhao, Chen, Wurtz, Jonathan, Weinberg, Phillip, Hamdan, Majd, Zhdanov, Andrii, Cantu, Sergio H., Zhou, Hengyun, Bravo, Rodrigo Araiza, Bagnall, Kevin, Basham, James I., Campo, Joseph, Choukri, Adam, DeAngelo, Robert, Frederick, Paige, Haines, David, Hammett, Julian, Hsu, Ning, Hu, Ming-Guang, Huber, Florian, Jepsen, Paul Niklas, Jia, Ningyuan, Karolyshyn, Thomas, Kwon, Minho, Long, John, Lopatin, Jonathan, Lukin, Alexander, Macrì, Tommaso, Marković, Ognjen, Martínez-Martínez, Luis A., Meng, Xianmei, Ostroumov, Evgeny, Paquette, David, Robinson, John, Rodriguez, Pedro Sales, Singh, Anshuman, Sinha, Nandan, Thoreen, Henry, Wan, Noel, Waxman-Lenz, Daniel, Wong, Tak, Wu, Kai-Hsin, Lopes, Pedro L. S., Boger, Yuval, Gemelke, Nathan, Kitagawa, Takuya, Keesling, Alexander, Gao, Xun, Bylinskii, Alexei, Yelin, Susanne F., Liu, Fangli, Wang, Sheng-Tao
Quantum machine learning has gained considerable attention as quantum technology advances, presenting a promising approach for efficiently learning complex data patterns. Despite this promise, most contemporary quantum methods require significant res
Externí odkaz:
http://arxiv.org/abs/2407.02553
The development and use of large-scale quantum computers relies on integrating quantum error-correcting (QEC) schemes into the quantum computing pipeline. A fundamental part of the QEC protocol is the decoding of the syndrome to identify a recovery o
Externí odkaz:
http://arxiv.org/abs/2405.03776
Efficiently solving unbalanced three-phase power flow in distribution grids is pivotal for grid analysis and simulation. There is a pressing need for scalable algorithms capable of handling large-scale unbalanced power grids that can provide accurate
Externí odkaz:
http://arxiv.org/abs/2403.00892
Dense retrievers compress source documents into (possibly lossy) vector representations, yet there is little analysis of what information is lost versus preserved, and how it affects downstream tasks. We conduct the first analysis of the information
Externí odkaz:
http://arxiv.org/abs/2402.15925
Autor:
Jiang, Zhengbao, Sun, Zhiqing, Shi, Weijia, Rodriguez, Pedro, Zhou, Chunting, Neubig, Graham, Lin, Xi Victoria, Yih, Wen-tau, Iyer, Srinivasan
In order for large language model (LLM)-based assistants to effectively adapt to evolving information needs, it must be possible to update their factual knowledge through continued training on new data. The standard recipe for doing so involves conti
Externí odkaz:
http://arxiv.org/abs/2402.12847
A submonoid of the additive group $\mathbb{Q}$ is called a Puiseux monoid if it consists of nonnegative rationals. Given a monoid $M$, the set consisting of all nonempty finite subsets of $M$ is also a monoid under the Minkowski sum, and it is called
Externí odkaz:
http://arxiv.org/abs/2401.12444
Autor:
Bluvstein, Dolev, Evered, Simon J., Geim, Alexandra A., Li, Sophie H., Zhou, Hengyun, Manovitz, Tom, Ebadi, Sepehr, Cain, Madelyn, Kalinowski, Marcin, Hangleiter, Dominik, Ataides, J. Pablo Bonilla, Maskara, Nishad, Cong, Iris, Gao, Xun, Rodriguez, Pedro Sales, Karolyshyn, Thomas, Semeghini, Giulia, Gullans, Michael J., Greiner, Markus, Vuletic, Vladan, Lukin, Mikhail D.
Publikováno v:
Nature (2023)
Suppressing errors is the central challenge for useful quantum computing, requiring quantum error correction for large-scale processing. However, the overhead in the realization of error-corrected ``logical'' qubits, where information is encoded acro
Externí odkaz:
http://arxiv.org/abs/2312.03982