Zobrazeno 1 - 10
of 16 127
pro vyhledávání: '"A., Darshan"'
Gastrointestinal (GI) bleeding is a serious medical condition that presents significant diagnostic challenges, particularly in settings with limited access to healthcare resources. Wireless Capsule Endoscopy (WCE) has emerged as a powerful diagnostic
Externí odkaz:
http://arxiv.org/abs/2412.18591
Autor:
Tanaka, Takumi S., Silverman, John D., Shimasaku, Kazuhiro, Arita, Junya, Akins, Hollis B., Inayoshi, Kohei, Ding, Xuheng, Onoue, Masafusa, Liu, Zhaoxuan, Casey, Caitlin M., Lambrides, Erini, Kokorev, Vasily, Jin, Shuowen, Faisst, Andreas L., Drakos, Nicole, Shen, Yue, Li, Junyao, Zhuang, Mingyang, Fei, Qinyue, Ito, Kei, Ren, Wenke, Matsui, Suin, Ando, Makoto, Hatano, Shun, Fujii, Michiko S., Kartaltepe, Jeyhan S., Koekemoer, Anton M., Liu, Daizhong, McCracken, Henry Joy, Rhodes, Jason, Robertson, Brant E., Franco, Maximilien, Andika, Irham T., Cloonan, Aidan P., Fan, Xiaohui, Gozaliasl, Ghassem, Harish, Santosh, Hayward, Christopher C., Huertas-Company, Marc, Kakkad, Darshan, Kinugawa, Tomoya, Roy, Namrata, Shuntov, Marko, Talia, Margherita, Toft, Sune, Vijayan, Aswin P., Zhang, Yiyang
``Little Red Dots'' (LRDs) are an abundant high-redshift population newly discovered by the James Webb Space Telescope (JWST). They are characterized by a red color in the rest-frame optical band, compact morphology, and broad Balmer emission lines (
Externí odkaz:
http://arxiv.org/abs/2412.14246
Autor:
Deshpande, Darshan, Ravi, Selvan Sunitha, CH-Wang, Sky, Mielczarek, Bartosz, Kannappan, Anand, Qian, Rebecca
The LLM-as-judge paradigm is increasingly being adopted for automated evaluation of model outputs. While LLM judges have shown promise on constrained evaluation tasks, closed source LLMs display critical shortcomings when deployed in real world appli
Externí odkaz:
http://arxiv.org/abs/2412.14140
Active Learning (AL) is a user-interactive approach aimed at reducing annotation costs by selecting the most crucial examples to label. Although AL has been extensively studied for image classification tasks, the specific scenario of interactive imag
Externí odkaz:
http://arxiv.org/abs/2412.02310
Image restoration aims to recover high-quality images from degraded observations. When the degradation process is known, the recovery problem can be formulated as an inverse problem, and in a Bayesian context, the goal is to sample a clean reconstruc
Externí odkaz:
http://arxiv.org/abs/2411.15295
Autor:
Delaunay, Pierre, Bouthillier, Xavier, Breuleux, Olivier, Ortiz-Gagné, Satya, Bilaniuk, Olexa, Normandin, Fabrice, Bergeron, Arnaud, Carrez, Bruno, Alain, Guillaume, Blanc, Soline, Osterrath, Frédéric, Viviano, Joseph, Patil, Roger Creus-Castanyer Darshan, Awal, Rabiul, Zhang, Le
AI workloads, particularly those driven by deep learning, are introducing novel usage patterns to high-performance computing (HPC) systems that are not comprehensively captured by standard HPC benchmarks. As one of the largest academic research cente
Externí odkaz:
http://arxiv.org/abs/2411.11940
Autor:
Balasubramanian, S, Subramaniam, M Sai, Talasu, Sai Sriram, Krishna, P Yedu, Sai, Manepalli Pranav Phanindra, Mukkamala, Ravi, Gera, Darshan
Deep neural networks (DNNS) excel at learning from static datasets but struggle with continual learning, where data arrives sequentially. Catastrophic forgetting, the phenomenon of forgetting previously learned knowledge, is a primary challenge. This
Externí odkaz:
http://arxiv.org/abs/2410.23751
Autor:
Koeplinger, David, Gandhi, Darshan, Nandkar, Pushkar, Sheeley, Nathan, Musaddiq, Matheen, Zhang, Leon, Goodbar, Reid, Shaffer, Matthew, Wang, Han, Wang, Angela, Wang, Mingran, Prabhakar, Raghu
Token generation speed is critical to power the next wave of AI inference applications. GPUs significantly underperform during token generation due to synchronization overheads at kernel boundaries, utilizing only 21% of their peak memory bandwidth.
Externí odkaz:
http://arxiv.org/abs/2410.23668
Autor:
Middleton, Alicia, Choi, Steve K., Walker, Samantha, Austermann, Jason, Burgoyne, James R., Butler, Victoria, Chapman, Scott C., Crites, Abigail T., Duell, Cody J., Freundt, Rodrigo G., Huber, Anthony I., Huber, Zachary B., Hubmayr, Johannes, Keller, Ben, Lin, Lawrence T., Niemack, Michael D., Patel, Darshan, Sinclair, Adrian K., Smith, Ema, Vaskuri, Anna, Vavagiakis, Eve M., Vissers, Michael, Wang, Yuhan, Wheeler, Jordan
Prime-Cam, one of the primary instruments for the Fred Young Submillimeter Telescope (FYST) developed by the CCAT Collaboration, will house up to seven instrument modules, with the first operating at 280 GHz. Each module will include three arrays of
Externí odkaz:
http://arxiv.org/abs/2410.21396
Activation functions introduce non-linearity into Neural Networks, enabling them to learn complex patterns. Different activation functions vary in speed and accuracy, ranging from faster but less accurate options like ReLU to slower but more accurate
Externí odkaz:
http://arxiv.org/abs/2410.10887