Zobrazeno 1 - 10
of 1 406
pro vyhledávání: '"Borisyuk, A."'
We study user history modeling via Transformer encoders in deep learning recommendation models (DLRM). Such architectures can significantly improve recommendation quality, but usually incur high latency cost necessitating infrastructure upgrades or v
Externí odkaz:
http://arxiv.org/abs/2412.06924
Autor:
Borisyuk, Fedor, Song, Qingquan, Zhou, Mingzhou, Parameswaran, Ganesh, Arun, Madhu, Popuri, Siva, Bingol, Tugrul, Pei, Zhuotao, Lee, Kuang-Hsuan, Zheng, Lu, Shao, Qizhan, Naqvi, Ali, Zhou, Sen, Gupta, Aman
This paper introduces LiNR, LinkedIn's large-scale, GPU-based retrieval system. LiNR supports a billion-sized index on GPU models. We discuss our experiences and challenges in creating scalable, differentiable search indexes using TensorFlow and PyTo
Externí odkaz:
http://arxiv.org/abs/2407.13218
Autor:
Liu, Andrew, Borisyuk, Alla
The environments where individuals live can present diverse navigation challenges, resulting in varying navigation abilities and strategies. Inspired by differing urban layouts and the Dual Solutions Paradigm test used for human navigators, we develo
Externí odkaz:
http://arxiv.org/abs/2407.03436
Autor:
Wang, Ruofan, Prabhakar, Prakruthi, Srivastava, Gaurav, Wang, Tianqi, Jalali, Zeinab S., Bharill, Varun, Ouyang, Yunbo, Nigam, Aastha, Venugopalan, Divya, Gupta, Aman, Borisyuk, Fedor, Keerthi, Sathiya, Muralidharan, Ajith
In the realm of recommender systems, the ubiquitous adoption of deep neural networks has emerged as a dominant paradigm for modeling diverse business objectives. As user bases continue to expand, the necessity of personalization and frequent model up
Externí odkaz:
http://arxiv.org/abs/2403.00803
Autor:
Shen, Jianqiang, Juan, Yuchin, Zhang, Shaobo, Liu, Ping, Pu, Wen, Vasudevan, Sriram, Song, Qingquan, Borisyuk, Fedor, Shen, Kay Qianqi, Wei, Haichao, Ren, Yunxiang, Chiou, Yeou S., Kuang, Sicong, Yin, Yuan, Zheng, Ben, Wu, Muchen, Gharghabi, Shaghayegh, Wang, Xiaoqing, Xue, Huichao, Guo, Qi, Hewlett, Daniel, Simon, Luke, Hong, Liangjie, Zhang, Wenjing
Web-scale search systems typically tackle the scalability challenge with a two-step paradigm: retrieval and ranking. The retrieval step, also known as candidate selection, often involves extracting standardized entities, creating an inverted index, a
Externí odkaz:
http://arxiv.org/abs/2402.13435
Autor:
Liu, Ping, Wei, Haichao, Hou, Xiaochen, Shen, Jianqiang, He, Shihai, Shen, Kay Qianqi, Chen, Zhujun, Borisyuk, Fedor, Hewlett, Daniel, Wu, Liang, Veeraraghavan, Srikant, Tsun, Alex, Jiang, Chengming, Zhang, Wenjing
We present LinkSAGE, an innovative framework that integrates Graph Neural Networks (GNNs) into large-scale personalized job matching systems, designed to address the complex dynamics of LinkedIns extensive professional network. Our approach capitaliz
Externí odkaz:
http://arxiv.org/abs/2402.13430
Autor:
Borisyuk, Fedor, He, Shihai, Ouyang, Yunbo, Ramezani, Morteza, Du, Peng, Hou, Xiaochen, Jiang, Chengming, Pasumarthy, Nitin, Bannur, Priya, Tiwana, Birjodh, Liu, Ping, Dangi, Siddharth, Sun, Daqi, Pei, Zhoutao, Shi, Xiao, Zhu, Sirou, Shen, Qianqi, Lee, Kuang-Hsuan, Stein, David, Li, Baolei, Wei, Haichao, Ghoting, Amol, Ghosh, Souvik
In this paper, we present LiGNN, a deployed large-scale Graph Neural Networks (GNNs) Framework. We share our insight on developing and deployment of GNNs at large scale at LinkedIn. We present a set of algorithmic improvements to the quality of GNN r
Externí odkaz:
http://arxiv.org/abs/2402.11139
Autor:
Borisyuk, Fedor, Zhou, Mingzhou, Song, Qingquan, Zhu, Siyu, Tiwana, Birjodh, Parameswaran, Ganesh, Dangi, Siddharth, Hertel, Lars, Xiao, Qiang, Hou, Xiaochen, Ouyang, Yunbo, Gupta, Aman, Singh, Sheallika, Liu, Dan, Cheng, Hailing, Le, Lei, Hung, Jonathan, Keerthi, Sathiya, Wang, Ruoyan, Zhang, Fengyu, Kothari, Mohit, Zhu, Chen, Sun, Daqi, Dai, Yun, Luan, Xun, Zhu, Sirou, Wang, Zhiwei, Daftary, Neil, Shen, Qianqi, Jiang, Chengming, Wei, Haichao, Varshney, Maneesh, Ghoting, Amol, Ghosh, Souvik
We present LiRank, a large-scale ranking framework at LinkedIn that brings to production state-of-the-art modeling architectures and optimization methods. We unveil several modeling improvements, including Residual DCN, which adds attention and resid
Externí odkaz:
http://arxiv.org/abs/2402.06859
Autor:
Xiao, Qiang Charles, Muralidharan, Ajith, Tiwana, Birjodh, Jia, Johnson, Borisyuk, Fedor, Gupta, Aman, Woodard, Dawn
In this paper, we propose a generic model-based re-ranking framework, MultiSlot ReRanker, which simultaneously optimizes relevance, diversity, and freshness. Specifically, our Sequential Greedy Algorithm (SGA) is efficient enough (linear time complex
Externí odkaz:
http://arxiv.org/abs/2401.06293
We have studied the potential barriers for the penetration of atomic beryllium or boron inside the C60 fullerene by performing ab initio density functional theory (DFT) calculations with three variants for the exchange and correlation: B3LYP (hybrid
Externí odkaz:
http://arxiv.org/abs/2312.13928