no code implementations • 2 Apr 2024 • Junxiong Wang, Ali Mousavi, Omar Attia, Ronak Pradeep, Saloni Potdar, Alexander M. Rush, Umar Farooq Minhas, Yunyao Li
Existing generative approaches demonstrate improved accuracy compared to classification approaches under the standardized ZELDA benchmark.
Ranked #1 on Entity Linking on KORE50
no code implementations • 24 Jan 2024 • Junxiong Wang, Tushaar Gangavarapu, Jing Nathan Yan, Alexander M. Rush
We propose MambaByte, a token-free adaptation of the Mamba SSM trained autoregressively on byte sequences.
1 code implementation • 21 Jul 2023 • Kaiwen Wang, Junxiong Wang, Yueying Li, Nathan Kallus, Immanuel Trummer, Wen Sun
Join order selection (JOS) is the problem of ordering join operations to minimize total query execution cost and it is the core NP-hard combinatorial optimization problem of query optimization.
1 code implementation • 20 Dec 2022 • Junxiong Wang, Jing Nathan Yan, Albert Gu, Alexander M. Rush
Even so, BiGS is able to match BERT pretraining accuracy on GLUE and can be extended to long-form pretraining of 4096 tokens without approximation.
1 code implementation • 14 Oct 2021 • Junxiong Wang, Debabrota Basu, Immanuel Trummer
In black-box optimization problems, we aim to maximize an unknown objective function, where the function is only accessible through feedbacks of an evaluation or simulation oracle.
no code implementations • 24 Nov 2014 • Junxiong Wang, Hongzhi Wang, Chenxu Zhao
Currently, many machine learning algorithms contain lots of iterations.