no code implementations • Findings (EMNLP) 2021 • Yusheng Huang, Weijia Jia
Document-level event extraction is critical to various natural language processing tasks for providing structured information.
no code implementations • 21 Nov 2023 • Yusheng Huang, Yong Deng
Visibility graph (VG) transformation is a technique used to convert a time series into a graph based on specific visibility criteria.
1 code implementation • 10 Oct 2023 • Yusheng Huang, Zhouhan Lin
Multimodal information extraction is attracting research attention nowadays, which requires aggregating representations from different modalities.
1 code implementation • 6 Oct 2023 • Yusheng Huang, Zhouhan Lin
Document-level relation extraction (DocRE) is a task that focuses on identifying relations between entities within a document.
1 code implementation • 10 Apr 2023 • Yusheng Huang, Jiexing Qi, Xinbing Wang, Zhouhan Lin
We further employ the asymmetric focusing mechanism to decouple the gradient contribution from the negative and positive samples.
1 code implementation • 19 Feb 2023 • Jiexing Qi, Shuhao Li, Zhixin Guo, Yusheng Huang, Chenghu Zhou, Weinan Zhang, Xinbing Wang, Zhouhan Lin
In this work, we first collect a large-scale institution name normalization dataset LoT-insts1, which contains over 25k classes that exhibit a naturally long-tailed distribution.
Ranked #1 on Long-tail Learning on Lot-insts
no code implementations • 7 Oct 2021 • Qianli Zhou, Yusheng Huang, Yong Deng
Smets proposes the Pignistic Probability Transformation (PPT) as the decision layer in the Transferable Belief Model (TBM), which argues when there is no more information, we have to make a decision using a Probability Mass Function (PMF).
no code implementations • 8 Jan 2021 • Yusheng Huang, Dong Chu, Joel Weijia Lai, Yong Deng, Kang Hao Cheong
Based on the D-Path and T-Point, a newly accelerated PPA named OPPA-D using the proposed termination criterion is developed which is superior to all other baseline algorithms according to the experiments conducted in this paper.
1 code implementation • 19 Oct 2020 • Yusheng Huang, Dong Chu, Yong Deng, Kang Hao Cheong
Motivated by the lack of related PPA-based research, a novel framework, the capacitated physarum polycephalum inspired algorithm (CPPA), is proposed to allow capacity constraints toward link flow in the PPA.