Zobrazeno 1 - 10
of 340
pro vyhledávání: '"Ng, Patrick P."'
Autor:
Yuan, Jiaqing, Pan, Lin, Hang, Chung-Wei, Guo, Jiang, Jiang, Jiarong, Min, Bonan, Ng, Patrick, Wang, Zhiguo
Large language models (LLMs) have shown remarkable performance on a variety of NLP tasks, and are being rapidly adopted in a wide range of use cases. It is therefore of vital importance to holistically evaluate the factuality of their generated outpu
Externí odkaz:
http://arxiv.org/abs/2404.16164
Current approaches of knowledge editing struggle to effectively propagate updates to interconnected facts. In this work, we delve into the barriers that hinder the appropriate propagation of updated knowledge within these models for accurate reasonin
Externí odkaz:
http://arxiv.org/abs/2401.17585
Autor:
Li, Alexander Hanbo, Shang, Mingyue, Spiliopoulou, Evangelia, Ma, Jie, Ng, Patrick, Wang, Zhiguo, Min, Bonan, Wang, William, McKeown, Kathleen, Castelli, Vittorio, Roth, Dan, Xiang, Bing
We present a novel approach for structured data-to-text generation that addresses the limitations of existing methods that primarily focus on specific types of structured data. Our proposed method aims to improve performance in multi-task training, z
Externí odkaz:
http://arxiv.org/abs/2308.05317
Autor:
Fu, Xingyu, Zhang, Sheng, Kwon, Gukyeong, Perera, Pramuditha, Zhu, Henghui, Zhang, Yuhao, Li, Alexander Hanbo, Wang, William Yang, Wang, Zhiguo, Castelli, Vittorio, Ng, Patrick, Roth, Dan, Xiang, Bing
The open-ended Visual Question Answering (VQA) task requires AI models to jointly reason over visual and natural language inputs using world knowledge. Recently, pre-trained Language Models (PLM) such as GPT-3 have been applied to the task and shown
Externí odkaz:
http://arxiv.org/abs/2305.18842
Autor:
Wang, Sijia, Li, Alexander Hanbo, Zhu, Henry, Zhang, Sheng, Hang, Chung-Wei, Perera, Pramuditha, Ma, Jie, Wang, William, Wang, Zhiguo, Castelli, Vittorio, Xiang, Bing, Ng, Patrick
Entities can be expressed in diverse formats, such as texts, images, or column names and cell values in tables. While existing entity linking (EL) models work well on per modality configuration, such as text-only EL, visual grounding, or schema linki
Externí odkaz:
http://arxiv.org/abs/2305.17337
Autor:
Lan, Wuwei, Wang, Zhiguo, Chauhan, Anuj, Zhu, Henghui, Li, Alexander, Guo, Jiang, Zhang, Sheng, Hang, Chung-Wei, Lilien, Joseph, Hu, Yiqun, Pan, Lin, Dong, Mingwen, Wang, Jun, Jiang, Jiarong, Ash, Stephen, Castelli, Vittorio, Ng, Patrick, Xiang, Bing
A practical text-to-SQL system should generalize well on a wide variety of natural language questions, unseen database schemas, and novel SQL query structures. To comprehensively evaluate text-to-SQL systems, we introduce a UNIfied benchmark for Text
Externí odkaz:
http://arxiv.org/abs/2305.16265
Autor:
Chang, Shuaichen, Wang, Jun, Dong, Mingwen, Pan, Lin, Zhu, Henghui, Li, Alexander Hanbo, Lan, Wuwei, Zhang, Sheng, Jiang, Jiarong, Lilien, Joseph, Ash, Steve, Wang, William Yang, Wang, Zhiguo, Castelli, Vittorio, Ng, Patrick, Xiang, Bing
Neural text-to-SQL models have achieved remarkable performance in translating natural language questions into SQL queries. However, recent studies reveal that text-to-SQL models are vulnerable to task-specific perturbations. Previous curated robustne
Externí odkaz:
http://arxiv.org/abs/2301.08881
Autor:
Zhao, Yiyun, Jiang, Jiarong, Hu, Yiqun, Lan, Wuwei, Zhu, Henry, Chauhan, Anuj, Li, Alexander, Pan, Lin, Wang, Jun, Hang, Chung-Wei, Zhang, Sheng, Dong, Marvin, Lilien, Joe, Ng, Patrick, Wang, Zhiguo, Castelli, Vittorio, Xiang, Bing
Recently, there has been increasing interest in synthesizing data to improve downstream text-to-SQL tasks. In this paper, we first examined the existing synthesized datasets and discovered that state-of-the-art text-to-SQL algorithms did not further
Externí odkaz:
http://arxiv.org/abs/2212.08785
Autor:
Chen, Jifan, Zhang, Yuhao, Liu, Lan, Dong, Rui, Chen, Xinchi, Ng, Patrick, Wang, William Yang, Huang, Zhiheng
There has been great progress in unifying various table-to-text tasks using a single encoder-decoder model trained via multi-task learning (Xie et al., 2022). However, existing methods typically encode task information with a simple dataset name as a
Externí odkaz:
http://arxiv.org/abs/2212.08780
Autor:
Yu, Donghan, Zhang, Sheng, Ng, Patrick, Zhu, Henghui, Li, Alexander Hanbo, Wang, Jun, Hu, Yiqun, Wang, William, Wang, Zhiguo, Xiang, Bing
Question answering over knowledge bases (KBs) aims to answer natural language questions with factual information such as entities and relations in KBs. Previous methods either generate logical forms that can be executed over KBs to obtain final answe
Externí odkaz:
http://arxiv.org/abs/2210.00063