no code implementations • 27 Feb 2024 • Xiangqing Shen, Yurun Song, Siwei Wu, Rui Xia
In this work, we draw inspiration from a commonsense knowledge base ConceptNet in natural language processing, and systematically define the types of visual commonsense.
no code implementations • 5 Oct 2023 • Siwei Wu, Xiangqing Shen, Rui Xia
Firstly, EDIT generates open questions related to the dialogue context as the potential user's intention; Then, EDIT answers those questions by interacting with LLMs and searching in domain-specific knowledge bases respectively, and use LLMs to choose the proper answers to questions as extra knowledge; Finally, EDIT enhances response generation by explicitly integrating those extra knowledge.
1 code implementation • 26 May 2023 • Siwei Wu, Xiangqing Shen, Rui Xia
To address the two problems, we propose a new CSKG completion framework based on Contrastive Pretraining and Node Clustering (CPNC).
no code implementations • 5 Apr 2023 • Yaochen Zhu, Xiangqing Shen, Rui Xia
For another, the multimodal reasoning task emphasized the prediction of future states and behaviors but often neglected the incorporation of individual personality traits.
1 code implementation • 14 Oct 2022 • Xiangqing Shen, Siwei Wu, Rui Xia
Both automatic and human evaluation on an annotated subgraph of ATOMIC demonstrate the advantage of Rel-CSKGC over strong baselines.